Version bump to v1.13.0

Store worker statistics under worker id and not internal redis key
Fix unit tests

apiserver/apierrors/errors.conf

@@ -27,7 +27,7 @@
     24: ["not_public_object", "object is not public"]
 
     # Auth / Login
-    75: ["invalid_access_key", "access key not found for user"]
+    75: ["invalid_access_key", "access key not found"]
 
     # Tasks
     100: ["task_error", "general task error"]
@@ -50,6 +50,12 @@
     130: ["task_not_found", "task not found"]
     131: ["events_not_added", "events not added"]
 
+    # Reports
+    150: ["operation_supported_on_reports_only", "passed task is not report"]
+
+    # Pipelines
+    160: ["cannot_remove_all_runs", "at least one pipeline run should be left"]
+
     # Models
     200: ["model_error", "general task error"]
     201: ["invalid_model_id", "invalid model id"]
@@ -70,12 +76,14 @@
     402: ["project_has_tasks", "project has associated tasks"]
     403: ["project_not_found", "project not found"]
     405: ["project_has_models", "project has associated models"]
+    406: ["project_has_datasets", "project has associated non-empty datasets"]
     407: ["invalid_project_name", "invalid project name"]
     408: ["cannot_update_project_location", "Cannot update project location. Use projects.move instead"]
     409: ["project_path_exceeds_max", "Project path exceed the maximum allowed depth"]
     410: ["project_source_and_destination_are_the_same", "Project has the same source and destination paths"]
     411: ["project_cannot_be_moved_under_itself", "Project can not be moved under itself in the projects hierarchy"]
     412: ["project_cannot_be_merged_into_its_child", "Project can not be merged into its own child"]
+    413: ["project_has_pipelines", "project has associated pipelines with active controllers"]
 
     # Queues
     701: ["invalid_queue_id", "invalid queue id"]

apiserver/apimodels/__init__.py

@@ -61,6 +61,13 @@ class ListField(fields.ListField):
             item.validate()
 
 
+class ScalarField(fields.BaseField):
+
+    """String field."""
+
+    types = (str, int, float, bool)
+
+
 class DictField(fields.BaseField):
     types = (dict,)
 

apiserver/apimodels/events.py

@@ -36,7 +36,7 @@ class MultiTaskScalarMetricsIterHistogramRequest(HistogramRequestBase):
             Length(
                 minimum_value=1,
                 maximum_value=config.get(
-                    "services.tasks.multi_task_histogram_limit", 10
+                    "services.tasks.multi_task_histogram_limit", 100
                 ),
             )
         ],
@@ -155,6 +155,13 @@ class TaskMetricsRequest(MultiTasksRequestBase):
     event_type: EventType = ActualEnumField(EventType, required=True)
 
 
+class MultiTaskPlotsRequest(MultiTasksRequestBase):
+    iters: int = IntField(default=1)
+    scroll_id: str = StringField()
+    no_scroll: bool = BoolField(default=False)
+    last_iters_per_task_metric: bool = BoolField(default=True)
+
+
 class TaskPlotsRequest(Base):
     task: str = StringField(required=True)
     iters: int = IntField(default=1)

apiserver/apimodels/models.py

@@ -44,10 +44,12 @@ class ModelRequest(models.Base):
 
 class DeleteModelRequest(ModelRequest):
     force = fields.BoolField(default=False)
+    delete_external_artifacts = fields.BoolField(default=True)
 
 
 class ModelsDeleteManyRequest(BatchRequest):
     force = fields.BoolField(default=False)
+    delete_external_artifacts = fields.BoolField(default=True)
 
 
 class PublishModelRequest(ModelRequest):
@@ -79,3 +81,4 @@ class AddOrUpdateMetadataRequest(AddOrUpdateMetadata):
 
 class ModelsGetRequest(models.Base):
     include_stats = fields.BoolField(default=False)
+    allow_public = fields.BoolField(default=True)

apiserver/apimodels/organization.py

@@ -1,6 +1,11 @@
-from jsonmodels import fields, models
+from enum import auto
+from typing import Sequence
 
-from apiserver.apimodels import DictField
+from jsonmodels import fields, models
+from jsonmodels.validators import Length
+
+from apiserver.apimodels import DictField, ActualEnumField, ScalarField
+from apiserver.utilities.stringenum import StringEnum
 
 
 class Filter(models.Base):
@@ -19,5 +24,39 @@ class EntitiesCountRequest(models.Base):
     models = DictField()
     pipelines = DictField()
     datasets = DictField()
+    reports = DictField()
     active_users = fields.ListField(str)
     search_hidden = fields.BoolField(default=False)
+    allow_public = fields.BoolField(default=True)
+
+
+class EntityType(StringEnum):
+    task = auto()
+    model = auto()
+
+
+class ValueMapping(models.Base):
+    key = ScalarField(nullable=True)
+    value = ScalarField(nullable=True)
+
+
+class FieldMapping(models.Base):
+    field = fields.StringField(required=True)
+    name = fields.StringField()
+    values: Sequence[ValueMapping] = fields.ListField(items_types=[ValueMapping])
+
+
+class PrepareDownloadForGetAllRequest(models.Base):
+    entity_type = ActualEnumField(EntityType)
+    allow_public = fields.BoolField(default=True)
+    search_hidden = fields.BoolField(default=False)
+    only_fields = fields.ListField(
+        items_types=[str], validators=[Length(1)], required=True
+    )
+    field_mappings: Sequence[FieldMapping] = fields.ListField(
+        items_types=[FieldMapping], validators=[Length(1)], required=True
+    )
+
+
+class DownloadForGetAllRequest(models.Base):
+    prepare_id = fields.StringField(required=True)

apiserver/apimodels/pipelines.py

@@ -1,4 +1,5 @@
 from jsonmodels import models, fields
+from jsonmodels.validators import Length
 
 from apiserver.apimodels import ListField
 
@@ -8,6 +9,11 @@ class Arg(models.Base):
     value = fields.StringField(required=True)
 
 
+class DeleteRunsRequest(models.Base):
+    project = fields.StringField(required=True)
+    ids = ListField([str], required=True, validators=[Length(1)])
+
+
 class StartPipelineRequest(models.Base):
     task = fields.StringField(required=True)
     queue = fields.StringField(required=True)

apiserver/apimodels/projects.py

@@ -1,8 +1,11 @@
+from enum import Enum, auto
+
 from jsonmodels import models, fields
 
 from apiserver.apimodels import ListField, ActualEnumField, DictField
 from apiserver.apimodels.organization import TagsRequest
 from apiserver.database.model import EntityVisibility
+from apiserver.utilities.stringenum import StringEnum
 
 
 class ProjectRequest(models.Base):
@@ -20,6 +23,7 @@ class MoveRequest(ProjectRequest):
 class DeleteRequest(ProjectRequest):
     force = fields.BoolField(default=False)
     delete_contents = fields.BoolField(default=False)
+    delete_external_artifacts = fields.BoolField(default=True)
 
 
 class ProjectOrNoneRequest(models.Base):
@@ -27,6 +31,10 @@ class ProjectOrNoneRequest(models.Base):
     include_subprojects = fields.BoolField(default=True)
 
 
+class GetUniqueMetricsRequest(ProjectOrNoneRequest):
+    model_metrics = fields.BoolField(default=False)
+
+
 class GetParamsRequest(ProjectOrNoneRequest):
     page = fields.IntField(default=0)
     page_size = fields.IntField(default=500)
@@ -43,17 +51,38 @@ class MultiProjectRequest(models.Base):
 
 class ProjectTaskParentsRequest(MultiProjectRequest):
     tasks_state = ActualEnumField(EntityVisibility)
+    task_name = fields.StringField()
 
 
-class ProjectHyperparamValuesRequest(MultiProjectRequest):
+class EntityTypeEnum(StringEnum):
+    task = auto()
+    model = auto()
+
+
+class ProjectUserNamesRequest(MultiProjectRequest):
+    entity = ActualEnumField(EntityTypeEnum, default=EntityTypeEnum.task)
+
+
+class MultiProjectPagedRequest(MultiProjectRequest):
+    allow_public = fields.BoolField(default=True)
+    page = fields.IntField(default=0)
+    page_size = fields.IntField(default=500)
+
+
+class ProjectHyperparamValuesRequest(MultiProjectPagedRequest):
     section = fields.StringField(required=True)
     name = fields.StringField(required=True)
-    allow_public = fields.BoolField(default=True)
+    pattern = fields.StringField()
 
 
-class ProjectModelMetadataValuesRequest(MultiProjectRequest):
+class ProjectModelMetadataValuesRequest(MultiProjectPagedRequest):
     key = fields.StringField(required=True)
-    allow_public = fields.BoolField(default=True)
+
+
+class ProjectChildrenType(Enum):
+    pipeline = "pipeline"
+    report = "report"
+    dataset = "dataset"
 
 
 class ProjectsGetRequest(models.Base):
@@ -62,8 +91,12 @@ class ProjectsGetRequest(models.Base):
     include_stats_filter = DictField()
     stats_with_children = fields.BoolField(default=True)
     stats_for_state = ActualEnumField(EntityVisibility, default=EntityVisibility.active)
-    non_public = fields.BoolField(default=False)
+    non_public = fields.BoolField(default=False)  # legacy, use allow_public instead
     active_users = fields.ListField(str)
     check_own_contents = fields.BoolField(default=False)
     shallow_search = fields.BoolField(default=False)
     search_hidden = fields.BoolField(default=False)
+    allow_public = fields.BoolField(default=True)
+    children_type = ActualEnumField(ProjectChildrenType)
+    children_tags = fields.ListField(str)
+    children_tags_filter = DictField()

apiserver/apimodels/reports.py

@@ -0,0 +1,84 @@
+from typing import Sequence
+
+from jsonmodels import validators
+from jsonmodels.fields import StringField, ListField, BoolField, EmbeddedField, IntField
+from jsonmodels.models import Base
+from jsonmodels.validators import Length
+
+from apiserver.apimodels.events import MetricVariants, HistogramRequestBase
+
+
+class UpdateReportRequest(Base):
+    task = StringField(required=True)
+    name = StringField(nullable=True, validators=Length(minimum_value=3))
+    tags = ListField(items_types=[str])
+    comment = StringField()
+    report = StringField()
+    report_assets = ListField(items_types=[str])
+
+
+class CreateReportRequest(Base):
+    name = StringField(required=True, validators=Length(minimum_value=3))
+    tags = ListField(items_types=[str])
+    comment = StringField()
+    report = StringField()
+    project = StringField()
+    report_assets = ListField(items_types=[str])
+
+
+class PublishReportRequest(Base):
+    task = StringField(required=True)
+    message = StringField(default="")
+
+
+class ArchiveReportRequest(Base):
+    task = StringField(required=True)
+    message = StringField(default="")
+
+
+class ShareReportRequest(Base):
+    task = StringField(required=True)
+    share = BoolField(default=True)
+
+
+class DeleteReportRequest(Base):
+    task = StringField(required=True)
+    force = BoolField(default=False)
+
+
+class MoveReportRequest(Base):
+    task = StringField(required=True)
+    project = StringField()
+    project_name = StringField()
+
+
+class EventsRequest(Base):
+    iters = IntField(default=1, validators=validators.Min(1))
+    metrics: Sequence[MetricVariants] = ListField(items_types=MetricVariants)
+
+
+class PlotEventsRequest(EventsRequest):
+    last_iters_per_task_metric: bool = BoolField(default=True)
+
+
+class ScalarMetricsIterHistogram(HistogramRequestBase):
+    metrics: Sequence[MetricVariants] = ListField(items_types=MetricVariants)
+
+
+class SingleValueMetrics(Base):
+    pass
+
+
+class GetTasksDataRequest(Base):
+    debug_images: EventsRequest = EmbeddedField(EventsRequest)
+    plots: PlotEventsRequest = EmbeddedField(PlotEventsRequest)
+    scalar_metrics_iter_histogram: ScalarMetricsIterHistogram = EmbeddedField(
+        ScalarMetricsIterHistogram
+    )
+    single_value_metrics: SingleValueMetrics = EmbeddedField(SingleValueMetrics)
+    allow_public = BoolField(default=True)
+    model_events: bool = BoolField(default=False)
+
+
+class GetAllRequest(Base):
+    allow_public = BoolField(default=True)

apiserver/apimodels/tasks.py

@@ -96,6 +96,11 @@ class UpdateRequest(TaskUpdateRequest):
     status_message = StringField(default="")
 
 
+class DequeueRequest(UpdateRequest):
+    remove_from_all_queues = BoolField(default=False)
+    new_status = StringField()
+
+
 class EnqueueRequest(UpdateRequest):
     queue = StringField()
     queue_name = StringField()
@@ -274,6 +279,11 @@ class StopManyRequest(TaskBatchRequest):
     force = BoolField(default=False)
 
 
+class DequeueManyRequest(TaskBatchRequest):
+    remove_from_all_queues = BoolField(default=False)
+    new_status = StringField()
+
+
 class EnqueueManyRequest(TaskBatchRequest):
     queue = StringField()
     queue_name = StringField()
@@ -318,3 +328,13 @@ class DeleteModelsRequest(TaskRequest):
     models: Sequence[ModelItemKey] = ListField(
         [ModelItemKey], validators=Length(minimum_value=1)
     )
+
+
+class GetAllReq(models.Base):
+    allow_public = BoolField(default=True)
+    search_hidden = BoolField(default=False)
+
+
+class UpdateTagsRequest(BatchRequest):
+    add_tags = ListField([str])
+    remove_tags = ListField([str])

apiserver/apimodels/workers.py

@@ -12,9 +12,8 @@ from jsonmodels.fields import (
 )
 from jsonmodels.models import Base
 
-from apiserver.apimodels import make_default, ListField, EnumField, JsonSerializableMixin
-
-DEFAULT_TIMEOUT = 10 * 60
+from apiserver.apimodels import ListField, EnumField, JsonSerializableMixin
+from apiserver.config_repo import config
 
 
 class WorkerRequest(Base):
@@ -24,9 +23,10 @@ class WorkerRequest(Base):
 
 
 class RegisterRequest(WorkerRequest):
-    timeout = make_default(
-        IntField, DEFAULT_TIMEOUT
-    )()  # registration timeout in seconds (default is 10min)
+    timeout = IntField(
+        default=int(config.get("services.workers.default_worker_timeout_sec", 10 * 60))
+    )
+    """ registration timeout in seconds (default is 10min) """
     queues = ListField(six.string_types)  # list of queues this worker listens to
 
 
@@ -106,6 +106,10 @@ class GetAllResponse(Base):
     workers = ListField(WorkerResponseEntry)
 
 
+class GetCountRequest(GetAllRequest):
+    last_seen = IntField(default=0)
+
+
 class StatsBase(Base):
     worker_ids = ListField(str)
 

apiserver/bll/event/event_bll.py

@@ -5,10 +5,10 @@ import zlib
 from collections import defaultdict
 from contextlib import closing
 from datetime import datetime
-from operator import attrgetter
 from typing import Sequence, Set, Tuple, Optional, List, Mapping, Union
 
 import elasticsearch
+from boltons.iterutils import chunked_iter
 from elasticsearch.helpers import BulkIndexError
 from mongoengine import Q
 from nested_dict import nested_dict
@@ -23,12 +23,14 @@ from apiserver.bll.event.event_common import (
     get_metric_variants_condition,
     uncompress_plot,
     get_max_metric_and_variant_counts,
+    PlotFields,
 )
 from apiserver.bll.event.events_iterator import EventsIterator, TaskEventsResult
 from apiserver.bll.event.history_debug_image_iterator import HistoryDebugImageIterator
 from apiserver.bll.event.history_plots_iterator import HistoryPlotsIterator
 from apiserver.bll.event.metric_debug_images_iterator import MetricDebugImagesIterator
 from apiserver.bll.event.metric_plots_iterator import MetricPlotsIterator
+from apiserver.bll.model import ModelBLL
 from apiserver.bll.util import parallel_chunked_decorator
 from apiserver.database import utils as dbutils
 from apiserver.database.model.model import Model
@@ -45,25 +47,19 @@ from apiserver.utilities.dicts import nested_get
 from apiserver.utilities.json import loads
 
 # noinspection PyTypeChecker
-EVENT_TYPES: Set[str] = set(map(attrgetter("value"), EventType))
+EVENT_TYPES: Set[str] = set(et.value for et in EventType if et != EventType.all)
 LOCKED_TASK_STATUSES = (TaskStatus.publishing, TaskStatus.published)
-MAX_LONG = 2 ** 63 - 1
-MIN_LONG = -(2 ** 63)
+MAX_LONG = 2**63 - 1
+MIN_LONG = -(2**63)
 
 
 log = config.logger(__file__)
-
-
-class PlotFields:
-    valid_plot = "valid_plot"
-    plot_len = "plot_len"
-    plot_str = "plot_str"
-    plot_data = "plot_data"
-    source_urls = "source_urls"
+async_task_events_delete = config.get("services.tasks.async_events_delete", False)
+async_delete_threshold = config.get("services.tasks.async_events_delete_threshold", 100_000)
 
 
 class EventBLL(object):
-    id_fields = ("task", "iter", "metric", "variant", "key")
+    event_id_fields = ("task", "iter", "metric", "variant", "key")
     empty_scroll = "FFFF"
     img_source_regex = re.compile(
         r"['\"]source['\"]:\s?['\"]([a-z][a-z0-9+\-.]*://.*?)['\"]",
@@ -101,47 +97,76 @@ class EventBLL(object):
         return self._metrics
 
     @staticmethod
-    def _get_valid_tasks(company_id, task_ids: Set, allow_locked_tasks=False) -> Set:
-        """Verify that task exists and can be updated"""
-        if not task_ids:
+    def _get_valid_entities(company_id, ids: Mapping[str, bool], model=False) -> Set:
+        """Verify that task or model exists and can be updated"""
+        if not ids:
             return set()
 
         with translate_errors_context():
-            query = Q(id__in=task_ids, company=company_id)
-            if not allow_locked_tasks:
-                query &= Q(status__nin=LOCKED_TASK_STATUSES)
-            res = Task.objects(query).only("id")
-            return {r.id for r in res}
+            allow_locked = {id_ for id_, allowed in ids.items() if allowed}
+            not_locked = {id_ for id_, allowed in ids.items() if not allowed}
+            res = set()
+            allow_locked_q = Q()
+            not_locked_q = (
+                Q(ready__ne=True) if model else Q(status__nin=LOCKED_TASK_STATUSES)
+            )
+            for requested_ids, locked_q in (
+                (allow_locked, allow_locked_q),
+                (not_locked, not_locked_q),
+            ):
+                if not requested_ids:
+                    continue
+                query = Q(id__in=requested_ids, company=company_id)
+                res.update(
+                    (Model if model else Task).objects(query & locked_q).scalar("id")
+                )
 
-    @staticmethod
-    def _get_valid_models(company_id, model_ids: Set, allow_locked_models=False) -> Set:
-        """Verify that task exists and can be updated"""
-        if not model_ids:
-            return set()
-
-        with translate_errors_context():
-            query = Q(id__in=model_ids, company=company_id)
-            if not allow_locked_models:
-                query &= Q(ready__ne=True)
-            res = Model.objects(query).only("id")
-            return {r.id for r in res}
+            return res
 
     def add_events(
-        self, company_id, events, worker, allow_locked=False
+        self,
+        company_id: str,
+        user_id: str,
+        events: Sequence[dict],
+        worker: str,
     ) -> Tuple[int, int, dict]:
-        model_events = events[0].get("model_event", False)
+        task_ids = {}
+        model_ids = {}
         for event in events:
-            if event.get("model_event", model_events) != model_events:
+            if event.get("model_event", False):
+                model = event.pop("model", None)
+                if model is not None:
+                    event["task"] = model
+                entity_ids = model_ids
+            else:
+                event["model_event"] = False
+                entity_ids = task_ids
+
+            id_ = event.get("task")
+            allow_locked = event.pop("allow_locked", False)
+            if not id_:
+                continue
+
+            allowed_for_entity = entity_ids.get(id_)
+            if allowed_for_entity is None:
+                entity_ids[id_] = allow_locked
+            elif allowed_for_entity != allow_locked:
                 raise errors.bad_request.ValidationError(
-                    "Inconsistent model_event setting in the passed events"
-                )
-            if event.pop("allow_locked", allow_locked) != allow_locked:
-                raise errors.bad_request.ValidationError(
-                    "Inconsistent allow_locked setting in the passed events"
+                    f"Inconsistent allow_locked setting in the passed events for {id_}"
                 )
 
+        found_in_both = set(task_ids).intersection(set(model_ids))
+        if found_in_both:
+            raise errors.bad_request.ValidationError(
+                "Inconsistent model_event setting in the passed events",
+                tasks=found_in_both,
+            )
+        valid_models = self._get_valid_entities(company_id, ids=model_ids, model=True)
+        valid_tasks = self._get_valid_entities(company_id, ids=task_ids)
+
         actions: List[dict] = []
-        task_or_model_ids = set()
+        used_task_ids = set()
+        used_model_ids = set()
         task_iteration = defaultdict(lambda: 0)
         task_last_scalar_events = nested_dict(
             3, dict
@@ -151,28 +176,6 @@ class EventBLL(object):
         )  # task_id -> metric_hash -> event_type -> MetricEvent
         errors_per_type = defaultdict(int)
         invalid_iteration_error = f"Iteration number should not exceed {MAX_LONG}"
-        if model_events:
-            for event in events:
-                model = event.pop("model", None)
-                if model is not None:
-                    event["task"] = model
-            valid_entities = self._get_valid_models(
-                company_id,
-                model_ids={
-                    event["task"] for event in events if event.get("task") is not None
-                },
-                allow_locked_models=allow_locked,
-            )
-            entity_name = "model"
-        else:
-            valid_entities = self._get_valid_tasks(
-                company_id,
-                task_ids={
-                    event["task"] for event in events if event.get("task") is not None
-                },
-                allow_locked_tasks=allow_locked,
-            )
-            entity_name = "task"
 
         for event in events:
             # remove spaces from event type
@@ -186,7 +189,8 @@ class EventBLL(object):
                 errors_per_type[f"Invalid event type {event_type}"] += 1
                 continue
 
-            if model_events and event_type == EventType.task_log.value:
+            model_event = event["model_event"]
+            if model_event and event_type == EventType.task_log.value:
                 errors_per_type[f"Task log events are not supported for models"] += 1
                 continue
 
@@ -195,8 +199,12 @@ class EventBLL(object):
                 errors_per_type["Event must have a 'task' field"] += 1
                 continue
 
-            if task_or_model_id not in valid_entities:
-                errors_per_type[f"Invalid {entity_name} id {task_or_model_id}"] += 1
+            if (model_event and task_or_model_id not in valid_models) or (
+                not model_event and task_or_model_id not in valid_tasks
+            ):
+                errors_per_type[
+                    f"Invalid {'model' if model_event else 'task'} id {task_or_model_id}"
+                ] += 1
                 continue
 
             event["type"] = event_type
@@ -218,13 +226,10 @@ class EventBLL(object):
             # force iter to be a long int
             iter = event.get("iter")
             if iter is not None:
-                if model_events:
-                    iter = 0
-                else:
-                    iter = int(iter)
-                    if iter > MAX_LONG or iter < MIN_LONG:
-                        errors_per_type[invalid_iteration_error] += 1
-                        continue
+                iter = int(iter)
+                if iter > MAX_LONG or iter < MIN_LONG:
+                    errors_per_type[invalid_iteration_error] += 1
+                    continue
                 event["iter"] = iter
 
             # used to have "values" to indicate array. no need anymore
@@ -234,7 +239,6 @@ class EventBLL(object):
 
             event["metric"] = event.get("metric") or ""
             event["variant"] = event.get("variant") or ""
-            event["model_event"] = model_events
 
             index_name = get_index_name(company_id, event_type)
             es_action = {
@@ -243,31 +247,33 @@ class EventBLL(object):
                 "_source": event,
             }
 
-            # for "log" events, don't assing custom _id - whatever is sent, is written (not overwritten)
+            # for "log" events, don't assign custom _id - whatever is sent, is written (not overwritten)
             if event_type != EventType.task_log.value:
                 es_action["_id"] = self._get_event_id(event)
             else:
                 es_action["_id"] = dbutils.id()
 
-            task_or_model_ids.add(task_or_model_id)
             if (
                 iter is not None
-                and not model_events
                 and event.get("metric") not in self._skip_iteration_for_metric
             ):
                 task_iteration[task_or_model_id] = max(
                     iter, task_iteration[task_or_model_id]
                 )
 
-            if not model_events:
+            if model_event:
+                used_model_ids.add(task_or_model_id)
+            else:
+                used_task_ids.add(task_or_model_id)
                 self._update_last_metric_events_for_task(
-                    last_events=task_last_events[task_or_model_id], event=event,
+                    last_events=task_last_events[task_or_model_id],
+                    event=event,
+                )
+            if event_type == EventType.metrics_scalar.value:
+                self._update_last_scalar_events_for_task(
+                    last_events=task_last_scalar_events[task_or_model_id],
+                    event=event,
                 )
-                if event_type == EventType.metrics_scalar.value:
-                    self._update_last_scalar_events_for_task(
-                        last_events=task_last_scalar_events[task_or_model_id],
-                        event=event,
-                    )
 
             actions.append(es_action)
 
@@ -305,31 +311,40 @@ class EventBLL(object):
                         else:
                             errors_per_type["Error when indexing events batch"] += 1
 
-                if not model_events:
-                    remaining_tasks = set()
-                    now = datetime.utcnow()
-                    for task_or_model_id in task_or_model_ids:
-                        # Update related tasks. For reasons of performance, we prefer to update
-                        # all of them and not only those who's events were successful
-                        updated = self._update_task(
-                            company_id=company_id,
-                            task_id=task_or_model_id,
-                            now=now,
-                            iter_max=task_iteration.get(task_or_model_id),
-                            last_scalar_events=task_last_scalar_events.get(
-                                task_or_model_id
-                            ),
-                            last_events=task_last_events.get(task_or_model_id),
-                        )
+            now = datetime.utcnow()
+            for model_id in used_model_ids:
+                ModelBLL.update_statistics(
+                    company_id=company_id,
+                    user_id=user_id,
+                    model_id=model_id,
+                    last_update=now,
+                    last_iteration_max=task_iteration.get(model_id),
+                    last_scalar_events=task_last_scalar_events.get(model_id),
+                )
+            remaining_tasks = set()
+            for task_id in used_task_ids:
+                # Update related tasks. For reasons of performance, we prefer to update
+                # all of them and not only those who's events were successful
+                updated = self._update_task(
+                    company_id=company_id,
+                    user_id=user_id,
+                    task_id=task_id,
+                    now=now,
+                    iter_max=task_iteration.get(task_id),
+                    last_scalar_events=task_last_scalar_events.get(task_id),
+                    last_events=task_last_events.get(task_id),
+                )
+                if not updated:
+                    remaining_tasks.add(task_id)
+                    continue
 
-                        if not updated:
-                            remaining_tasks.add(task_or_model_id)
-                            continue
-
-                    if remaining_tasks:
-                        TaskBLL.set_last_update(
-                            remaining_tasks, company_id, last_update=now
-                        )
+            if remaining_tasks:
+                TaskBLL.set_last_update(
+                    remaining_tasks,
+                    company_id=company_id,
+                    user_id=user_id,
+                    last_update=now,
+                )
 
             # this is for backwards compatibility with streaming bulk throwing exception on those
             invalid_iterations_count = errors_per_type.get(invalid_iteration_error)
@@ -424,8 +439,22 @@ class EventBLL(object):
                 for k in ("value", "metric", "variant", "iter", "timestamp")
                 if k in event
             }
-            event_data["min_value"] = min(value, last_event.get("min_value", value))
-            event_data["max_value"] = max(value, last_event.get("max_value", value))
+            last_event_min_value = last_event.get("min_value", value)
+            last_event_min_value_iter = last_event.get("min_value_iter", event_iter)
+            if value < last_event_min_value:
+                event_data["min_value"] = value
+                event_data["min_value_iter"] = event_iter
+            else:
+                event_data["min_value"] = last_event_min_value
+                event_data["min_value_iter"] = last_event_min_value_iter
+            last_event_max_value = last_event.get("max_value", value)
+            last_event_max_value_iter = last_event.get("max_value_iter", event_iter)
+            if value > last_event_max_value:
+                event_data["max_value"] = value
+                event_data["max_value_iter"] = event_iter
+            else:
+                event_data["max_value"] = last_event_max_value
+                event_data["max_value_iter"] = last_event_max_value_iter
             last_events[metric_hash][variant_hash] = event_data
 
     def _update_last_metric_events_for_task(self, last_events, event):
@@ -445,9 +474,10 @@ class EventBLL(object):
 
     def _update_task(
         self,
-        company_id,
-        task_id,
-        now,
+        company_id: str,
+        user_id: str,
+        task_id: str,
+        now: datetime,
         iter_max=None,
         last_scalar_events=None,
         last_events=None,
@@ -463,8 +493,9 @@ class EventBLL(object):
             return False
 
         return TaskBLL.update_statistics(
-            task_id,
-            company_id,
+            task_id=task_id,
+            company_id=company_id,
+            user_id=user_id,
             last_update=now,
             last_iteration_max=iter_max,
             last_scalar_events=last_scalar_events,
@@ -472,7 +503,9 @@ class EventBLL(object):
         )
 
     def _get_event_id(self, event):
-        id_values = (str(event[field]) for field in self.id_fields if field in event)
+        id_values = (
+            str(event[field]) for field in self.event_id_fields if field in event
+        )
         return hashlib.md5("-".join(id_values).encode()).hexdigest()
 
     def scroll_task_events(
@@ -516,29 +549,45 @@ class EventBLL(object):
 
         return events, next_scroll_id, total_events
 
-    def get_last_iterations_per_event_metric_variant(
+    def get_task_plots(
         self,
         company_id: str,
         task_id: str,
-        num_last_iterations: int,
-        event_type: EventType,
+        last_iterations_per_plot: int,
         metric_variants: MetricVariants = None,
     ):
-        if check_empty_data(self.es, company_id=company_id, event_type=event_type):
-            return []
+        event_type = EventType.metrics_plot
+        if check_empty_data(self.es, company_id, event_type):
+            return TaskEventsResult()
 
-        must = [{"term": {"task": task_id}}]
+        plot_valid_condition = {
+            "bool": {
+                "should": [
+                    {"term": {PlotFields.valid_plot: True}},
+                    {
+                        "bool": {
+                            "must_not": {"exists": {"field": PlotFields.valid_plot}}
+                        }
+                    },
+                ]
+            }
+        }
+        must = [plot_valid_condition, {"term": {"task": task_id}}]
         if metric_variants:
             must.append(get_metric_variants_condition(metric_variants))
+
         query = {"bool": {"must": must}}
         search_args = dict(es=self.es, company_id=company_id, event_type=event_type)
         max_metrics, max_variants = get_max_metric_and_variant_counts(
-            query=query, **search_args,
+            query=query,
+            **search_args,
         )
-        max_variants = int(max_variants // num_last_iterations)
+        max_variants = int(max_variants // last_iterations_per_plot)
 
-        es_req: dict = {
+        es_req = {
+            "sort": [{"iter": {"order": "desc"}}],
             "size": 0,
+            "query": query,
             "aggs": {
                 "metrics": {
                     "terms": {
@@ -554,11 +603,10 @@ class EventBLL(object):
                                 "order": {"_key": "asc"},
                             },
                             "aggs": {
-                                "iters": {
-                                    "terms": {
-                                        "field": "iter",
-                                        "size": num_last_iterations,
-                                        "order": {"_key": "desc"},
+                                "events": {
+                                    "top_hits": {
+                                        "sort": {"iter": {"order": "desc"}},
+                                        "size": last_iterations_per_plot,
                                     }
                                 }
                             },
@@ -566,117 +614,23 @@ class EventBLL(object):
                     },
                 }
             },
-            "query": query,
         }
 
         with translate_errors_context():
-            es_res = search_company_events(body=es_req, **search_args)
+            es_response = search_company_events(body=es_req, ignore=404, **search_args)
 
-        if "aggregations" not in es_res:
-            return []
-
-        return [
-            (metric["key"], variant["key"], iter["key"])
-            for metric in es_res["aggregations"]["metrics"]["buckets"]
-            for variant in metric["variants"]["buckets"]
-            for iter in variant["iters"]["buckets"]
-        ]
-
-    def get_task_plots(
-        self,
-        company_id: str,
-        tasks: Sequence[str],
-        last_iterations_per_plot: int = None,
-        sort=None,
-        size: int = 500,
-        scroll_id: str = None,
-        no_scroll: bool = False,
-        metric_variants: MetricVariants = None,
-        model_events: bool = False,
-    ):
-        if scroll_id == self.empty_scroll:
+        aggs_result = es_response.get("aggregations")
+        if not aggs_result:
             return TaskEventsResult()
 
-        if scroll_id:
-            with translate_errors_context():
-                es_res = self.es.scroll(scroll_id=scroll_id, scroll="1h")
-        else:
-            event_type = EventType.metrics_plot
-            if check_empty_data(self.es, company_id=company_id, event_type=event_type):
-                return TaskEventsResult()
-
-            plot_valid_condition = {
-                "bool": {
-                    "should": [
-                        {"term": {PlotFields.valid_plot: True}},
-                        {
-                            "bool": {
-                                "must_not": {"exists": {"field": PlotFields.valid_plot}}
-                            }
-                        },
-                    ]
-                }
-            }
-            must = [plot_valid_condition]
-
-            if last_iterations_per_plot is None or model_events:
-                must.append({"terms": {"task": tasks}})
-                if metric_variants:
-                    must.append(get_metric_variants_condition(metric_variants))
-            else:
-                should = []
-                for i, task_id in enumerate(tasks):
-                    last_iters = self.get_last_iterations_per_event_metric_variant(
-                        company_id=company_id,
-                        task_id=task_id,
-                        num_last_iterations=last_iterations_per_plot,
-                        event_type=event_type,
-                        metric_variants=metric_variants,
-                    )
-                    if not last_iters:
-                        continue
-
-                    for metric, variant, iter in last_iters:
-                        should.append(
-                            {
-                                "bool": {
-                                    "must": [
-                                        {"term": {"task": task_id}},
-                                        {"term": {"metric": metric}},
-                                        {"term": {"variant": variant}},
-                                        {"term": {"iter": iter}},
-                                    ]
-                                }
-                            }
-                        )
-                if not should:
-                    return TaskEventsResult()
-                must.append({"bool": {"should": should}})
-
-            if sort is None:
-                sort = [{"timestamp": {"order": "asc"}}]
-
-            es_req = {
-                "sort": sort,
-                "size": min(size, 10000),
-                "query": {"bool": {"must": must}},
-            }
-
-            with translate_errors_context():
-                es_res = search_company_events(
-                    self.es,
-                    company_id=company_id,
-                    event_type=event_type,
-                    body=es_req,
-                    ignore=404,
-                    **({} if no_scroll else {"scroll": "1h"}),
-                )
-
-        events, total_events, next_scroll_id = self._get_events_from_es_res(es_res)
+        events = [
+            hit["_source"]
+            for metrics_bucket in aggs_result["metrics"]["buckets"]
+            for variants_bucket in metrics_bucket["variants"]["buckets"]
+            for hit in variants_bucket["events"]["hits"]["hits"]
+        ]
         self.uncompress_plots(events)
-        return TaskEventsResult(
-            events=events, next_scroll_id=next_scroll_id, total_events=total_events
-        )
+        return TaskEventsResult(events=events, total_events=len(events))
 
     def _get_events_from_es_res(self, es_res: dict) -> Tuple[list, int, Optional[str]]:
         """
@@ -693,9 +647,11 @@ class EventBLL(object):
         return events, total_events, next_scroll_id
 
     def get_debug_image_urls(
-        self, company_id: str, task_id: str, after_key: dict = None
+        self, company_id: str, task_ids: Sequence[str], after_key: dict = None
     ) -> Tuple[Sequence[str], Optional[dict]]:
-        if check_empty_data(self.es, company_id, EventType.metrics_image):
+        if not task_ids or check_empty_data(
+            self.es, company_id, EventType.metrics_image
+        ):
             return [], None
 
         es_req = {
@@ -711,7 +667,10 @@ class EventBLL(object):
             },
             "query": {
                 "bool": {
-                    "must": [{"term": {"task": task_id}}, {"exists": {"field": "url"}}]
+                    "must": [
+                        {"terms": {"task": task_ids}},
+                        {"exists": {"field": "url"}},
+                    ]
                 }
             },
         }
@@ -729,9 +688,13 @@ class EventBLL(object):
         return [bucket["key"]["url"] for bucket in res["buckets"]], res.get("after_key")
 
     def get_plot_image_urls(
-        self, company_id: str, task_id: str, scroll_id: Optional[str]
+        self, company_id: str, task_ids: Sequence[str], scroll_id: Optional[str]
     ) -> Tuple[Sequence[dict], Optional[str]]:
-        if scroll_id == self.empty_scroll:
+        if (
+            scroll_id == self.empty_scroll
+            or not task_ids
+            or check_empty_data(self.es, company_id, EventType.metrics_plot)
+        ):
             return [], None
 
         if scroll_id:
@@ -746,7 +709,7 @@ class EventBLL(object):
                 "query": {
                     "bool": {
                         "must": [
-                            {"term": {"task": task_id}},
+                            {"terms": {"task": task_ids}},
                             {"exists": {"field": PlotFields.source_urls}},
                         ]
                     }
@@ -765,17 +728,16 @@ class EventBLL(object):
 
     def get_task_events(
         self,
-        company_id: str,
-        task_id: str,
+        company_id: Union[str, Sequence[str]],
+        task_id: Union[str, Sequence[str]],
         event_type: EventType,
-        metric=None,
-        variant=None,
+        metrics: MetricVariants = None,
         last_iter_count=None,
         sort=None,
         size=500,
         scroll_id=None,
         no_scroll=False,
-        model_events=False,
+        last_iters_per_task_metric=False,
     ) -> TaskEventsResult:
         if scroll_id == self.empty_scroll:
             return TaskEventsResult()
@@ -784,38 +746,65 @@ class EventBLL(object):
             with translate_errors_context():
                 es_res = self.es.scroll(scroll_id=scroll_id, scroll="1h")
         else:
-            if check_empty_data(self.es, company_id=company_id, event_type=event_type):
+            company_ids = [company_id] if isinstance(company_id, str) else company_id
+            company_ids = [
+                c_id
+                for c_id in set(company_ids)
+                if not check_empty_data(self.es, c_id, event_type)
+            ]
+            if not company_ids:
                 return TaskEventsResult()
 
             task_ids = [task_id] if isinstance(task_id, str) else task_id
 
             must = []
-            if metric:
-                must.append({"term": {"metric": metric}})
-            if variant:
-                must.append({"term": {"variant": variant}})
+            if metrics:
+                must.append(get_metric_variants_condition(metrics))
 
-            if last_iter_count is None or model_events:
+            if last_iter_count is None:
                 must.append({"terms": {"task": task_ids}})
             else:
-                tasks_iters = self.get_last_iters(
-                    company_id=company_id,
-                    event_type=event_type,
-                    task_id=task_ids,
-                    iters=last_iter_count,
-                )
-                should = [
-                    {
-                        "bool": {
-                            "must": [
-                                {"term": {"task": task}},
-                                {"terms": {"iter": last_iters}},
-                            ]
+                if last_iters_per_task_metric:
+                    task_metric_iters = self.get_last_iters_per_metric(
+                        company_id=company_ids,
+                        event_type=event_type,
+                        task_id=task_ids,
+                        iters=last_iter_count,
+                        metrics=metrics,
+                    )
+                    should = [
+                        {
+                            "bool": {
+                                "must": [
+                                    {"term": {"task": task}},
+                                    {"term": {"metric": metric}},
+                                    {"terms": {"iter": last_iters}},
+                                ]
+                            }
                         }
-                    }
-                    for task, last_iters in tasks_iters.items()
-                    if last_iters
-                ]
+                        for (task, metric), last_iters in task_metric_iters.items()
+                        if last_iters
+                    ]
+                else:
+                    tasks_iters = self.get_last_iters(
+                        company_id=company_ids,
+                        event_type=event_type,
+                        task_id=task_ids,
+                        iters=last_iter_count,
+                        metrics=metrics,
+                    )
+                    should = [
+                        {
+                            "bool": {
+                                "must": [
+                                    {"term": {"task": task}},
+                                    {"terms": {"iter": last_iters}},
+                                ]
+                            }
+                        }
+                        for task, last_iters in tasks_iters.items()
+                        if last_iters
+                    ]
                 if not should:
                     return TaskEventsResult()
                 must.append({"bool": {"should": should}})
@@ -832,7 +821,7 @@ class EventBLL(object):
             with translate_errors_context():
                 es_res = search_company_events(
                     self.es,
-                    company_id=company_id,
+                    company_id=company_ids,
                     event_type=event_type,
                     body=es_req,
                     ignore=404,
@@ -856,7 +845,8 @@ class EventBLL(object):
         query = {"bool": {"must": [{"term": {"task": task_id}}]}}
         search_args = dict(es=self.es, company_id=company_id, event_type=event_type)
         max_metrics, max_variants = get_max_metric_and_variant_counts(
-            query=query, **search_args,
+            query=query,
+            **search_args,
         )
         es_req = {
             "size": 0,
@@ -910,8 +900,10 @@ class EventBLL(object):
         }
         search_args = dict(es=self.es, company_id=company_id, event_type=event_type)
         max_metrics, max_variants = get_max_metric_and_variant_counts(
-            query=query, **search_args,
+            query=query,
+            **search_args,
         )
+        max_variants = int(max_variants // 2)
         es_req = {
             "size": 0,
             "query": query,
@@ -1013,22 +1005,98 @@ class EventBLL(object):
 
         return iterations, vectors
 
-    def get_last_iters(
+    def get_last_iters_per_metric(
         self,
-        company_id: str,
+        company_id: Union[str, Sequence[str]],
         event_type: EventType,
         task_id: Union[str, Sequence[str]],
         iters: int,
-    ) -> Mapping[str, Sequence]:
-        if check_empty_data(self.es, company_id=company_id, event_type=event_type):
+        metrics: MetricVariants = None,
+    ) -> Mapping[Tuple[str, str], Sequence]:
+        company_ids = [company_id] if isinstance(company_id, str) else company_id
+        company_ids = [
+            c_id
+            for c_id in set(company_ids)
+            if not check_empty_data(self.es, c_id, event_type)
+        ]
+        if not company_ids:
             return {}
 
         task_ids = [task_id] if isinstance(task_id, str) else task_id
+        must = [{"terms": {"task": task_ids}}]
+        if metrics:
+            must.append(get_metric_variants_condition(metrics))
+
+        max_tasks = min(len(task_ids), 1000)
+        max_metrics = 10_000 // (max_tasks * iters)
         es_req: dict = {
             "size": 0,
             "aggs": {
                 "tasks": {
-                    "terms": {"field": "task"},
+                    "terms": {"field": "task", "size": max_tasks},
+                    "aggs": {
+                        "metrics": {
+                            "terms": {"field": "metric", "size": max_metrics},
+                            "aggs": {
+                                "iters": {
+                                    "terms": {
+                                        "field": "iter",
+                                        "size": iters,
+                                        "order": {"_key": "desc"},
+                                    }
+                                }
+                            },
+                        }
+                    },
+                }
+            },
+            "query": {"bool": {"must": must}},
+        }
+
+        with translate_errors_context():
+            es_res = search_company_events(
+                self.es,
+                company_id=company_ids,
+                event_type=event_type,
+                body=es_req,
+            )
+        if "aggregations" not in es_res:
+            return {}
+
+        return {
+            (tb["key"], mb["key"]): [ib["key"] for ib in mb["iters"]["buckets"]]
+            for tb in es_res["aggregations"]["tasks"]["buckets"]
+            for mb in tb["metrics"]["buckets"]
+        }
+
+    def get_last_iters(
+        self,
+        company_id: Union[str, Sequence[str]],
+        event_type: EventType,
+        task_id: Union[str, Sequence[str]],
+        iters: int,
+        metrics: MetricVariants = None,
+    ) -> Mapping[str, Sequence]:
+        company_ids = [company_id] if isinstance(company_id, str) else company_id
+        company_ids = [
+            c_id
+            for c_id in set(company_ids)
+            if not check_empty_data(self.es, c_id, event_type)
+        ]
+        if not company_ids:
+            return {}
+
+        task_ids = [task_id] if isinstance(task_id, str) else task_id
+        must = [{"terms": {"task": task_ids}}]
+        if metrics:
+            must.append(get_metric_variants_condition(metrics))
+
+        max_tasks = min(len(task_ids), 1000)
+        es_req: dict = {
+            "size": 0,
+            "aggs": {
+                "tasks": {
+                    "terms": {"field": "task", "size": max_tasks},
                     "aggs": {
                         "iters": {
                             "terms": {
@@ -1040,12 +1108,15 @@ class EventBLL(object):
                     },
                 }
             },
-            "query": {"bool": {"must": [{"terms": {"task": task_ids}}]}},
+            "query": {"bool": {"must": must}},
         }
 
         with translate_errors_context():
             es_res = search_company_events(
-                self.es, company_id=company_id, event_type=event_type, body=es_req,
+                self.es,
+                company_id=company_ids,
+                event_type=event_type,
+                body=es_req,
             )
 
         if "aggregations" not in es_res:
@@ -1096,18 +1167,26 @@ class EventBLL(object):
 
         return {"refresh": True}
 
-    def delete_task_events(
-        self, company_id, task_id, allow_locked=False, model=False, async_delete=False,
-    ):
+    def delete_task_events(self, company_id, task_id, allow_locked=False, model=False):
         if model:
             self._validate_model_state(
-                company_id=company_id, model_id=task_id, allow_locked=allow_locked,
+                company_id=company_id,
+                model_id=task_id,
+                allow_locked=allow_locked,
             )
         else:
             self._validate_task_state(
                 company_id=company_id, task_id=task_id, allow_locked=allow_locked
             )
-
+        async_delete = async_task_events_delete
+        if async_delete:
+            total = self.events_iterator.count_task_events(
+                event_type=EventType.all,
+                company_id=company_id,
+                task_ids=[task_id],
+            )
+            if total <= async_delete_threshold:
+                async_delete = False
         es_req = {"query": {"term": {"task": task_id}}}
         with translate_errors_context():
             es_res = delete_company_events(
@@ -1165,24 +1244,37 @@ class EventBLL(object):
             return es_res.get("deleted", 0)
 
     def delete_multi_task_events(
-        self, company_id: str, task_ids: Sequence[str], async_delete=False
+        self, company_id: str, task_ids: Sequence[str], model=False
     ):
         """
-        Delete mutliple task events. No check is done for tasks write access
+        Delete multiple task events. No check is done for tasks write access
         so it should be checked by the calling code
         """
-        es_req = {"query": {"terms": {"task": task_ids}}}
+        deleted = 0
         with translate_errors_context():
-            es_res = delete_company_events(
-                es=self.es,
-                company_id=company_id,
-                event_type=EventType.all,
-                body=es_req,
-                **self._get_events_deletion_params(async_delete),
-            )
+            async_delete = async_task_events_delete
+            if async_delete and len(task_ids) < 100:
+                total = self.events_iterator.count_task_events(
+                    event_type=EventType.all,
+                    company_id=company_id,
+                    task_ids=task_ids,
+                )
+                if total <= async_delete_threshold:
+                    async_delete = False
+            for tasks in chunked_iter(task_ids, 100):
+                es_req = {"query": {"terms": {"task": tasks}}}
+                es_res = delete_company_events(
+                    es=self.es,
+                    company_id=company_id,
+                    event_type=EventType.all,
+                    body=es_req,
+                    **self._get_events_deletion_params(async_delete),
+                )
+                if not async_delete:
+                    deleted += es_res.get("deleted", 0)
 
         if not async_delete:
-            return es_res.get("deleted", 0)
+            return deleted
 
     def clear_scroll(self, scroll_id: str):
         if scroll_id == self.empty_scroll:

apiserver/bll/event/event_common.py

@@ -8,6 +8,7 @@ from elasticsearch import Elasticsearch
 
 from apiserver.config_repo import config
 from apiserver.database.errors import translate_errors_context
+from apiserver.database.model.task.task import Task
 from apiserver.tools import safe_get
 
 
@@ -22,6 +23,7 @@ class EventType(Enum):
 
 SINGLE_SCALAR_ITERATION = -(2 ** 31)
 MetricVariants = Mapping[str, Sequence[str]]
+TaskCompanies = Mapping[str, Sequence[Task]]
 
 
 class EventSettings:
@@ -52,14 +54,17 @@ class EventSettings:
         return int(self._max_es_allowed_aggregation_buckets * percentage)
 
 
-def get_index_name(company_id: str, event_type: str):
+def get_index_name(company_id: Union[str, Sequence[str]], event_type: str):
     event_type = event_type.lower().replace(" ", "_")
-    return f"events-{event_type}-{company_id.lower()}"
+    if isinstance(company_id, str):
+        company_id = [company_id]
+
+    return ",".join(f"events-{event_type}-{(c_id or '').lower()}" for c_id in company_id)
 
 
 def check_empty_data(es: Elasticsearch, company_id: str, event_type: EventType) -> bool:
     es_index = get_index_name(company_id, event_type.value)
-    if not es.indices.exists(es_index):
+    if not es.indices.exists(index=es_index):
         return True
     return False
 

apiserver/bll/event/event_metrics.py

@@ -18,11 +18,11 @@ from apiserver.bll.event.event_common import (
     get_metric_variants_condition,
     get_max_metric_and_variant_counts,
     SINGLE_SCALAR_ITERATION,
+    TaskCompanies,
 )
 from apiserver.bll.event.scalar_key import ScalarKey, ScalarKeyEnum
 from apiserver.config_repo import config
 from apiserver.database.errors import translate_errors_context
-from apiserver.database.model.task.task import Task
 from apiserver.tools import safe_get
 
 log = config.logger(__file__)
@@ -108,37 +108,51 @@ class EventMetrics:
 
     def compare_scalar_metrics_average_per_iter(
         self,
-        company_id,
-        tasks: Sequence[Task],
+        companies: TaskCompanies,
         samples,
         key: ScalarKeyEnum,
+        metric_variants: MetricVariants = None,
     ):
         """
         Compare scalar metrics for different tasks per metric and variant
         The amount of points in each histogram should not exceed the requested samples
         """
         event_type = EventType.metrics_scalar
-        if check_empty_data(self.es, company_id=company_id, event_type=event_type):
+        companies = {
+            company_id: tasks
+            for company_id, tasks in companies.items()
+            if not check_empty_data(
+                self.es, company_id=company_id, event_type=event_type
+            )
+        }
+        if not companies:
             return {}
 
-        task_name_by_id = {t.id: t.name for t in tasks}
         get_scalar_average_per_iter = partial(
             self._get_scalar_average_per_iter_core,
-            company_id=company_id,
             event_type=event_type,
             samples=samples,
             key=ScalarKey.resolve(key),
+            metric_variants=metric_variants,
             run_parallel=False,
         )
-        task_ids = [t.id for t in tasks]
+        task_ids, company_ids = zip(
+            *(
+                (t.id, t.company)
+                for t in itertools.chain.from_iterable(companies.values())
+            )
+        )
         with ThreadPoolExecutor(max_workers=EventSettings.max_workers) as pool:
             task_metrics = zip(
-                task_ids, pool.map(get_scalar_average_per_iter, task_ids)
+                task_ids, pool.map(get_scalar_average_per_iter, task_ids, company_ids)
             )
 
+        task_names = {
+            t.id: t.name for t in itertools.chain.from_iterable(companies.values())
+        }
         res = defaultdict(lambda: defaultdict(dict))
         for task_id, task_data in task_metrics:
-            task_name = task_name_by_id[task_id]
+            task_name = task_names[task_id]
             for metric_key, metric_data in task_data.items():
                 for variant_key, variant_data in metric_data.items():
                     variant_data["name"] = task_name
@@ -147,18 +161,27 @@ class EventMetrics:
         return res
 
     def get_task_single_value_metrics(
-        self, company_id: str, tasks: Sequence[Task]
+        self, companies: TaskCompanies
     ) -> Mapping[str, dict]:
         """
         For the requested tasks return all the events delivered for the single iteration (-2**31)
         """
-        if check_empty_data(
-            self.es, company_id=company_id, event_type=EventType.metrics_scalar
-        ):
+        companies = {
+            company_id: [t.id for t in tasks]
+            for company_id, tasks in companies.items()
+            if not check_empty_data(
+                self.es, company_id=company_id, event_type=EventType.metrics_scalar
+            )
+        }
+        if not companies:
             return {}
 
-        task_ids = [t.id for t in tasks]
-        task_events = self._get_task_single_value_metrics(company_id, task_ids)
+        with ThreadPoolExecutor(max_workers=EventSettings.max_workers) as pool:
+            task_events = list(
+                itertools.chain.from_iterable(
+                    pool.map(self._get_task_single_value_metrics, companies.items())
+                ),
+            )
 
         def _get_value(event: dict):
             return {
@@ -172,8 +195,9 @@ class EventMetrics:
         }
 
     def _get_task_single_value_metrics(
-        self, company_id: str, task_ids: Sequence[str]
+        self, tasks: Tuple[str, Sequence[str]]
     ) -> Sequence[dict]:
+        company_id, task_ids = tasks
         es_req = {
             "size": 10000,
             "query": {

apiserver/bll/event/events_iterator.py

@@ -64,13 +64,13 @@ class EventsIterator:
         self,
         event_type: EventType,
         company_id: str,
-        task_id: str,
+        task_ids: Sequence[str],
         metric_variants: MetricVariants = None,
     ) -> int:
         if check_empty_data(self.es, company_id, event_type):
             return 0
 
-        query, _ = self._get_initial_query_and_must(task_id, metric_variants)
+        query, _ = self._get_initial_query_and_must(task_ids, metric_variants)
         es_req = {
             "query": query,
         }
@@ -100,7 +100,7 @@ class EventsIterator:
         For the last key-field value all the events are brought (even if the resulting size exceeds batch_size)
         so that events with this value will not be lost between the calls.
         """
-        query, must = self._get_initial_query_and_must(task_id, metric_variants)
+        query, must = self._get_initial_query_and_must([task_id], metric_variants)
 
         # retrieve the next batch of events
         es_req = {
@@ -158,14 +158,14 @@ class EventsIterator:
 
     @staticmethod
     def _get_initial_query_and_must(
-        task_id: str, metric_variants: MetricVariants = None
+        task_ids: Sequence[str], metric_variants: MetricVariants = None
     ) -> Tuple[dict, list]:
         if not metric_variants:
-            must = [{"term": {"task": task_id}}]
-            query = {"term": {"task": task_id}}
+            query = {"terms": {"task": task_ids}}
+            must = [query]
         else:
             must = [
-                {"term": {"task": task_id}},
+                {"terms": {"task": task_ids}},
                 get_metric_variants_condition(metric_variants),
             ]
             query = {"bool": {"must": must}}

apiserver/bll/event/metric_events_iterator.py

@@ -75,18 +75,25 @@ class MetricEventsIterator:
 
     def get_task_events(
         self,
-        company_id: str,
+        companies: Mapping[str, str],
         task_metrics: Mapping[str, dict],
         iter_count: int,
         navigate_earlier: bool = True,
         refresh: bool = False,
         state_id: str = None,
     ) -> MetricEventsResult:
-        if check_empty_data(self.es, company_id, self.event_type):
+        companies = {
+            task_id: company_id
+            for task_id, company_id in companies.items()
+            if not check_empty_data(
+                self.es, company_id=company_id, event_type=self.event_type
+            )
+        }
+        if not companies:
             return MetricEventsResult()
 
         def init_state(state_: MetricEventsScrollState):
-            state_.tasks = self._init_task_states(company_id, task_metrics)
+            state_.tasks = self._init_task_states(companies, task_metrics)
 
         def validate_state(state_: MetricEventsScrollState):
             """
@@ -95,7 +102,7 @@ class MetricEventsIterator:
             Refresh the state if requested
             """
             if refresh:
-                self._reinit_outdated_task_states(company_id, state_, task_metrics)
+                self._reinit_outdated_task_states(companies, state_, task_metrics)
 
         with self.cache_manager.get_or_create_state(
             state_id=state_id, init_state=init_state, validate_state=validate_state
@@ -112,7 +119,7 @@ class MetricEventsIterator:
                     pool.map(
                         partial(
                             self._get_task_metric_events,
-                            company_id=company_id,
+                            companies=companies,
                             iter_count=iter_count,
                             navigate_earlier=navigate_earlier,
                             specific_variants_requested=specific_variants_requested,
@@ -125,7 +132,7 @@ class MetricEventsIterator:
 
     def _reinit_outdated_task_states(
         self,
-        company_id,
+        companies: Mapping[str, str],
         state: MetricEventsScrollState,
         task_metrics: Mapping[str, dict],
     ):
@@ -133,9 +140,7 @@ class MetricEventsIterator:
         Determine the metrics for which new event_type events were added
         since their states were initialized and re-init these states
         """
-        tasks = Task.objects(id__in=list(task_metrics), company=company_id).only(
-            "id", "metric_stats"
-        )
+        tasks = Task.objects(id__in=list(task_metrics)).only("id", "metric_stats")
 
         def get_last_update_times_for_task_metrics(
             task: Task,
@@ -175,7 +180,7 @@ class MetricEventsIterator:
             if metrics_to_recalc:
                 task_metrics_to_recalc[task] = metrics_to_recalc
 
-        updated_task_states = self._init_task_states(company_id, task_metrics_to_recalc)
+        updated_task_states = self._init_task_states(companies, task_metrics_to_recalc)
 
         def merge_with_updated_task_states(
             old_state: TaskScrollState, updates: Sequence[TaskScrollState]
@@ -205,14 +210,14 @@ class MetricEventsIterator:
         ]
 
     def _init_task_states(
-        self, company_id: str, task_metrics: Mapping[str, dict]
+        self, companies: Mapping[str, str], task_metrics: Mapping[str, dict]
     ) -> Sequence[TaskScrollState]:
         """
         Returned initialized metric scroll stated for the requested task metrics
         """
         with ThreadPoolExecutor(EventSettings.max_workers) as pool:
             task_metric_states = pool.map(
-                partial(self._init_metric_states_for_task, company_id=company_id),
+                partial(self._init_metric_states_for_task, companies=companies),
                 task_metrics.items(),
             )
 
@@ -232,13 +237,14 @@ class MetricEventsIterator:
         pass
 
     def _init_metric_states_for_task(
-        self, task_metrics: Tuple[str, dict], company_id: str
+        self, task_metrics: Tuple[str, dict], companies: Mapping[str, str]
     ) -> Sequence[MetricState]:
         """
         Return metric scroll states for the task filled with the variant states
         for the variants that reported any event_type events
         """
         task, metrics = task_metrics
+        company_id = companies[task]
         must = [{"term": {"task": task}}, *self._get_extra_conditions()]
         if metrics:
             must.append(get_metric_variants_condition(metrics))
@@ -319,7 +325,7 @@ class MetricEventsIterator:
     def _get_task_metric_events(
         self,
         task_state: TaskScrollState,
-        company_id: str,
+        companies: Mapping[str, str],
         iter_count: int,
         navigate_earlier: bool,
         specific_variants_requested: bool,
@@ -391,7 +397,10 @@ class MetricEventsIterator:
         }
         with translate_errors_context():
             es_res = search_company_events(
-                self.es, company_id=company_id, event_type=self.event_type, body=es_req,
+                self.es,
+                company_id=companies[task_state.task],
+                event_type=self.event_type,
+                body=es_req,
             )
         if "aggregations" not in es_res:
             return task_state.task, []

apiserver/bll/model/__init__.py

@@ -5,7 +5,8 @@ from mongoengine import Q
 
 from apiserver.apierrors import errors
 from apiserver.apimodels.models import ModelTaskPublishResponse
-from apiserver.bll.task.utils import deleted_prefix
+from apiserver.bll.task.utils import deleted_prefix, get_last_metric_updates
+from apiserver.config_repo import config
 from apiserver.database.model import EntityVisibility
 from apiserver.database.model.model import Model
 from apiserver.database.model.task.task import Task, TaskStatus
@@ -13,6 +14,8 @@ from .metadata import Metadata
 
 
 class ModelBLL:
+    event_bll = None
+
     @classmethod
     def get_company_model_by_id(
         cls, company_id: str, model_id: str, only_fields=None
@@ -28,11 +31,7 @@ class ModelBLL:
 
     @staticmethod
     def assert_exists(
-        company_id,
-        model_ids,
-        only=None,
-        allow_public=False,
-        return_models=True,
+        company_id, model_ids, only=None, allow_public=False, return_models=True,
     ) -> Optional[Sequence[Model]]:
         model_ids = [model_ids] if isinstance(model_ids, str) else model_ids
         ids = set(model_ids)
@@ -58,8 +57,9 @@ class ModelBLL:
         cls,
         model_id: str,
         company_id: str,
+        user_id: str,
         force_publish_task: bool = False,
-        publish_task_func: Callable[[str, str, bool], dict] = None,
+        publish_task_func: Callable[[str, str, str, bool], dict] = None,
     ) -> Tuple[int, ModelTaskPublishResponse]:
         model = cls.get_company_model_by_id(company_id=company_id, model_id=model_id)
         if model.ready:
@@ -74,18 +74,25 @@ class ModelBLL:
             )
             if task and task.status != TaskStatus.published:
                 task_publish_res = publish_task_func(
-                    model.task, company_id, force_publish_task
+                    model.task, company_id, user_id, force_publish_task
                 )
                 published_task = ModelTaskPublishResponse(
                     id=model.task, data=task_publish_res
                 )
 
-        updated = model.update(upsert=False, ready=True, last_update=datetime.utcnow())
+        now = datetime.utcnow()
+        updated = model.update(
+            upsert=False,
+            ready=True,
+            last_update=now,
+            last_change=now,
+            last_changed_by=user_id,
+        )
         return updated, published_task
 
     @classmethod
     def delete_model(
-        cls, model_id: str, company_id: str, force: bool
+        cls, model_id: str, company_id: str, user_id: str, force: bool, delete_external_artifacts: bool = True,
     ) -> Tuple[int, Model]:
         model = cls.get_company_model_by_id(
             company_id=company_id,
@@ -111,49 +118,88 @@ class ModelBLL:
 
         if model.task:
             task = Task.objects(id=model.task).first()
-            if task and task.status == TaskStatus.published:
-                if not force:
-                    raise errors.bad_request.ModelCreatingTaskExists(
-                        "and published, use force=True to delete", task=model.task
-                    )
-                if task.models.output and model_id in task.models.output:
-                    now = datetime.utcnow()
+            if task:
+                now = datetime.utcnow()
+                if task.status == TaskStatus.published:
+                    if not force:
+                        raise errors.bad_request.ModelCreatingTaskExists(
+                            "and published, use force=True to delete", task=model.task
+                        )
                     Task._get_collection().update_one(
                         filter={"_id": model.task, "models.output.model": model_id},
                         update={
                             "$set": {
                                 "models.output.$[elem].model": deleted_model_id,
                                 "output.error": f"model deleted on {now.isoformat()}",
+                                "last_change": now,
+                                "last_changed_by": user_id,
                             },
-                            "last_change": now,
                         },
                         array_filters=[{"elem.model": model_id}],
                         upsert=False,
                     )
+                else:
+                    task.update(
+                        pull__models__output__model=model_id,
+                        set__last_change=now,
+                        set__last_changed_by=user_id,
+                    )
 
+        delete_external_artifacts = delete_external_artifacts and config.get(
+            "services.async_urls_delete.enabled", True
+        )
+        if delete_external_artifacts:
+            from apiserver.bll.task.task_cleanup import (
+                collect_debug_image_urls,
+                collect_plot_image_urls,
+                _schedule_for_delete,
+            )
+            urls = set()
+            urls.update(collect_debug_image_urls(company_id, model_id))
+            urls.update(collect_plot_image_urls(company_id, model_id))
+            if model.uri:
+                urls.add(model.uri)
+            if urls:
+                _schedule_for_delete(
+                    task_id=model_id,
+                    company=company_id,
+                    user=user_id,
+                    urls=urls,
+                    can_delete_folders=False,
+                )
+
+        if not cls.event_bll:
+            from apiserver.bll.event import EventBLL
+            cls.event_bll = EventBLL()
+
+        cls.event_bll.delete_task_events(company_id, model_id, allow_locked=True, model=True)
         del_count = Model.objects(id=model_id, company=company_id).delete()
         return del_count, model
 
     @classmethod
-    def archive_model(cls, model_id: str, company_id: str):
+    def archive_model(cls, model_id: str, company_id: str, user_id: str):
         cls.get_company_model_by_id(
             company_id=company_id, model_id=model_id, only_fields=("id",)
         )
+        now = datetime.utcnow()
         archived = Model.objects(company=company_id, id=model_id).update(
             add_to_set__system_tags=EntityVisibility.archived.value,
-            last_update=datetime.utcnow(),
+            last_change=now,
+            last_changed_by=user_id,
         )
 
         return archived
 
     @classmethod
-    def unarchive_model(cls, model_id: str, company_id: str):
+    def unarchive_model(cls, model_id: str, company_id: str, user_id: str):
         cls.get_company_model_by_id(
             company_id=company_id, model_id=model_id, only_fields=("id",)
         )
+        now = datetime.utcnow()
         unarchived = Model.objects(company=company_id, id=model_id).update(
             pull__system_tags=EntityVisibility.archived.value,
-            last_update=datetime.utcnow(),
+            last_change=now,
+            last_changed_by=user_id,
         )
 
         return unarchived
@@ -178,12 +224,43 @@ class ModelBLL:
                         "labels_count": {"$size": {"$objectToArray": "$labels"}}
                     }
                 },
-                {
-                    "$project": {"labels_count": 1},
-                },
+                {"$project": {"labels_count": 1}},
             ]
         )
-        return {
-            r.pop("_id"): r
-            for r in result
+        return {r.pop("_id"): r for r in result}
+
+    @staticmethod
+    def update_statistics(
+        company_id: str,
+        user_id: str,
+        model_id: str,
+        last_update: datetime = None,
+        last_iteration_max: int = None,
+        last_scalar_events: Dict[str, Dict[str, dict]] = None,
+    ):
+        last_update = last_update or datetime.utcnow()
+        updates = {
+            "last_update": datetime.utcnow(),
+            "last_change": last_update,
+            "last_changed_by": user_id,
         }
+        if last_iteration_max is not None:
+            updates.update(max__last_iteration=last_iteration_max)
+
+        raw_updates = {}
+        if last_scalar_events is not None:
+            raw_updates = {}
+            if last_scalar_events is not None:
+                get_last_metric_updates(
+                    task_id=model_id,
+                    last_scalar_events=last_scalar_events,
+                    raw_updates=raw_updates,
+                    extra_updates=updates,
+                    model_events=True,
+                )
+
+        ret = Model.objects(id=model_id).update_one(**updates)
+        if ret and raw_updates:
+            Model.objects(id=model_id).update_one(__raw__=[{"$set": raw_updates}])
+
+        return ret

apiserver/bll/model/metadata.py

@@ -5,7 +5,6 @@ from mongoengine import Document
 from apiserver.apierrors import errors
 from apiserver.apimodels.metadata import MetadataItem
 from apiserver.database.model.base import GetMixin
-from apiserver.service_repo import APICall
 from apiserver.utilities.parameter_key_escaper import (
     ParameterKeyEscaper,
     mongoengine_safe,
@@ -87,13 +86,13 @@ class Metadata:
         return paths
 
     @classmethod
-    def escape_query_parameters(cls, call: APICall) -> dict:
-        if not call.data:
-            return call.data
+    def escape_query_parameters(cls, call_data: dict) -> dict:
+        if not call_data:
+            return call_data
 
-        keys = list(call.data)
+        keys = list(call_data)
         call_data = {
-            safe_key: call.data[key]
+            safe_key: call_data[key]
             for key, safe_key in zip(keys, Metadata.escape_paths(keys))
         }
 

apiserver/bll/organization/__init__.py

@@ -1,8 +1,11 @@
 from collections import defaultdict
 from enum import Enum
-from typing import Sequence, Dict
+from typing import Sequence, Dict, Type
 
+from apiserver.apierrors import errors
+from apiserver.bll.util import update_project_time
 from apiserver.config_repo import config
+from apiserver.database.model.model import AttributedDocument
 from apiserver.database.model.model import Model
 from apiserver.database.model.task.task import Task
 from apiserver.redis_manager import redman
@@ -22,6 +25,51 @@ class OrgBLL:
         self._task_tags = _TagsCache(Task, self.redis)
         self._model_tags = _TagsCache(Model, self.redis)
 
+    def edit_entity_tags(
+        self,
+        company_id,
+        entity_cls: Type[AttributedDocument],
+        entity_ids: Sequence[str],
+        add_tags: Sequence[str],
+        remove_tags: Sequence[str],
+    ) -> int:
+        if entity_cls not in (Task, Model):
+            raise errors.bad_request.ValidationError(
+                "Tags editing can be called on tasks or models only"
+            )
+        if not entity_ids:
+            raise errors.bad_request.ValidationError(
+                "No entity ids provided for editing tags"
+            )
+        if not (add_tags or remove_tags):
+            raise errors.bad_request.ValidationError(
+                "Either add tags or remove tags should be provided"
+            )
+
+        updated = 0
+        if add_tags:
+            updated += entity_cls.objects(company=company_id, id__in=entity_ids).update(
+                add_to_set__tags=add_tags
+            )
+        if remove_tags:
+            updated += entity_cls.objects(company=company_id, id__in=entity_ids).update(
+                pull_all__tags=remove_tags
+            )
+        if not updated:
+            return 0
+
+        projects = entity_cls.objects(company=company_id, id__in=entity_ids).distinct(
+            "project"
+        )
+        update_project_time(project_ids=projects)
+        self.update_tags(
+            company_id,
+            entity=Tags.Task if entity_cls is Task else Tags.Model,
+            projects=projects,
+            tags=add_tags or remove_tags
+        )
+        return updated
+
     def get_tags(
         self,
         company_id: str,
@@ -50,10 +98,10 @@ class OrgBLL:
         return ret
 
     def update_tags(
-        self, company_id: str, entity: Tags, project: str, tags=None, system_tags=None,
+        self, company_id: str, entity: Tags, projects: Sequence[str], tags=None, system_tags=None,
     ):
         tags_cache = self._get_tags_cache_for_entity(entity)
-        tags_cache.update_tags(company_id, project, tags, system_tags)
+        tags_cache.update_tags(company_id, projects, tags, system_tags)
 
     def reset_tags(self, company_id: str, entity: Tags, projects: Sequence[str]):
         tags_cache = self._get_tags_cache_for_entity(entity)

apiserver/bll/organization/tags_cache.py

@@ -107,7 +107,7 @@ class _TagsCache:
 
         return ret
 
-    def update_tags(self, company_id: str, project: str, tags=None, system_tags=None):
+    def update_tags(self, company_id: str, projects: Sequence[str], tags=None, system_tags=None):
         """
         Updates tags. If reset is set then both tags and system_tags
         are recalculated. Otherwise only those that are not 'None'
@@ -123,7 +123,7 @@ class _TagsCache:
         if not fields:
             return
 
-        self._delete_redis_keys(company_id, projects=[project], fields=fields)
+        self._delete_redis_keys(company_id, projects=projects, fields=fields)
 
     def reset_tags(self, company_id: str, projects: Sequence[str]):
         self._delete_redis_keys(

apiserver/bll/project/project_bll.py

@@ -1,8 +1,7 @@
-import itertools
 from collections import defaultdict
 from datetime import datetime, timedelta
 from functools import reduce
-from itertools import groupby
+from itertools import groupby, chain
 from operator import itemgetter
 from typing import (
     Sequence,
@@ -15,15 +14,16 @@ from typing import (
     Callable,
     Mapping,
     Any,
+    Union,
 )
 
-from boltons.iterutils import partition
 from mongoengine import Q, Document
 
 from apiserver import database
 from apiserver.apierrors import errors
+from apiserver.apimodels.projects import ProjectChildrenType
 from apiserver.config_repo import config
-from apiserver.database.model import EntityVisibility, AttributedDocument
+from apiserver.database.model import EntityVisibility, AttributedDocument, User
 from apiserver.database.model.base import GetMixin
 from apiserver.database.model.model import Model
 from apiserver.database.model.project import Project
@@ -40,16 +40,25 @@ from .sub_projects import (
     _ids_with_children,
     _ids_with_parents,
     _get_project_depth,
+    ProjectsChildren,
 )
 
 log = config.logger(__file__)
 max_depth = config.get("services.projects.sub_projects.max_depth", 10)
+reports_project_name = ".reports"
+datasets_project_name = ".datasets"
+pipelines_project_name = ".pipelines"
+reports_tag = "reports"
+dataset_tag = "dataset"
+pipeline_tag = "pipeline"
 
 
 class ProjectBLL:
+    child_classes = (Task, Model)
+
     @classmethod
     def merge_project(
-        cls, company, source_id: str, destination_id: str
+        cls, company: str, source_id: str, destination_id: str
     ) -> Tuple[int, int, Set[str]]:
         """
         Move all the tasks and sub projects from the source project to the destination
@@ -81,7 +90,7 @@ class ProjectBLL:
             )
 
         moved_entities = 0
-        for entity_type in (Task, Model):
+        for entity_type in cls.child_classes:
             moved_entities += entity_type.objects(
                 company=company,
                 project=source_id,
@@ -297,18 +306,21 @@ class ProjectBLL:
         """
         Move a batch of entities to `project` or a project named `project_name` (create if does not exist)
         """
-        project = cls.find_or_create(
-            user=user,
-            company=company,
-            project_id=project,
-            project_name=project_name,
-            description="",
-        )
-        extra = (
-            {"set__last_change": datetime.utcnow()}
-            if hasattr(entity_cls, "last_change")
-            else {}
-        )
+        if project_name or project:
+            project = cls.find_or_create(
+                user=user,
+                company=company,
+                project_id=project,
+                project_name=project_name,
+                description="",
+            )
+
+        extra = {}
+        if hasattr(entity_cls, "last_change"):
+            extra["set__last_change"] = datetime.utcnow()
+        if hasattr(entity_cls, "last_changed_by"):
+            extra["set__last_changed_by"] = user
+
         entity_cls.objects(company=company, id__in=ids).update(
             set__project=project, **extra
         )
@@ -394,6 +406,18 @@ class ProjectBLL:
                                 "$completed",
                                 {"$gt": ["$completed", time_thresh]},
                                 additional_cond,
+                                {
+                                    "$not": {
+                                        "$in": [
+                                            "$status",
+                                            [
+                                                TaskStatus.queued,
+                                                TaskStatus.in_progress,
+                                                TaskStatus.failed,
+                                            ],
+                                        ]
+                                    }
+                                },
                             ]
                         },
                         "then": 1,
@@ -507,7 +531,7 @@ class ProjectBLL:
     def aggregate_project_data(
         func: Callable[[T, T], T],
         project_ids: Sequence[str],
-        child_projects: Mapping[str, Sequence[Project]],
+        child_projects: ProjectsChildren,
         data: Mapping[str, T],
     ) -> Dict[str, T]:
         """
@@ -527,7 +551,10 @@ class ProjectBLL:
 
     @classmethod
     def get_dataset_stats(
-        cls, company: str, project_ids: Sequence[str], users: Sequence[str] = None,
+        cls,
+        company: str,
+        project_ids: Sequence[str],
+        users: Sequence[str] = None,
     ) -> Dict[str, dict]:
         if not project_ids:
             return {}
@@ -559,6 +586,140 @@ class ProjectBLL:
             for r in Task.aggregate(task_runtime_pipeline)
         }
 
+    @staticmethod
+    def _get_projects_children(
+        project_ids: Sequence[str],
+        search_hidden: bool,
+        allowed_ids: Sequence[str],
+    ) -> Tuple[ProjectsChildren, Set[str]]:
+        child_projects = _get_sub_projects(
+            project_ids,
+            _only=("id", "name"),
+            search_hidden=search_hidden,
+            allowed_ids=allowed_ids,
+        )
+        return (
+            child_projects,
+            {c.id for c in chain.from_iterable(child_projects.values())},
+        )
+
+    @staticmethod
+    def _get_children_info(
+        project_ids: Sequence[str], child_projects: ProjectsChildren
+    ) -> dict:
+        return {
+            project: sorted(
+                [{"id": c.id, "name": c.name} for c in child_projects.get(project, [])],
+                key=itemgetter("name"),
+            )
+            for project in project_ids
+        }
+
+    @classmethod
+    def _get_project_dataset_stats_core(
+        cls,
+        company: str,
+        project_ids: Sequence[str],
+        project_field: str,
+        entity_class: Type[AttributedDocument],
+        include_children: bool = True,
+        filter_: Mapping[str, Any] = None,
+        users: Sequence[str] = None,
+        selected_project_ids: Sequence[str] = None,
+    ) -> Tuple[Dict[str, dict], Dict[str, dict]]:
+        if not project_ids:
+            return {}, {}
+
+        child_projects = {}
+        project_ids_with_children = set(project_ids)
+        if include_children:
+            child_projects, children_ids = cls._get_projects_children(
+                project_ids,
+                search_hidden=True,
+                allowed_ids=selected_project_ids,
+            )
+            project_ids_with_children |= children_ids
+
+        pipeline = [
+            {
+                "$match": cls.get_match_conditions(
+                    company=company,
+                    project_ids=list(project_ids_with_children),
+                    filter_=filter_,
+                    users=users,
+                    project_field=project_field,
+                )
+            },
+            {"$project": {project_field: 1, "tags": 1}},
+            {
+                "$group": {
+                    "_id": f"${project_field}",
+                    "count": {"$sum": 1},
+                    "tags": {"$push": "$tags"},
+                }
+            },
+        ]
+        res = entity_class.aggregate(pipeline)
+
+        project_stats = {
+            result["_id"]: {
+                "count": result.get("count", 0),
+                "tags": set(chain.from_iterable(result.get("tags", []))),
+            }
+            for result in res
+        }
+
+        def concat_dataset_stats(a: dict, b: dict) -> dict:
+            return {
+                "count": a.get("count", 0) + b.get("count", 0),
+                "tags": a.get("tags", {}) | b.get("tags", {}),
+            }
+
+        top_project_stats = cls.aggregate_project_data(
+            func=concat_dataset_stats,
+            project_ids=project_ids,
+            child_projects=child_projects,
+            data=project_stats,
+        )
+        for _, stat in top_project_stats.items():
+            stat["tags"] = sorted(list(stat.get("tags", {})))
+
+        empty_stats = {"count": 0, "tags": []}
+        stats = {
+            project: {"datasets": top_project_stats.get(project, empty_stats)}
+            for project in project_ids
+        }
+        return stats, cls._get_children_info(project_ids, child_projects)
+
+    @classmethod
+    def get_project_dataset_stats(
+        cls,
+        company: str,
+        project_ids: Sequence[str],
+        include_children: bool = True,
+        filter_: Mapping[str, Any] = None,
+        users: Sequence[str] = None,
+        selected_project_ids: Sequence[str] = None,
+    ) -> Tuple[Dict[str, dict], Dict[str, dict]]:
+        filter_ = filter_ or {}
+        filter_system_tags = filter_.get("system_tags")
+        if not isinstance(filter_system_tags, list):
+            filter_system_tags = []
+        if dataset_tag not in filter_system_tags:
+            filter_system_tags.append(dataset_tag)
+            filter_["system_tags"] = filter_system_tags
+
+        return cls._get_project_dataset_stats_core(
+            company=company,
+            project_ids=project_ids,
+            project_field="parent",
+            entity_class=Project,
+            include_children=include_children,
+            filter_=filter_,
+            users=users,
+            selected_project_ids=selected_project_ids,
+        )
+
     @classmethod
     def get_project_stats(
         cls,
@@ -569,24 +730,21 @@ class ProjectBLL:
         search_hidden: bool = False,
         filter_: Mapping[str, Any] = None,
         users: Sequence[str] = None,
-        user_active_project_ids: Sequence[str] = None,
+        selected_project_ids: Sequence[str] = None,
     ) -> Tuple[Dict[str, dict], Dict[str, dict]]:
         if not project_ids:
             return {}, {}
 
-        child_projects = (
-            _get_sub_projects(
+        child_projects = {}
+        project_ids_with_children = set(project_ids)
+        if include_children:
+            child_projects, children_ids = cls._get_projects_children(
                 project_ids,
-                _only=("id", "name"),
                 search_hidden=search_hidden,
-                allowed_ids=user_active_project_ids,
+                allowed_ids=selected_project_ids,
             )
-            if include_children
-            else {}
-        )
-        project_ids_with_children = set(project_ids) | {
-            c.id for c in itertools.chain.from_iterable(child_projects.values())
-        }
+            project_ids_with_children |= children_ids
+
         status_count_pipeline, runtime_pipeline = cls.make_projects_get_all_pipelines(
             company,
             project_ids=list(project_ids_with_children),
@@ -690,14 +848,7 @@ class ProjectBLL:
             for project in project_ids
         }
 
-        children = {
-            project: sorted(
-                [{"id": c.id, "name": c.name} for c in child_projects.get(project, [])],
-                key=itemgetter("name"),
-            )
-            for project in project_ids
-        }
-        return stats, children
+        return stats, cls._get_children_info(project_ids, child_projects)
 
     @classmethod
     def get_active_users(
@@ -722,7 +873,7 @@ class ProjectBLL:
             projects_query &= Q(id__in=project_ids)
 
         res = set(Project.objects(projects_query).distinct(field="user"))
-        for cls_ in (Task, Model):
+        for cls_ in cls.child_classes:
             res |= set(cls_.objects(query).distinct(field="user"))
 
         return res
@@ -751,61 +902,102 @@ class ProjectBLL:
         return tags, system_tags
 
     @classmethod
-    def get_projects_with_active_user(
+    def get_projects_with_selected_children(
         cls,
         company: str,
-        users: Sequence[str],
+        users: Sequence[str] = None,
         project_ids: Optional[Sequence[str]] = None,
         allow_public: bool = True,
+        children_type: ProjectChildrenType = None,
+        children_tags: Sequence[str] = None,
+        children_tags_filter: dict = None,
     ) -> Tuple[Sequence[str], Sequence[str]]:
         """
-        Get the projects ids where user created any tasks including all the parents of these projects
+        Get the projects ids matching children_condition (if passed) or where the passed user created any tasks
+        including all the parents of these projects
         If project ids are specified then filter the results by these project ids
         """
-        query = Q(user__in=users)
+        if not (users or children_type):
+            raise errors.bad_request.ValidationError(
+                "Either active users or children_condition should be specified"
+            )
 
-        if allow_public:
-            query &= get_company_or_none_constraint(company)
+        query = (
+            get_company_or_none_constraint(company)
+            if allow_public
+            else Q(company=company)
+        )
+        if users:
+            query &= Q(user__in=users)
+
+        project_query = None
+        if children_tags_filter:
+            child_query = query & GetMixin.get_list_filter_query(
+                "tags", children_tags_filter
+            )
+        elif children_tags:
+            child_query = query & GetMixin.get_list_field_query("tags", children_tags)
         else:
-            query &= Q(company=company)
+            child_query = query
+
+        if children_type == ProjectChildrenType.dataset:
+            child_queries = {
+                Project: child_query
+                & Q(system_tags__in=[dataset_tag], basename__ne=datasets_project_name)
+            }
+        elif children_type == ProjectChildrenType.pipeline:
+            child_queries = {
+                Project: child_query
+                & Q(system_tags__in=[pipeline_tag], basename__ne=pipelines_project_name)
+            }
+        elif children_type == ProjectChildrenType.report:
+            child_queries = {Task: child_query & Q(system_tags__in=[reports_tag])}
+        else:
+            project_query = query
+            child_queries = {entity_cls: query for entity_cls in cls.child_classes}
 
-        user_projects_query = query
         if project_ids:
             ids_with_children = _ids_with_children(project_ids)
-            query &= Q(project__in=ids_with_children)
-            user_projects_query &= Q(id__in=ids_with_children)
+            if project_query:
+                project_query &= Q(id__in=ids_with_children)
+            for child_cls in child_queries:
+                child_queries[child_cls] &= (
+                    Q(parent__in=ids_with_children)
+                    if child_cls is Project
+                    else Q(project__in=ids_with_children)
+                )
 
-        res = {p.id for p in Project.objects(user_projects_query).only("id")}
-        for cls_ in (Task, Model):
-            res |= set(cls_.objects(query).distinct(field="project"))
+        res = (
+            set(Project.objects(project_query).scalar("id")) if project_query else set()
+        )
+        for cls_, query_ in child_queries.items():
+            res |= set(
+                cls_.objects(query_).distinct(
+                    field="id" if cls_ is Project else "project"
+                )
+            )
 
         res = list(res)
         if not res:
             return res, res
 
-        user_active_project_ids = _ids_with_parents(res)
+        selected_project_ids = _ids_with_parents(res)
         filtered_ids = (
-            list(set(user_active_project_ids) & set(project_ids))
+            list(set(selected_project_ids) & set(project_ids))
             if project_ids
-            else list(user_active_project_ids)
+            else list(selected_project_ids)
         )
 
-        return filtered_ids, user_active_project_ids
+        return filtered_ids, selected_project_ids
 
-    @classmethod
-    def get_task_parents(
-        cls,
-        company_id: str,
-        projects: Sequence[str],
-        include_subprojects: bool,
+    @staticmethod
+    def _get_project_query(
+        company: str,
+        projects: Sequence,
+        include_subprojects: bool = True,
         state: Optional[EntityVisibility] = None,
-    ) -> Sequence[dict]:
-        """
-        Get list of unique parent tasks sorted by task name for the passed company projects
-        If projects is None or empty then get parents for all the company tasks
-        """
-        query = Q(company=company_id)
-
+    ) -> Q:
+        query = get_company_or_none_constraint(company)
         if projects:
             if include_subprojects:
                 projects = _ids_with_children(projects)
@@ -818,6 +1010,25 @@ class ProjectBLL:
         elif state == EntityVisibility.active:
             query &= Q(system_tags__nin=[EntityVisibility.archived.value])
 
+        return query
+
+    @classmethod
+    def get_task_parents(
+        cls,
+        company_id: str,
+        projects: Sequence[str],
+        include_subprojects: bool,
+        state: Optional[EntityVisibility] = None,
+        name: str = None,
+    ) -> Sequence[dict]:
+        """
+        Get list of unique parent tasks sorted by task name for the passed company projects
+        If projects is None or empty then get parents for all the company tasks
+        """
+        query = cls._get_project_query(
+            company_id, projects, include_subprojects=include_subprojects, state=state
+        )
+
         parent_ids = set(Task.objects(query).distinct("parent"))
         if not parent_ids:
             return []
@@ -825,23 +1036,37 @@ class ProjectBLL:
         parents = Task.get_many_with_join(
             company_id,
             query=Q(id__in=parent_ids),
+            query_dict={"name": name} if name else None,
             allow_public=True,
             override_projection=("id", "name", "project.name"),
         )
+
         return sorted(parents, key=itemgetter("name"))
 
+    @classmethod
+    def get_entity_users(
+        cls,
+        company: str,
+        entity_cls: Type[Union[Task, Model]],
+        projects: Sequence[str],
+        include_subprojects: bool,
+    ) -> Sequence[dict]:
+        query = cls._get_project_query(
+            company, projects, include_subprojects=include_subprojects
+        )
+        user_ids = entity_cls.objects(query).distinct(field="user")
+        if not user_ids:
+            return []
+        users = User.objects(id__in=user_ids).only("id", "name")
+        return [{"id": u.id, "name": u.name} for u in users]
+
     @classmethod
     def get_task_types(cls, company, project_ids: Optional[Sequence]) -> set:
         """
         Return the list of unique task types used by company and public tasks
         If project ids passed then only tasks from these projects are considered
         """
-        query = get_company_or_none_constraint(company)
-        if project_ids:
-            project_ids = _ids_with_children(project_ids)
-            query &= Q(project__in=project_ids)
-        else:
-            query &= Q(system_tags__nin=[EntityVisibility.hidden.value])
+        query = cls._get_project_query(company, project_ids)
         res = Task.objects(query).distinct(field="type")
         return set(res).intersection(external_task_types)
 
@@ -851,10 +1076,7 @@ class ProjectBLL:
         Return the list of unique frameworks used by company and public models
         If project ids passed then only models from these projects are considered
         """
-        query = get_company_or_none_constraint(company)
-        if project_ids:
-            project_ids = _ids_with_children(project_ids)
-            query &= Q(project__in=project_ids)
+        query = cls._get_project_query(company, project_ids)
         return Model.objects(query).distinct(field="framework")
 
     @staticmethod
@@ -863,10 +1085,11 @@ class ProjectBLL:
         project_ids: Sequence[str],
         filter_: Mapping[str, Any],
         users: Sequence[str],
+        project_field: str = "project",
     ):
         conditions = {
             "company": {"$in": [None, "", company]},
-            "project": {"$in": project_ids},
+            project_field: {"$in": project_ids},
         }
         if users:
             conditions["user"] = {"$in": users}
@@ -874,29 +1097,129 @@ class ProjectBLL:
         if not filter_:
             return conditions
 
+        or_conditions = []
         for field, field_filter in filter_.items():
-            if not (
-                field_filter
-                and isinstance(field_filter, list)
-                and all(isinstance(t, str) for t in field_filter)
-            ):
+            if not (field_filter and isinstance(field_filter, (list, dict))):
                 raise errors.bad_request.ValidationError(
-                    f"List of strings expected for the field: {field}"
+                    f"Non empty list or dictionary expected for the field: {field}"
                 )
-            exclude, include = partition(field_filter, lambda x: x.startswith("-"))
-            conditions[field] = {
-                **({"$in": include} if include else {}),
-                **({"$nin": [e[1:] for e in exclude]} if exclude else {}),
-            }
+
+            if isinstance(field_filter, list):
+                if not all(isinstance(t, str) for t in field_filter):
+                    raise errors.bad_request.ValidationError(
+                        f"Only string values are allowed in the list filter: {field}"
+                    )
+                helper = GetMixin.NewListFieldBucketHelper(
+                    field, data=field_filter, legacy=True
+                )
+                op = (
+                    Q.OR
+                    if helper.explicit_operator and helper.global_operator == Q.OR
+                    else Q.AND
+                )
+                db_query = {op: helper.actions}
+            else:
+                helper = GetMixin.ListQueryFilter.from_data(field, field_filter)
+                db_query = helper.db_query
+
+            for op, actions in db_query.items():
+                field_conditions = {}
+                for action, values in actions.items():
+                    value = list(set(values))
+                    for key in reversed(action.split("__")):
+                        value = {f"${key}": value}
+                    field_conditions.update(value)
+
+                if op == Q.OR and len(field_conditions) > 1:
+                    or_conditions.append(
+                        {
+                            "$or": [
+                                {field: {db_modifier: cond}}
+                                for db_modifier, cond in field_conditions.items()
+                            ]
+                        }
+                    )
+                else:
+                    conditions[field] = field_conditions
+
+        if or_conditions:
+            if len(or_conditions) == 1:
+                conditions = next(iter(or_conditions))
+            else:
+                conditions["$and"] = [c for c in or_conditions]
 
         return conditions
 
+    @classmethod
+    def _calc_own_datasets_core(
+        cls,
+        company: str,
+        project_ids: Sequence[str],
+        project_field: str,
+        entity_class: Type[AttributedDocument],
+        filter_: Mapping[str, Any] = None,
+        users: Sequence[str] = None,
+    ) -> Dict[str, dict]:
+        """
+        Returns the amount of hyper datasets per requested project
+        """
+        if not project_ids:
+            return {}
+
+        pipeline = [
+            {
+                "$match": cls.get_match_conditions(
+                    company=company,
+                    project_ids=project_ids,
+                    filter_=filter_,
+                    users=users,
+                    project_field=project_field,
+                )
+            },
+            {"$project": {project_field: 1}},
+            {"$group": {"_id": f"${project_field}", "count": {"$sum": 1}}},
+        ]
+        datasets = {
+            data["_id"]: data["count"] for data in entity_class.aggregate(pipeline)
+        }
+
+        return {pid: {"own_datasets": datasets.get(pid, 0)} for pid in project_ids}
+
+    @classmethod
+    def calc_own_datasets(
+        cls,
+        company: str,
+        project_ids: Sequence[str],
+        filter_: Mapping[str, Any] = None,
+        users: Sequence[str] = None,
+    ) -> Dict[str, dict]:
+        """
+        Returns the amount of datasets per requested project
+        """
+        filter_ = filter_ or {}
+        filter_system_tags = filter_.get("system_tags")
+        if not isinstance(filter_system_tags, list):
+            filter_system_tags = []
+        if dataset_tag not in filter_system_tags:
+            filter_system_tags.append(dataset_tag)
+            filter_["system_tags"] = filter_system_tags
+
+        return cls._calc_own_datasets_core(
+            company=company,
+            project_ids=project_ids,
+            project_field="parent",
+            entity_class=Project,
+            filter_=filter_,
+            users=users,
+        )
+
     @classmethod
     def calc_own_contents(
         cls,
         company: str,
         project_ids: Sequence[str],
         filter_: Mapping[str, Any] = None,
+        specific_state: Optional[EntityVisibility] = None,
         users: Sequence[str] = None,
     ) -> Dict[str, dict]:
         """
@@ -907,6 +1230,20 @@ class ProjectBLL:
         if not project_ids:
             return {}
 
+        if specific_state:
+            filter_ = filter_ or {}
+            system_tags_filter = filter_.get("system_tags", [])
+            archived = EntityVisibility.archived.value
+            non_archived = f"-{EntityVisibility.archived.value}"
+            if not any(t in system_tags_filter for t in (archived, non_archived)):
+                filter_ = {k: v for k, v in filter_.items()}
+                filter_["system_tags"] = [
+                    archived
+                    if specific_state == EntityVisibility.archived
+                    else non_archived,
+                    *system_tags_filter,
+                ]
+
         pipeline = [
             {
                 "$match": cls.get_match_conditions(

apiserver/bll/project/project_cleanup.py

@@ -1,6 +1,9 @@
+from collections import defaultdict
+from datetime import datetime
 from typing import Tuple, Set, Sequence
 
 import attr
+from mongoengine import Q
 
 from apiserver.apierrors import errors
 from apiserver.bll.event import EventBLL
@@ -14,12 +17,19 @@ from apiserver.config_repo import config
 from apiserver.database.model import EntityVisibility
 from apiserver.database.model.model import Model
 from apiserver.database.model.project import Project
-from apiserver.database.model.task.task import Task, ArtifactModes, TaskType
+from apiserver.database.model.task.task import Task, ArtifactModes, TaskType, TaskStatus
+from .project_bll import (
+    ProjectBLL,
+    pipeline_tag,
+    pipelines_project_name,
+    dataset_tag,
+    datasets_project_name,
+    reports_tag,
+)
 from .sub_projects import _ids_with_children
 
 log = config.logger(__file__)
 event_bll = EventBLL()
-async_events_delete = config.get("services.tasks.async_events_delete", False)
 
 
 @attr.s(auto_attribs=True)
@@ -31,30 +41,83 @@ class DeleteProjectResult:
     urls: TaskUrls = None
 
 
+def _get_child_project_ids(
+    project_id: str,
+) -> Tuple[Sequence[str], Sequence[str], Sequence[str]]:
+    project_ids = _ids_with_children([project_id])
+    pipeline_ids = list(
+        Project.objects(
+            id__in=project_ids,
+            system_tags__in=[pipeline_tag],
+            basename__ne=pipelines_project_name,
+        ).scalar("id")
+    )
+    dataset_ids = list(
+        Project.objects(
+            id__in=project_ids,
+            system_tags__in=[dataset_tag],
+            basename__ne=datasets_project_name,
+        ).scalar("id")
+    )
+    return project_ids, pipeline_ids, dataset_ids
+
+
 def validate_project_delete(company: str, project_id: str):
     project = Project.get_for_writing(
         company=company, id=project_id, _only=("id", "path", "system_tags")
     )
     if not project:
         raise errors.bad_request.InvalidProjectId(id=project_id)
-    is_pipeline = "pipeline" in (project.system_tags or [])
-    project_ids = _ids_with_children([project_id])
+
+    project_ids, pipeline_ids, dataset_ids = _get_child_project_ids(project_id)
     ret = {}
-    for cls in (Task, Model):
-        ret[f"{cls.__name__.lower()}s"] = cls.objects(project__in=project_ids).count()
-    for cls in (Task, Model):
-        query = dict(
-            project__in=project_ids, system_tags__nin=[EntityVisibility.archived.value]
-        )
-        name = f"non_archived_{cls.__name__.lower()}s"
-        if not is_pipeline:
-            ret[name] = cls.objects(**query).count()
-        else:
-            ret[name] = (
-                cls.objects(**query, type=TaskType.controller).count()
-                if cls == Task
-                else 0
+    if pipeline_ids:
+        pipelines_with_active_controllers = Task.objects(
+            project__in=pipeline_ids,
+            type=TaskType.controller,
+            system_tags__nin=[EntityVisibility.archived.value],
+        ).distinct("project")
+        ret["pipelines"] = len(pipelines_with_active_controllers)
+    else:
+        ret["pipelines"] = 0
+    if dataset_ids:
+        datasets_with_data = Task.objects(
+            project__in=dataset_ids,
+            system_tags__nin=[EntityVisibility.archived.value],
+        ).distinct("project")
+        ret["datasets"] = len(datasets_with_data)
+    else:
+        ret["datasets"] = 0
+
+    project_ids = list(set(project_ids) - set(pipeline_ids) - set(dataset_ids))
+    if project_ids:
+        in_project_query = Q(project__in=project_ids)
+        for cls in (Task, Model):
+            query = (
+                in_project_query & Q(system_tags__nin=[reports_tag])
+                if cls is Task
+                else in_project_query
             )
+            ret[f"{cls.__name__.lower()}s"] = cls.objects(query).count()
+            ret[f"non_archived_{cls.__name__.lower()}s"] = cls.objects(
+                query & Q(system_tags__nin=[EntityVisibility.archived.value])
+            ).count()
+        ret["reports"] = Task.objects(
+            in_project_query & Q(system_tags__in=[reports_tag])
+        ).count()
+        ret["non_archived_reports"] = Task.objects(
+            in_project_query
+            & Q(
+                system_tags__in=[reports_tag],
+                system_tags__nin=[EntityVisibility.archived.value],
+            )
+        ).count()
+    else:
+        for cls in (Task, Model):
+            ret[f"{cls.__name__.lower()}s"] = 0
+            ret[f"non_archived_{cls.__name__.lower()}s"] = 0
+        ret["reports"] = 0
+        ret["non_archived_reports"] = 0
 
     return ret
 
@@ -65,7 +128,7 @@ def delete_project(
     project_id: str,
     force: bool,
     delete_contents: bool,
-    delete_external_artifacts=True,
+    delete_external_artifacts: bool,
 ) -> Tuple[DeleteProjectResult, Set[str]]:
     project = Project.get_for_writing(
         company=company, id=project_id, _only=("id", "path", "system_tags")
@@ -74,39 +137,58 @@ def delete_project(
         raise errors.bad_request.InvalidProjectId(id=project_id)
 
     delete_external_artifacts = delete_external_artifacts and config.get(
-        "services.async_urls_delete.enabled", False
+        "services.async_urls_delete.enabled", True
     )
-    is_pipeline = "pipeline" in (project.system_tags or [])
-    project_ids = _ids_with_children([project_id])
+    project_ids, pipeline_ids, dataset_ids = _get_child_project_ids(project_id)
     if not force:
-        query = dict(
-            project__in=project_ids, system_tags__nin=[EntityVisibility.archived.value]
-        )
-        if not is_pipeline:
+        if pipeline_ids:
+            active_controllers = Task.objects(
+                project__in=pipeline_ids,
+                type=TaskType.controller,
+                system_tags__nin=[EntityVisibility.archived.value],
+            ).only("id")
+            if active_controllers:
+                raise errors.bad_request.ProjectHasPipelines(
+                    "please archive all the controllers or use force=true",
+                    id=project_id,
+                )
+        if dataset_ids:
+            datasets_with_data = Task.objects(
+                project__in=dataset_ids,
+                system_tags__nin=[EntityVisibility.archived.value],
+            ).only("id")
+            if datasets_with_data:
+                raise errors.bad_request.ProjectHasDatasets(
+                    "please delete all the dataset versions or use force=true",
+                    id=project_id,
+                )
+
+        regular_projects = list(set(project_ids) - set(pipeline_ids) - set(dataset_ids))
+        if regular_projects:
             for cls, error in (
                 (Task, errors.bad_request.ProjectHasTasks),
                 (Model, errors.bad_request.ProjectHasModels),
             ):
-                non_archived = cls.objects(**query).only("id")
+                non_archived = cls.objects(
+                    project__in=regular_projects,
+                    system_tags__nin=[EntityVisibility.archived.value],
+                ).only("id")
                 if non_archived:
-                    raise error("use force=true to delete", id=project_id)
-        else:
-            non_archived = Task.objects(**query, type=TaskType.controller).only("id")
-            if non_archived:
-                raise errors.bad_request.ProjectHasTasks(
-                    "please archive all the runs inside the project", id=project_id
-                )
+                    raise error("use force=true", id=project_id)
 
     if not delete_contents:
-        for cls in (Model, Task):
-            updated_count = cls.objects(project__in=project_ids).update(project=None)
-        res = DeleteProjectResult(disassociated_tasks=updated_count)
+        disassociated = defaultdict(int)
+        for cls in ProjectBLL.child_classes:
+            disassociated[cls] = cls.objects(project__in=project_ids).update(
+                project=None
+            )
+        res = DeleteProjectResult(disassociated_tasks=disassociated[Task])
     else:
         deleted_models, model_event_urls, model_urls = _delete_models(
-            company=company, projects=project_ids
+            company=company, user=user, projects=project_ids
         )
         deleted_tasks, task_event_urls, artifact_urls = _delete_tasks(
-            company=company, projects=project_ids
+            company=company, user=user, projects=project_ids
         )
         event_urls = task_event_urls | model_event_urls
         if delete_external_artifacts:
@@ -135,7 +217,9 @@ def delete_project(
     return res, affected
 
 
-def _delete_tasks(company: str, projects: Sequence[str]) -> Tuple[int, Set, Set]:
+def _delete_tasks(
+    company: str, user: str, projects: Sequence[str]
+) -> Tuple[int, Set, Set]:
     """
     Delete only the task themselves and their non published version.
     Child models under the same project are deleted separately.
@@ -146,14 +230,24 @@ def _delete_tasks(company: str, projects: Sequence[str]) -> Tuple[int, Set, Set]
     if not tasks:
         return 0, set(), set()
 
-    task_ids = {t.id for t in tasks}
-    Task.objects(parent__in=task_ids, project__nin=projects).update(parent=None)
-    Model.objects(task__in=task_ids, project__nin=projects).update(task=None)
+    task_ids = list({t.id for t in tasks})
+    now = datetime.utcnow()
+    Task.objects(parent__in=task_ids, project__nin=projects).update(
+        parent=None,
+        last_change=now,
+        last_changed_by=user,
+    )
+    Model.objects(task__in=task_ids, project__nin=projects).update(
+        task=None,
+        last_change=now,
+        last_changed_by=user,
+    )
 
-    event_urls, artifact_urls = set(), set()
+    event_urls = collect_debug_image_urls(company, task_ids) | collect_plot_image_urls(
+        company, task_ids
+    )
+    artifact_urls = set()
     for task in tasks:
-        event_urls.update(collect_debug_image_urls(company, task.id))
-        event_urls.update(collect_plot_image_urls(company, task.id))
         if task.execution and task.execution.artifacts:
             artifact_urls.update(
                 {
@@ -163,15 +257,13 @@ def _delete_tasks(company: str, projects: Sequence[str]) -> Tuple[int, Set, Set]
                 }
             )
 
-    event_bll.delete_multi_task_events(
-        company, list(task_ids), async_delete=async_events_delete
-    )
+    event_bll.delete_multi_task_events(company, task_ids)
     deleted = tasks.delete()
     return deleted, event_urls, artifact_urls
 
 
 def _delete_models(
-    company: str, projects: Sequence[str]
+    company: str, user: str, projects: Sequence[str]
 ) -> Tuple[int, Set[str], Set[str]]:
     """
     Delete project models and update the tasks from other projects
@@ -182,39 +274,54 @@ def _delete_models(
         return 0, set(), set()
 
     model_ids = list({m.id for m in models})
-
+    deleted = "__DELETED__"
     Task._get_collection().update_many(
         filter={
             "project": {"$nin": projects},
             "models.input.model": {"$in": model_ids},
         },
-        update={"$set": {"models.input.$[elem].model": None}},
+        update={"$set": {"models.input.$[elem].model": deleted}},
         array_filters=[{"elem.model": {"$in": model_ids}}],
         upsert=False,
     )
 
     model_tasks = list({m.task for m in models if m.task})
     if model_tasks:
+        now = datetime.utcnow()
+        # update published tasks
         Task._get_collection().update_many(
             filter={
                 "_id": {"$in": model_tasks},
                 "project": {"$nin": projects},
                 "models.output.model": {"$in": model_ids},
+                "status": TaskStatus.published,
+            },
+            update={
+                "$set": {
+                    "models.output.$[elem].model": deleted,
+                    "last_change": now,
+                    "last_changed_by": user,
+                }
             },
-            update={"$set": {"models.output.$[elem].model": None}},
             array_filters=[{"elem.model": {"$in": model_ids}}],
             upsert=False,
         )
+        # update unpublished tasks
+        Task.objects(
+            id__in=model_tasks,
+            project__nin=projects,
+            status__ne=TaskStatus.published,
+        ).update(
+            pull__models__output__model__in=model_ids,
+            set__last_change=now,
+            set__last_changed_by=user,
+        )
 
-    event_urls, model_urls = set(), set()
-    for m in models:
-        event_urls.update(collect_debug_image_urls(company, m.id))
-        event_urls.update(collect_plot_image_urls(company, m.id))
-        if m.uri:
-            model_urls.add(m.uri)
-
-    event_bll.delete_multi_task_events(
-        company, model_ids, async_delete=async_events_delete
+    event_urls = collect_debug_image_urls(company, model_ids) | collect_plot_image_urls(
+        company, model_ids
     )
+    model_urls = {m.uri for m in models if m.uri}
+
+    event_bll.delete_multi_task_events(company, model_ids, model=True)
     deleted = models.delete()
     return deleted, event_urls, model_urls

apiserver/bll/project/project_queries.py

@@ -140,7 +140,12 @@ class ProjectQueries:
         name: str,
         include_subprojects: bool,
         allow_public: bool = True,
+        pattern: str = None,
+        page: int = 0,
+        page_size: int = 500,
     ) -> ParamValues:
+        page = max(0, page)
+        page_size = max(1, page_size)
         company_constraint = self._get_company_constraint(company_id, allow_public)
         project_constraint = self._get_project_constraint(
             project_ids, include_subprojects
@@ -160,7 +165,20 @@ class ProjectQueries:
         if not last_updated_task:
             return 0, []
 
-        redis_key = f"hyperparam_values_{company_id}_{'_'.join(project_ids)}_{section}_{name}_{allow_public}"
+        redis_key = "_".join(
+            str(part)
+            for part in (
+                "hyperparam_values",
+                company_id,
+                "_".join(project_ids),
+                section,
+                name,
+                allow_public,
+                pattern,
+                page,
+                page_size,
+            )
+        )
         last_update = last_updated_task.last_update or datetime.utcnow()
         cached_res = self._get_cached_param_values(
             key=redis_key,
@@ -172,19 +190,27 @@ class ProjectQueries:
         if cached_res:
             return cached_res
 
-        max_values = config.get("services.tasks.hyperparam_values.max_count", 100)
-        pipeline = [
-            {
-                "$match": {
-                    **company_constraint,
-                    **project_constraint,
-                    key_path: {"$exists": True},
+        match_condition = {
+            **company_constraint,
+            **project_constraint,
+            key_path: {"$exists": True},
+        }
+        if pattern:
+            match_condition["$expr"] = {
+                "$regexMatch": {
+                    "input": f"${key_path}.value",
+                    "regex": pattern,
+                    "options": "i",
                 }
-            },
+            }
+
+        pipeline = [
+            {"$match": match_condition},
             {"$project": {"value": f"${key_path}.value"}},
             {"$group": {"_id": "$value"}},
             {"$sort": {"_id": 1}},
-            {"$limit": max_values},
+            {"$skip": page * page_size},
+            {"$limit": page_size},
             {
                 "$group": {
                     "_id": 1,
@@ -209,7 +235,11 @@ class ProjectQueries:
 
     @classmethod
     def get_unique_metric_variants(
-        cls, company_id, project_ids: Sequence[str], include_subprojects: bool
+        cls,
+        company_id,
+        project_ids: Sequence[str],
+        include_subprojects: bool,
+        model_metrics: bool = False,
     ):
         pipeline = [
             {
@@ -246,7 +276,8 @@ class ProjectQueries:
             {"$sort": OrderedDict({"_id.metric": 1, "_id.variant": 1})},
         ]
 
-        result = Task.aggregate(pipeline)
+        entity_cls = Model if model_metrics else Task
+        result = entity_cls.aggregate(pipeline)
         return [r["metrics"][0] for r in result]
 
     @classmethod
@@ -306,7 +337,11 @@ class ProjectQueries:
         key: str,
         include_subprojects: bool,
         allow_public: bool = True,
+        page: int = 0,
+        page_size: int = 500,
     ) -> ParamValues:
+        page = max(0, page)
+        page_size = max(1, page_size)
         company_constraint = self._get_company_constraint(company_id, allow_public)
         project_constraint = self._get_project_constraint(
             project_ids, include_subprojects
@@ -326,7 +361,7 @@ class ProjectQueries:
         if not last_updated_model:
             return 0, []
 
-        redis_key = f"modelmetadata_values_{company_id}_{'_'.join(project_ids)}_{key}_{allow_public}"
+        redis_key = f"modelmetadata_values_{company_id}_{'_'.join(project_ids)}_{key}_{allow_public}_{page}_{page_size}"
         last_update = last_updated_model.last_update or datetime.utcnow()
         cached_res = self._get_cached_param_values(
             key=redis_key, last_update=last_update
@@ -334,7 +369,6 @@ class ProjectQueries:
         if cached_res:
             return cached_res
 
-        max_values = config.get("services.models.metadata_values.max_count", 100)
         pipeline = [
             {
                 "$match": {
@@ -346,7 +380,8 @@ class ProjectQueries:
             {"$project": {"value": f"${key_path}.value"}},
             {"$group": {"_id": "$value"}},
             {"$sort": {"_id": 1}},
-            {"$limit": max_values},
+            {"$skip": page * page_size},
+            {"$limit": page_size},
             {
                 "$group": {
                     "_id": 1,

apiserver/bll/project/sub_projects.py

@@ -14,14 +14,16 @@ def _get_project_depth(project_name: str) -> int:
     return len(list(filter(None, project_name.split(name_separator))))
 
 
-def _validate_project_name(project_name: str) -> Tuple[str, str]:
+def _validate_project_name(project_name: str, raise_if_empty=True) -> Tuple[str, str]:
     """
     Remove redundant '/' characters. Ensure that the project name is not empty
     Return the cleaned up project name and location
     """
-    name_parts = list(filter(None, project_name.split(name_separator)))
+    name_parts = [p.strip() for p in project_name.split(name_separator) if p]
     if not name_parts:
-        raise errors.bad_request.InvalidProjectName(name=project_name)
+        if raise_if_empty:
+            raise errors.bad_request.InvalidProjectName(name=project_name)
+        return "", ""
 
     return name_separator.join(name_parts), name_separator.join(name_parts[:-1])
 
@@ -34,7 +36,7 @@ def _ensure_project(
     If needed auto-create the project and all the missing projects in the path to it
     Return the project
     """
-    name = name.strip(name_separator)
+    name, location = _validate_project_name(name, raise_if_empty=False)
     if not name:
         return None
 
@@ -43,7 +45,6 @@ def _ensure_project(
         return project
 
     now = datetime.utcnow()
-    name, location = _validate_project_name(name)
     project = Project(
         id=database.utils.id(),
         user=user,
@@ -101,12 +102,15 @@ def _get_writable_project_from_name(
     return qs.first()
 
 
+ProjectsChildren = Mapping[str, Sequence[Project]]
+
+
 def _get_sub_projects(
     project_ids: Sequence[str],
     _only: Sequence[str] = ("id", "path"),
     search_hidden=True,
     allowed_ids: Sequence[str] = None,
-) -> Mapping[str, Sequence[Project]]:
+) -> ProjectsChildren:
     """
     Return the list of child projects of all the levels for the parent project ids
     """
@@ -140,8 +144,8 @@ def _ids_with_children(project_ids: Sequence[str]) -> Sequence[str]:
     """
     Return project ids with the ids of all the subprojects
     """
-    subprojects = Project.objects(path__in=project_ids).only("id")
-    return list({*project_ids, *(child.id for child in subprojects)})
+    children_ids = Project.objects(path__in=project_ids).scalar("id")
+    return list({*project_ids, *children_ids})
 
 
 def _update_subproject_names(
@@ -156,13 +160,17 @@ def _update_subproject_names(
     Optionally update the paths
     """
     updated = 0
+    now = datetime.utcnow()
     for child in children:
         child_suffix = name_separator.join(
-            child.name.split(name_separator)[len(old_name.split(name_separator)) :]
+            child.name.split(name_separator)[len(old_name.split(name_separator)):]
         )
-        updates = {"name": name_separator.join((project.name, child_suffix))}
+        updates = {
+            "name": name_separator.join((project.name, child_suffix)),
+            "last_update": now,
+        }
         if update_path:
-            updates["path"] = project.path + child.path[len(old_path) :]
+            updates["path"] = project.path + child.path[len(old_path):]
         updated += child.update(upsert=False, **updates)
 
     return updated
@@ -177,6 +185,7 @@ def _reposition_project_with_children(
     project.name = name_separator.join(
         filter(None, (new_location, project.name.split(name_separator)[-1]))
     )
+    project.last_update = datetime.utcnow()
     _save_under_parent(project, parent=parent)
 
     moved = 1 + _update_subproject_names(

apiserver/bll/queue/queue_bll.py

@@ -1,6 +1,6 @@
 from collections import defaultdict
 from datetime import datetime
-from typing import Callable, Sequence, Optional, Tuple
+from typing import Sequence, Optional, Tuple, Union
 
 from elasticsearch import Elasticsearch
 from mongoengine import Q
@@ -16,6 +16,8 @@ from apiserver.database.errors import translate_errors_context
 from apiserver.database.model.queue import Queue, Entry
 
 log = config.logger(__file__)
+MOVE_FIRST = "first"
+MOVE_LAST = "last"
 
 
 class QueueBLL(object):
@@ -133,7 +135,7 @@ class QueueBLL(object):
             self.get_by_id(company_id=company_id, queue_id=queue_id, only=("id",))
             return Queue.safe_update(company_id, queue_id, update_fields)
 
-    def delete(self, company_id: str, queue_id: str, force: bool) -> None:
+    def delete(self, company_id: str, user_id: str, queue_id: str, force: bool) -> None:
         """
         Delete the queue
         :raise errors.bad_request.InvalidQueueId: if the queue is not found
@@ -153,7 +155,13 @@ class QueueBLL(object):
                         task = Task.get_for_writing(
                             company=company_id,
                             id=item.task,
-                            _only=["id", "status", "enqueue_status", "project"],
+                            _only=[
+                                "id",
+                                "company",
+                                "status",
+                                "enqueue_status",
+                                "project",
+                            ],
                         )
                         if not task:
                             continue
@@ -163,6 +171,8 @@ class QueueBLL(object):
                             new_status=task.enqueue_status or TaskStatus.created,
                             status_reason="Queue was deleted",
                             status_message="",
+                            user_id=user_id,
+                            force=True,
                         ).execute(enqueue_status=None)
                     except Exception as ex:
                         log.exception(
@@ -233,6 +243,7 @@ class QueueBLL(object):
                     {
                         "name": w.id,
                         "ip": w.ip,
+                        "key": w.key,
                         "task": w.task.to_struct() if w.task else None,
                     }
                     for w in queue_workers.get(item["id"], [])
@@ -318,46 +329,131 @@ class QueueBLL(object):
             return len(entries_to_remove) if res else 0
 
     def reposition_task(
-        self,
-        company_id: str,
-        queue_id: str,
-        task_id: str,
-        pos_func: Callable[[int], int],
+        self, company_id: str, queue_id: str, task_id: str, move_count: Union[int, str],
     ) -> int:
         """
         Moves the task in the queue to the position calculated by pos_func
         Returns the updated task position in the queue
         """
-        with translate_errors_context():
-            queue = self.get_queue_with_task(
+
+        def get_queue_and_task_position():
+            q = self.get_queue_with_task(
                 company_id=company_id, queue_id=queue_id, task_id=task_id
             )
+            return q, next(i for i, e in enumerate(q.entries) if e.task == task_id)
 
-            position = next(i for i, e in enumerate(queue.entries) if e.task == task_id)
-            new_position = pos_func(position)
+        with translate_errors_context():
+            queue, position = get_queue_and_task_position()
+            if move_count == MOVE_FIRST:
+                new_position = 0
+            elif move_count == MOVE_LAST:
+                new_position = len(queue.entries) - 1
+            else:
+                new_position = position + move_count
+            if new_position == position:
+                return new_position
 
-            if new_position != position:
-                entry = queue.entries[position]
-                query = dict(id=queue_id, company=company_id)
-                updated = Queue.objects(entries__task=task_id, **query).update_one(
-                    pull__entries=entry, last_update=datetime.utcnow()
-                )
-                if not updated:
-                    raise errors.bad_request.RemovedDuringReposition(
-                        task=task_id, **query
-                    )
-                inst = {"$push": {"entries": {"$each": [entry.to_proper_dict()]}}}
-                if new_position >= 0:
-                    inst["$push"]["entries"]["$position"] = new_position
-                res = Queue.objects(entries__task__ne=task_id, **query).update_one(
-                    __raw__=inst
-                )
-                if not res:
-                    raise errors.bad_request.FailedAddingDuringReposition(
-                        task=task_id, **query
-                    )
+            without_entry = {
+                "$filter": {
+                    "input": "$entries",
+                    "as": "entry",
+                    "cond": {"$ne": ["$$entry.task", task_id]},
+                }
+            }
+            task_entry = {
+                "$filter": {
+                    "input": "$entries",
+                    "as": "entry",
+                    "cond": {"$eq": ["$$entry.task", task_id]},
+                }
+            }
+            if move_count == MOVE_FIRST:
+                operations = [
+                    {
+                        "$set": {
+                            "entries": {"$concatArrays": [task_entry, without_entry]}
+                        }
+                    }
+                ]
+            elif move_count == MOVE_LAST:
+                operations = [
+                    {
+                        "$set": {
+                            "entries": {"$concatArrays": [without_entry, task_entry]}
+                        }
+                    }
+                ]
+            else:
+                operations = [
+                    {
+                        "$set": {
+                            "new_pos": {
+                                "$add": [
+                                    {"$indexOfArray": ["$entries.task", task_id]},
+                                    move_count,
+                                ]
+                            },
+                            "without_entry": without_entry,
+                            "task_entry": task_entry,
+                        }
+                    },
+                    {
+                        "$set": {
+                            "entries": {
+                                "$switch": {
+                                    "branches": [
+                                        {
+                                            "case": {"$lte": ["$new_pos", 0]},
+                                            "then": {
+                                                "$concatArrays": [
+                                                    "$task_entry",
+                                                    "$without_entry",
+                                                ]
+                                            },
+                                        },
+                                        {
+                                            "case": {
+                                                "$gte": [
+                                                    "$new_pos",
+                                                    {"$size": "$without_entry"},
+                                                ]
+                                            },
+                                            "then": {
+                                                "$concatArrays": [
+                                                    "$without_entry",
+                                                    "$task_entry",
+                                                ]
+                                            },
+                                        },
+                                    ],
+                                    "default": {
+                                        "$concatArrays": [
+                                            {"$slice": ["$without_entry", "$new_pos"]},
+                                            "$task_entry",
+                                            {
+                                                "$slice": [
+                                                    "$without_entry",
+                                                    "$new_pos",
+                                                    {"$size": "$without_entry"},
+                                                ]
+                                            },
+                                        ]
+                                    },
+                                }
+                            }
+                        }
+                    },
+                    {"$unset": ["new_pos", "without_entry", "task_entry"]},
+                ]
 
-            return new_position
+            updated = Queue.objects(
+                id=queue_id, company=company_id, entries__task=task_id
+            ).update_one(__raw__=operations)
+
+            if not updated:
+                raise errors.bad_request.FailedAddingDuringReposition(task=task_id)
+
+            return get_queue_and_task_position()[1]
 
     def count_entries(self, company: str, queue_id: str) -> Optional[int]:
         res = next(

apiserver/bll/queue/queue_metrics.py

@@ -80,7 +80,7 @@ class QueueMetrics:
         logged = 0
         for q in queues:
             queue_doc = make_doc(q)
-            self.es.index(index=es_index, body=queue_doc)
+            self.es.index(index=es_index, document=queue_doc)
             redis_key = _queue_metrics_key_pattern.format(queue=q.id)
             redis.set(redis_key, json.dumps(queue_doc))
             logged += 1

apiserver/bll/statistics/stats_reporter.py

@@ -8,8 +8,7 @@ from typing import Sequence, Optional
 
 import dpath
 import requests
-from requests.adapters import HTTPAdapter
-from requests.packages.urllib3.util.retry import Retry
+from requests.adapters import HTTPAdapter, Retry
 
 from apiserver.bll.query import Builder as QueryBuilder
 from apiserver.bll.util import get_server_uuid

apiserver/bll/storage/__init__.py

@@ -0,0 +1,48 @@
+from copy import copy
+
+from boltons.cacheutils import cachedproperty
+from clearml.backend_config.bucket_config import (
+    S3BucketConfigurations,
+    AzureContainerConfigurations,
+    GSBucketConfigurations,
+)
+
+from apiserver.config_repo import config
+
+
+log = config.logger(__file__)
+
+
+class StorageBLL:
+    default_aws_configs: S3BucketConfigurations = None
+    conf = config.get("services.storage_credentials")
+
+    @cachedproperty
+    def _default_aws_configs(self) -> S3BucketConfigurations:
+        return S3BucketConfigurations.from_config(self.conf.get("aws.s3"))
+
+    @cachedproperty
+    def _default_azure_configs(self) -> AzureContainerConfigurations:
+        return AzureContainerConfigurations.from_config(self.conf.get("azure.storage"))
+
+    @cachedproperty
+    def _default_gs_configs(self) -> GSBucketConfigurations:
+        return GSBucketConfigurations.from_config(self.conf.get("google.storage"))
+
+    def get_azure_settings_for_company(
+        self,
+        company_id: str,
+    ) -> AzureContainerConfigurations:
+        return copy(self._default_azure_configs)
+
+    def get_gs_settings_for_company(
+        self,
+        company_id: str,
+    ) -> GSBucketConfigurations:
+        return copy(self._default_gs_configs)
+
+    def get_aws_settings_for_company(
+        self,
+        company_id: str,
+    ) -> S3BucketConfigurations:
+        return copy(self._default_aws_configs)

apiserver/bll/task/__init__.py

@@ -1,6 +1,5 @@
 from .task_bll import TaskBLL
 from .utils import (
     ChangeStatusRequest,
-    update_project_time,
     validate_status_change,
 )

apiserver/bll/task/artifacts.py

@@ -48,6 +48,7 @@ class Artifacts:
     def add_or_update_artifacts(
         cls,
         company_id: str,
+        user_id: str,
         task_id: str,
         artifacts: Sequence[ApiArtifact],
         force: bool,
@@ -63,12 +64,13 @@ class Artifacts:
             f"set__execution__artifacts__{mongoengine_safe(name)}": value
             for name, value in artifacts.items()
         }
-        return update_task(task, update_cmds=update_cmds)
+        return update_task(task, user_id=user_id, update_cmds=update_cmds)
 
     @classmethod
     def delete_artifacts(
         cls,
         company_id: str,
+        user_id: str,
         task_id: str,
         artifact_ids: Sequence[ArtifactId],
         force: bool,
@@ -83,4 +85,4 @@ class Artifacts:
             f"unset__execution__artifacts__{id_}": 1 for id_ in set(artifact_ids)
         }
 
-        return update_task(task, update_cmds=delete_cmds)
+        return update_task(task, user_id=user_id, update_cmds=delete_cmds)

apiserver/bll/task/hyperparams.py

@@ -63,6 +63,7 @@ class HyperParams:
     def delete_params(
         cls,
         company_id: str,
+        user_id: str,
         task_id: str,
         hyperparams: Sequence[HyperParamKey],
         force: bool,
@@ -94,13 +95,17 @@ class HyperParams:
             delete_cmds[f"unset__hyperparams__{section}__{name}"] = 1
 
         return update_task(
-            task, update_cmds=delete_cmds, set_last_update=not properties_only
+            task,
+            user_id=user_id,
+            update_cmds=delete_cmds,
+            set_last_update=not properties_only,
         )
 
     @classmethod
     def edit_params(
         cls,
         company_id: str,
+        user_id: str,
         task_id: str,
         hyperparams: Sequence[HyperParamItem],
         replace_hyperparams: str,
@@ -129,7 +134,10 @@ class HyperParams:
                     ] = value
 
         return update_task(
-            task, update_cmds=update_cmds, set_last_update=not properties_only
+            task,
+            user_id=user_id,
+            update_cmds=update_cmds,
+            set_last_update=not properties_only,
         )
 
     @classmethod
@@ -201,6 +209,7 @@ class HyperParams:
     def edit_configuration(
         cls,
         company_id: str,
+        user_id: str,
         task_id: str,
         configuration: Sequence[Configuration],
         replace_configuration: bool,
@@ -219,11 +228,16 @@ class HyperParams:
             for name, value in configuration.items():
                 update_cmds[f"set__configuration__{mongoengine_safe(name)}"] = value
 
-        return update_task(task, update_cmds=update_cmds)
+        return update_task(task, user_id=user_id, update_cmds=update_cmds)
 
     @classmethod
     def delete_configuration(
-        cls, company_id: str, task_id: str, configuration: Sequence[str], force: bool
+        cls,
+        company_id: str,
+        user_id: str,
+        task_id: str,
+        configuration: Sequence[str],
+        force: bool,
     ) -> int:
         task = get_task_for_update(company_id=company_id, task_id=task_id, force=force)
 
@@ -232,4 +246,4 @@ class HyperParams:
             for name in set(configuration)
         }
 
-        return update_task(task, update_cmds=delete_cmds)
+        return update_task(task, user_id=user_id, update_cmds=delete_cmds)

apiserver/bll/task/non_responsive_tasks_watchdog.py

@@ -1,7 +1,7 @@
 from datetime import timedelta, datetime
 from time import sleep
 
-from apiserver.bll.task import update_project_time
+from apiserver.bll.util import update_project_time
 from apiserver.config_repo import config
 from apiserver.database.model.task.task import TaskStatus, Task
 from apiserver.utilities.threads_manager import ThreadsManager
@@ -85,6 +85,7 @@ class NonResponsiveTasksWatchdog:
                     status_changed=now,
                     last_update=now,
                     last_change=now,
+                    last_changed_by="__apiserver__",
                 )
                 if updated:
                     project_ids.add(task.project)

apiserver/bll/task/task_bll.py

@@ -7,11 +7,12 @@ from redis import StrictRedis
 from six import string_types
 
 import apiserver.database.utils as dbutils
-from apiserver.apierrors import errors
+from apiserver.apierrors import errors, APIError
 from apiserver.apimodels.tasks import TaskInputModel
 from apiserver.bll.queue import QueueBLL
 from apiserver.bll.organization import OrgBLL, Tags
 from apiserver.bll.project import ProjectBLL
+from apiserver.bll.util import update_project_time
 from apiserver.config_repo import config
 from apiserver.database.errors import translate_errors_context
 from apiserver.database.model.model import Model
@@ -30,17 +31,20 @@ from apiserver.database.model.task.task import (
     TaskModelTypes,
 )
 from apiserver.database.model import EntityVisibility
-from apiserver.database.utils import get_company_or_none_constraint, id as create_id
+from apiserver.database.model.queue import Queue
+from apiserver.database.utils import (
+    get_company_or_none_constraint,
+    id as create_id,
+)
 from apiserver.es_factory import es_factory
 from apiserver.redis_manager import redman
-from apiserver.service_repo import APICall
 from apiserver.services.utils import validate_tags, escape_dict_field, escape_dict
 from .artifacts import artifacts_prepare_for_save
 from .param_utils import params_prepare_for_save
 from .utils import (
     ChangeStatusRequest,
-    update_project_time,
     deleted_prefix,
+    get_last_metric_updates,
 )
 
 log = config.logger(__file__)
@@ -77,7 +81,11 @@ class TaskBLL:
 
     @staticmethod
     def get_by_id(
-        company_id, task_id, required_status=None, only_fields=None, allow_public=False,
+        company_id,
+        task_id,
+        required_status=None,
+        only_fields=None,
+        allow_public=False,
     ):
         if only_fields:
             if isinstance(only_fields, string_types):
@@ -128,16 +136,16 @@ class TaskBLL:
                 return list(q)
 
     @staticmethod
-    def create(call: APICall, fields: dict):
-        identity = call.identity
+    def create(company: str, user: str, fields: dict):
         now = datetime.utcnow()
         return Task(
             id=create_id(),
-            user=identity.user,
-            company=identity.company,
+            user=user,
+            company=company,
             created=now,
             last_update=now,
             last_change=now,
+            last_changed_by=user,
             **fields,
         )
 
@@ -257,6 +265,16 @@ class TaskBLL:
                 not in [TaskSystemTags.development, EntityVisibility.archived.value]
             ]
 
+        def ensure_int_labels(execution: dict) -> dict:
+            if not execution:
+                return execution
+
+            model_labels = execution.get("model_labels")
+            if model_labels:
+                execution["model_labels"] = {k: int(v) for k, v in model_labels.items()}
+
+            return execution
+
         parent_task = (
             task.parent
             if task.parent and not task.parent.startswith(deleted_prefix)
@@ -269,6 +287,7 @@ class TaskBLL:
             created=now,
             last_update=now,
             last_change=now,
+            last_changed_by=user_id,
             name=name or task.name,
             comment=comment or task.comment,
             parent=parent or parent_task,
@@ -280,7 +299,7 @@ class TaskBLL:
             output=Output(destination=task.output.destination) if task.output else None,
             models=Models(input=input_models or task.models.input),
             container=escape_dict(container) or task.container,
-            execution=execution_dict,
+            execution=ensure_int_labels(execution_dict),
             configuration=params_dict.get("configuration") or task.configuration,
             hyperparams=params_dict.get("hyperparams") or task.hyperparams,
         )
@@ -301,7 +320,7 @@ class TaskBLL:
         org_bll.update_tags(
             company_id,
             Tags.Task,
-            project=new_task.project,
+            projects=[new_task.project],
             tags=updated_tags,
             system_tags=updated_system_tags,
         )
@@ -344,6 +363,7 @@ class TaskBLL:
     def set_last_update(
         task_ids: Collection[str],
         company_id: str,
+        user_id: str,
         last_update: datetime,
         **extra_updates,
     ):
@@ -364,6 +384,7 @@ class TaskBLL:
                 upsert=False,
                 last_update=last_update,
                 last_change=last_update,
+                last_changed_by=user_id,
                 **updates,
             )
         return count
@@ -372,6 +393,7 @@ class TaskBLL:
     def update_statistics(
         task_id: str,
         company_id: str,
+        user_id: str,
         last_update: datetime = None,
         last_iteration: int = None,
         last_iteration_max: int = None,
@@ -400,50 +422,21 @@ class TaskBLL:
         elif last_iteration_max is not None:
             extra_updates.update(max__last_iteration=last_iteration_max)
 
+        raw_updates = {}
         if last_scalar_events is not None:
-            max_values = config.get("services.tasks.max_last_metrics", 2000)
-            total_metrics = set()
-            if max_values:
-                query = dict(id=task_id)
-                to_add = sum(len(v) for m, v in last_scalar_events.items())
-                if to_add <= max_values:
-                    query[f"unique_metrics__{max_values-to_add}__exists"] = True
-                task = Task.objects(**query).only("unique_metrics").first()
-                if task and task.unique_metrics:
-                    total_metrics = set(task.unique_metrics)
-
-            new_metrics = []
-            for metric_key, metric_data in last_scalar_events.items():
-                for variant_key, variant_data in metric_data.items():
-                    metric = (
-                        f"{variant_data.get('metric')}/{variant_data.get('variant')}"
-                    )
-                    if max_values:
-                        if (
-                            len(total_metrics) >= max_values
-                            and metric not in total_metrics
-                        ):
-                            continue
-                        total_metrics.add(metric)
-
-                    new_metrics.append(metric)
-                    path = f"last_metrics__{metric_key}__{variant_key}"
-                    for key, value in variant_data.items():
-                        if key == "min_value":
-                            extra_updates[f"min__{path}__min_value"] = value
-                        elif key == "max_value":
-                            extra_updates[f"max__{path}__max_value"] = value
-                        elif key in ("metric", "variant", "value"):
-                            extra_updates[f"set__{path}__{key}"] = value
-            if new_metrics:
-                extra_updates["add_to_set__unique_metrics"] = new_metrics
+            get_last_metric_updates(
+                task_id=task_id,
+                last_scalar_events=last_scalar_events,
+                raw_updates=raw_updates,
+                extra_updates=extra_updates,
+            )
 
         if last_events is not None:
 
-            def events_per_type(metric_data: Dict[str, dict]) -> Dict[str, EventStats]:
+            def events_per_type(metric_data_: Dict[str, dict]) -> Dict[str, EventStats]:
                 return {
                     event_type: EventStats(last_update=event["timestamp"])
-                    for event_type, event in metric_data.items()
+                    for event_type, event in metric_data_.items()
                 }
 
             metric_stats = {
@@ -454,28 +447,54 @@ class TaskBLL:
             }
             extra_updates["metric_stats"] = metric_stats
 
-        return TaskBLL.set_last_update(
+        ret = TaskBLL.set_last_update(
             task_ids=[task_id],
             company_id=company_id,
+            user_id=user_id,
             last_update=last_update,
             **extra_updates,
         )
+        if ret and raw_updates:
+            Task.objects(id=task_id).update_one(__raw__=[{"$set": raw_updates}])
+
+        return ret
+
+    @staticmethod
+    def remove_task_from_all_queues(company_id: str, task_id: str) -> int:
+        return Queue.objects(company=company_id, entries__task=task_id).update(
+            pull__entries__task=task_id, last_update=datetime.utcnow()
+        )
 
     @classmethod
     def dequeue_and_change_status(
-        cls, task: Task, company_id: str, status_message: str, status_reason: str,
+        cls,
+        task: Task,
+        company_id: str,
+        user_id: str,
+        status_message: str,
+        status_reason: str,
+        remove_from_all_queues=False,
+        new_status=None,
     ):
         try:
-            cls.dequeue(task, company_id)
-        except errors.bad_request.InvalidQueueOrTaskNotQueued:
+            cls.dequeue(task, company_id, silent_fail=True)
+        except APIError:
             # dequeue may fail if the queue was deleted
             pass
 
+        if remove_from_all_queues:
+            cls.remove_task_from_all_queues(company_id=company_id, task_id=task.id)
+
+        if task.status not in [TaskStatus.queued, TaskStatus.in_progress]:
+            return {"updated": 0}
+
         return ChangeStatusRequest(
             task=task,
-            new_status=task.enqueue_status or TaskStatus.created,
+            new_status=new_status or task.enqueue_status or TaskStatus.created,
             status_reason=status_reason,
             status_message=status_message,
+            user_id=user_id,
+            force=True,
         ).execute(enqueue_status=None)
 
     @classmethod

apiserver/bll/task/task_cleanup.py

@@ -1,10 +1,11 @@
 from datetime import datetime
 from itertools import chain
 from operator import attrgetter
-from typing import Sequence, Set, Tuple
+from typing import Sequence, Set, Tuple, Union
 
 import attr
-from boltons.iterutils import partition, bucketize, first
+from boltons.iterutils import partition, bucketize, first, chunked_iter
+from furl import furl
 from mongoengine import NotUniqueError
 from pymongo.errors import DuplicateKeyError
 
@@ -25,7 +26,6 @@ from apiserver.database.utils import id as db_id
 
 log = config.logger(__file__)
 event_bll = EventBLL()
-async_events_delete = config.get("services.tasks.async_events_delete", False)
 
 
 @attr.s(auto_attribs=True)
@@ -68,49 +68,74 @@ class CleanupResult:
         )
 
 
-def collect_plot_image_urls(company: str, task_or_model: str) -> Set[str]:
+def collect_plot_image_urls(
+    company: str, task_or_model: Union[str, Sequence[str]]
+) -> Set[str]:
     urls = set()
-    next_scroll_id = None
-    while True:
-        events, next_scroll_id = event_bll.get_plot_image_urls(
-            company_id=company, task_id=task_or_model, scroll_id=next_scroll_id
-        )
-        if not events:
-            break
-        for event in events:
-            event_urls = event.get(PlotFields.source_urls)
-            if event_urls:
-                urls.update(set(event_urls))
+    task_ids = task_or_model if isinstance(task_or_model, list) else [task_or_model]
+    for tasks in chunked_iter(task_ids, 100):
+        next_scroll_id = None
+        while True:
+            events, next_scroll_id = event_bll.get_plot_image_urls(
+                company_id=company, task_ids=tasks, scroll_id=next_scroll_id
+            )
+            if not events:
+                break
+            for event in events:
+                event_urls = event.get(PlotFields.source_urls)
+                if event_urls:
+                    urls.update(set(event_urls))
 
     return urls
 
 
-def collect_debug_image_urls(company: str, task_or_model: str) -> Set[str]:
+def collect_debug_image_urls(
+    company: str, task_or_model: Union[str, Sequence[str]]
+) -> Set[str]:
     """
     Return the set of unique image urls
     Uses DebugImagesIterator to make sure that we do not retrieve recycled urls
     """
-    after_key = None
     urls = set()
-    while True:
-        res, after_key = event_bll.get_debug_image_urls(
-            company_id=company, task_id=task_or_model, after_key=after_key,
-        )
-        urls.update(res)
-        if not after_key:
-            break
+    task_ids = task_or_model if isinstance(task_or_model, list) else [task_or_model]
+    for tasks in chunked_iter(task_ids, 100):
+        after_key = None
+        while True:
+            res, after_key = event_bll.get_debug_image_urls(
+                company_id=company,
+                task_ids=tasks,
+                after_key=after_key,
+            )
+            urls.update(res)
+            if not after_key:
+                break
 
     return urls
 
 
 supported_storage_types = {
-    "https://": StorageType.fileserver,
-    "http://": StorageType.fileserver,
+    "s3://": StorageType.s3,
+    "azure://": StorageType.azure,
+    "gs://": StorageType.gs,
 }
 
+supported_storage_types.update(
+    {
+        p: StorageType.fileserver
+        for p in config.get(
+            "services.async_urls_delete.fileserver.url_prefixes",
+            ["https://", "http://"],
+        )
+    }
+)
+
 
 def _schedule_for_delete(
-    company: str, user: str, task_id: str, urls: Set[str], can_delete_folders: bool,
+    company: str,
+    user: str,
+    task_id: str,
+    urls: Set[str],
+    can_delete_folders: bool,
 ) -> Set[str]:
     urls_per_storage = bucketize(
         urls,
@@ -129,7 +154,14 @@ def _schedule_for_delete(
         for url in storage_urls:
             folder = None
             if delete_folders:
-                folder, _, _ = url.rpartition("/")
+                try:
+                    parsed = furl(url)
+                    if parsed.path and len(parsed.path.segments) > 1:
+                        folder = parsed.remove(
+                            args=True, fragment=True, path=parsed.path.segments[-1]
+                        ).url.rstrip("/")
+                except Exception as ex:
+                    pass
 
             to_delete = folder or url
             if to_delete in scheduled_to_delete:
@@ -185,7 +217,7 @@ def cleanup_task(
         task, force
     )
     delete_external_artifacts = delete_external_artifacts and config.get(
-        "services.async_urls_delete.enabled", False
+        "services.async_urls_delete.enabled", True
     )
     event_urls, artifact_urls, model_urls = set(), set(), set()
     if return_file_urls or delete_external_artifacts:
@@ -203,8 +235,13 @@ def cleanup_task(
 
     deleted_task_id = f"{deleted_prefix}{task.id}"
     updated_children = 0
+    now = datetime.utcnow()
     if update_children:
-        updated_children = Task.objects(parent=task.id).update(parent=deleted_task_id)
+        updated_children = Task.objects(parent=task.id).update(
+            parent=deleted_task_id,
+            last_change=now,
+            last_changed_by=user,
+        )
 
     deleted_models = 0
     updated_models = 0
@@ -212,37 +249,41 @@ def cleanup_task(
         if not models:
             continue
         if delete_output_models and allow_delete:
-            model_ids = set(m.id for m in models if m.id not in in_use_model_ids)
-            for m_id in model_ids:
+            model_ids = list({m.id for m in models if m.id not in in_use_model_ids})
+            if model_ids:
                 if return_file_urls or delete_external_artifacts:
-                    event_urls.update(collect_debug_image_urls(task.company, m_id))
-                    event_urls.update(collect_plot_image_urls(task.company, m_id))
-                try:
-                    event_bll.delete_task_events(
-                        task.company,
-                        m_id,
-                        allow_locked=True,
-                        model=True,
-                        async_delete=async_events_delete,
-                    )
-                except errors.bad_request.InvalidModelId as ex:
-                    log.info(f"Error deleting events for the model {m_id}: {str(ex)}")
+                    event_urls.update(collect_debug_image_urls(task.company, model_ids))
+                    event_urls.update(collect_plot_image_urls(task.company, model_ids))
+
+                event_bll.delete_multi_task_events(
+                    task.company,
+                    model_ids,
+                    model=True,
+                )
+                deleted_models += Model.objects(id__in=model_ids).delete()
 
-            deleted_models += Model.objects(id__in=list(model_ids)).delete()
             if in_use_model_ids:
-                Model.objects(id__in=list(in_use_model_ids)).update(unset__task=1)
+                Model.objects(id__in=list(in_use_model_ids)).update(
+                    unset__task=1,
+                    set__last_change=now,
+                    set__last_changed_by=user,
+                )
             continue
 
         if update_children:
             updated_models += Model.objects(id__in=[m.id for m in models]).update(
-                task=deleted_task_id
+                task=deleted_task_id,
+                last_change=now,
+                last_changed_by=user,
             )
         else:
-            Model.objects(id__in=[m.id for m in models]).update(unset__task=1)
+            Model.objects(id__in=[m.id for m in models]).update(
+                unset__task=1,
+                set__last_change=now,
+                set__last_changed_by=user,
+            )
 
-    event_bll.delete_task_events(
-        task.company, task.id, allow_locked=force, async_delete=async_events_delete
-    )
+    event_bll.delete_task_events(task.company, task.id, allow_locked=force)
 
     if delete_external_artifacts:
         scheduled = _schedule_for_delete(
@@ -285,7 +326,8 @@ def verify_task_children_and_ouptuts(
 
     model_fields = ["id", "ready", "uri"]
     published_models, draft_models = partition(
-        Model.objects(task=task.id).only(*model_fields), key=attrgetter("ready"),
+        Model.objects(task=task.id).only(*model_fields),
+        key=attrgetter("ready"),
     )
     if not force and published_models:
         raise errors.bad_request.TaskCannotBeDeleted(

apiserver/bll/task/task_operations.py

@@ -7,9 +7,9 @@ from apiserver.bll.task import (
     TaskBLL,
     validate_status_change,
     ChangeStatusRequest,
-    update_project_time,
 )
 from apiserver.bll.task.task_cleanup import cleanup_task, CleanupResult
+from apiserver.bll.util import update_project_time
 from apiserver.config_repo import config
 from apiserver.database.model import EntityVisibility
 from apiserver.database.model.model import Model
@@ -23,6 +23,7 @@ from apiserver.database.model.task.task import (
     Execution,
     DEFAULT_LAST_ITERATION,
 )
+from apiserver.database.utils import get_options
 from apiserver.utilities.dicts import nested_set
 
 log = config.logger(__file__)
@@ -30,7 +31,11 @@ queue_bll = QueueBLL()
 
 
 def archive_task(
-    task: Union[str, Task], company_id: str, status_message: str, status_reason: str,
+    task: Union[str, Task],
+    company_id: str,
+    user_id: str,
+    status_message: str,
+    status_reason: str,
 ) -> int:
     """
     Deque and archive task
@@ -42,6 +47,7 @@ def archive_task(
             company_id=company_id,
             only=(
                 "id",
+                "company",
                 "execution",
                 "status",
                 "project",
@@ -52,7 +58,12 @@ def archive_task(
         )
     try:
         TaskBLL.dequeue_and_change_status(
-            task, company_id, status_message, status_reason,
+            task,
+            company_id=company_id,
+            user_id=user_id,
+            status_message=status_message,
+            status_reason=status_reason,
+            remove_from_all_queues=True,
         )
     except APIError:
         # dequeue may fail if the task was not enqueued
@@ -63,36 +74,73 @@ def archive_task(
         status_reason=status_reason,
         add_to_set__system_tags=EntityVisibility.archived.value,
         last_change=datetime.utcnow(),
+        last_changed_by=user_id,
     )
 
 
 def unarchive_task(
-    task: str, company_id: str, status_message: str, status_reason: str,
+    task: str,
+    company_id: str,
+    user_id: str,
+    status_message: str,
+    status_reason: str,
 ) -> int:
     """
     Unarchive task. Return 1 if successful
     """
     task = TaskBLL.get_task_with_access(
-        task, company_id=company_id, only=("id",), requires_write_access=True,
+        task,
+        company_id=company_id,
+        only=("id",),
+        requires_write_access=True,
     )
     return task.update(
         status_message=status_message,
         status_reason=status_reason,
         pull__system_tags=EntityVisibility.archived.value,
         last_change=datetime.utcnow(),
+        last_changed_by=user_id,
     )
 
 
 def dequeue_task(
-    task_id: str, company_id: str, status_message: str, status_reason: str,
+    task_id: str,
+    company_id: str,
+    user_id: str,
+    status_message: str,
+    status_reason: str,
+    remove_from_all_queues: bool = False,
+    new_status=None,
 ) -> Tuple[int, dict]:
-    query = dict(id=task_id, company=company_id)
-    task = Task.get_for_writing(**query)
+    if new_status and new_status not in get_options(TaskStatus):
+        raise errors.bad_request.ValidationError(f"Invalid task status: {new_status}")
+
+    # get the task without write access to make sure that it actually exists
+    task = Task.get(
+        id=task_id,
+        company=company_id,
+        _only=(
+            "id",
+            "company",
+            "execution",
+            "status",
+            "project",
+            "enqueue_status",
+        ),
+        include_public=True,
+    )
     if not task:
-        raise errors.bad_request.InvalidTaskId(**query)
+        TaskBLL.remove_task_from_all_queues(company_id, task_id=task_id)
+        return 1, {"updated": 0}
 
     res = TaskBLL.dequeue_and_change_status(
-        task, company_id, status_message=status_message, status_reason=status_reason,
+        task,
+        company_id=company_id,
+        user_id=user_id,
+        status_message=status_message,
+        status_reason=status_reason,
+        remove_from_all_queues=remove_from_all_queues,
+        new_status=new_status,
     )
     return 1, res
 
@@ -100,6 +148,7 @@ def dequeue_task(
 def enqueue_task(
     task_id: str,
     company_id: str,
+    user_id: str,
     queue_id: str,
     status_message: str,
     status_reason: str,
@@ -139,6 +188,7 @@ def enqueue_task(
         status_message=status_message,
         allow_same_state_transition=False,
         force=force,
+        user_id=user_id,
     ).execute(enqueue_status=task.status)
 
     try:
@@ -151,6 +201,7 @@ def enqueue_task(
             new_status=task.status,
             force=True,
             status_reason="failed enqueueing",
+            user_id=user_id,
         ).execute(enqueue_status=None)
         raise
 
@@ -220,8 +271,10 @@ def delete_task(
         TaskBLL.dequeue_and_change_status(
             task,
             company_id=company_id,
+            user_id=user_id,
             status_message=status_message,
             status_reason=status_reason,
+            remove_from_all_queues=True,
         )
     except APIError:
         # dequeue may fail if the task was not enqueued
@@ -240,6 +293,7 @@ def delete_task(
     if move_to_trash:
         # make sure that whatever changes were done to the task are saved
         # the task itself will be deleted later in the move_tasks_to_trash operation
+        task.last_update = datetime.utcnow()
         task.save()
     else:
         task.delete()
@@ -274,6 +328,8 @@ def reset_task(
         # dequeue may fail if the task was not enqueued
         pass
 
+    TaskBLL.remove_task_from_all_queues(company_id=company_id, task_id=task.id)
+
     cleaned_up = cleanup_task(
         company=company_id,
         user=user_id,
@@ -296,11 +352,17 @@ def reset_task(
         unset__output__error=1,
         unset__last_worker=1,
         unset__last_worker_report=1,
+        unset__started=1,
+        unset__completed=1,
+        unset__published=1,
+        unset__active_duration=1,
+        unset__enqueue_status=1,
     )
 
     if clear_all:
         updates.update(
-            set__execution=Execution(), unset__script=1,
+            set__execution=Execution(),
+            unset__script=1,
         )
     else:
         updates.update(unset__execution__queue=1)
@@ -319,12 +381,8 @@ def reset_task(
         force=force,
         status_reason="reset",
         status_message="reset",
+        user_id=user_id,
     ).execute(
-        started=None,
-        completed=None,
-        published=None,
-        active_duration=None,
-        enqueue_status=None,
         **updates,
     )
 
@@ -334,8 +392,9 @@ def reset_task(
 def publish_task(
     task_id: str,
     company_id: str,
+    user_id: str,
     force: bool,
-    publish_model_func: Callable[[str, str], Any] = None,
+    publish_model_func: Callable[[str, str, str], Any] = None,
     status_message: str = "",
     status_reason: str = "",
 ) -> dict:
@@ -363,7 +422,7 @@ def publish_task(
                 .first()
             )
             if model and not model.ready:
-                publish_model_func(model.id, company_id)
+                publish_model_func(model.id, company_id, user_id)
 
         # set task status to published, and update (or set) it's new output (view and models)
         return ChangeStatusRequest(
@@ -372,6 +431,7 @@ def publish_task(
             force=force,
             status_reason=status_reason,
             status_message=status_message,
+            user_id=user_id,
         ).execute(published=datetime.utcnow(), output=output)
 
     except Exception as ex:
@@ -384,7 +444,12 @@ def publish_task(
 
 
 def stop_task(
-    task_id: str, company_id: str, user_name: str, status_reason: str, force: bool,
+    task_id: str,
+    company_id: str,
+    user_id: str,
+    user_name: str,
+    status_reason: str,
+    force: bool,
 ) -> dict:
     """
     Stop a running task. Requires task status 'in_progress' and
@@ -446,4 +511,5 @@ def stop_task(
         status_reason=status_reason,
         status_message=status_message,
         force=force,
+        user_id=user_id,
     ).execute()

apiserver/bll/task/utils.py

@@ -1,12 +1,13 @@
 from datetime import datetime
-from typing import Sequence, Union
 
 import attr
 import six
 
 from apiserver.apierrors import errors
+from apiserver.bll.util import update_project_time
+from apiserver.config_repo import config
 from apiserver.database.errors import translate_errors_context
-from apiserver.database.model.project import Project
+from apiserver.database.model.model import Model
 from apiserver.database.model.task.task import Task, TaskStatus, TaskSystemTags
 from apiserver.database.utils import get_options
 from apiserver.utilities.attrs import typed_attrs
@@ -26,6 +27,7 @@ class ChangeStatusRequest(object):
     force = attr.ib(type=bool, default=False)
     allow_same_state_transition = attr.ib(type=bool, default=True)
     current_status_override = attr.ib(default=None)
+    user_id = attr.ib(type=str, default=None)
 
     def execute(self, **kwargs):
         current_status = self.current_status_override or self.task.status
@@ -44,6 +46,7 @@ class ChangeStatusRequest(object):
             status_changed=now,
             last_update=now,
             last_change=now,
+            last_changed_by=self.user_id,
         )
 
         if self.new_status == TaskStatus.queued:
@@ -154,16 +157,6 @@ def get_possible_status_changes(current_status):
     return possible
 
 
-def update_project_time(project_ids: Union[str, Sequence[str]]):
-    if not project_ids:
-        return
-
-    if isinstance(project_ids, str):
-        project_ids = [project_ids]
-
-    return Project.objects(id__in=project_ids).update(last_update=datetime.utcnow())
-
-
 def get_task_for_update(
     company_id: str, task_id: str, allow_all_statuses: bool = False, force: bool = False
 ) -> Task:
@@ -187,9 +180,88 @@ def get_task_for_update(
     return task
 
 
-def update_task(task: Task, update_cmds: dict, set_last_update: bool = True):
+def update_task(
+    task: Task, user_id: str, update_cmds: dict, set_last_update: bool = True
+):
     now = datetime.utcnow()
-    last_updates = dict(last_change=now)
+    last_updates = dict(last_change=now, last_changed_by=user_id)
     if set_last_update:
         last_updates.update(last_update=now)
     return task.update(**update_cmds, **last_updates)
+
+
+def get_last_metric_updates(
+    task_id: str,
+    last_scalar_events: dict,
+    raw_updates: dict,
+    extra_updates: dict,
+    model_events: bool = False,
+):
+    max_values = config.get("services.tasks.max_last_metrics", 2000)
+    total_metrics = set()
+    if max_values:
+        query = dict(id=task_id)
+        to_add = sum(len(v) for m, v in last_scalar_events.items())
+        if to_add <= max_values:
+            query[f"unique_metrics__{max_values - to_add}__exists"] = True
+        db_cls = Model if model_events else Task
+        task = db_cls.objects(**query).only("unique_metrics").first()
+        if task and task.unique_metrics:
+            total_metrics = set(task.unique_metrics)
+
+    new_metrics = []
+
+    def add_last_metric_conditional_update(
+        metric_path: str, metric_value, iter_value: int, is_min: bool
+    ):
+        """
+        Build an aggregation for an atomic update of the min or max value and the corresponding iteration
+        """
+        if is_min:
+            field_prefix = "min"
+            op = "$gt"
+        else:
+            field_prefix = "max"
+            op = "$lt"
+
+        value_field = f"{metric_path}__{field_prefix}_value".replace("__", ".")
+        condition = {
+            "$or": [
+                {"$lte": [f"${value_field}", None]},
+                {op: [f"${value_field}", metric_value]},
+            ]
+        }
+        raw_updates[value_field] = {
+            "$cond": [condition, metric_value, f"${value_field}"]
+        }
+
+        value_iteration_field = f"{metric_path}__{field_prefix}_value_iteration".replace(
+            "__", "."
+        )
+        raw_updates[value_iteration_field] = {
+            "$cond": [condition, iter_value, f"${value_iteration_field}"]
+        }
+
+    for metric_key, metric_data in last_scalar_events.items():
+        for variant_key, variant_data in metric_data.items():
+            metric = f"{variant_data.get('metric')}/{variant_data.get('variant')}"
+            if max_values:
+                if len(total_metrics) >= max_values and metric not in total_metrics:
+                    continue
+                total_metrics.add(metric)
+
+            new_metrics.append(metric)
+            path = f"last_metrics__{metric_key}__{variant_key}"
+            for key, value in variant_data.items():
+                if key in ("min_value", "max_value"):
+                    add_last_metric_conditional_update(
+                        metric_path=path,
+                        metric_value=value,
+                        iter_value=variant_data.get(f"{key}_iter", 0),
+                        is_min=(key == "min_value"),
+                    )
+                elif key in ("metric", "variant", "value"):
+                    extra_updates[f"set__{path}__{key}"] = value
+
+    if new_metrics:
+        extra_updates["add_to_set__unique_metrics"] = new_metrics

apiserver/bll/util.py

@@ -1,76 +1,24 @@
 import functools
 import itertools
 from concurrent.futures.thread import ThreadPoolExecutor
+from datetime import datetime
 from typing import (
     Optional,
     Callable,
-    Dict,
-    Any,
-    Set,
     Iterable,
     Tuple,
     Sequence,
     TypeVar,
+    Union,
 )
 
 from boltons import iterutils
 
 from apiserver.apierrors import APIError
-from apiserver.database.model import AttributedDocument
+from apiserver.database.model.project import Project
 from apiserver.database.model.settings import Settings
 
 
-class SetFieldsResolver:
-    """
-    The class receives set fields dictionary
-    and for the set fields that require 'min' or 'max'
-    operation replace them with a simple set in case the
-    DB document does not have these fields set
-    """
-
-    SET_MODIFIERS = ("min", "max")
-
-    def __init__(self, set_fields: Dict[str, Any]):
-        self.orig_fields = {}
-        self.fields = {}
-        self.add_fields(**set_fields)
-
-    def add_fields(self, **set_fields: Any):
-        self.orig_fields.update(set_fields)
-        self.fields.update(
-            {
-                f: fname
-                for f, modifier, dunder, fname in (
-                    (f,) + f.partition("__") for f in set_fields.keys()
-                )
-                if dunder and modifier in self.SET_MODIFIERS
-            }
-        )
-
-    def _get_updated_name(self, doc: AttributedDocument, name: str) -> str:
-        if name in self.fields and doc.get_field_value(self.fields[name]) is None:
-            return self.fields[name]
-        return name
-
-    def get_fields(self, doc: AttributedDocument):
-        """
-        For the given document return the set fields instructions
-        with min/max operations replaced with a single set in case
-        the document does not have the field set
-        """
-        return {
-            self._get_updated_name(doc, name): value
-            for name, value in self.orig_fields.items()
-        }
-
-    def get_names(self) -> Set[str]:
-        """
-        Returns the names of the fields that had min/max modifiers
-        in the format suitable for projection (dot separated)
-        """
-        return set(name.replace("__", ".") for name in self.fields.values())
-
-
 @functools.lru_cache()
 def get_server_uuid() -> Optional[str]:
     return Settings.get_by_key("server.uuid")
@@ -132,3 +80,13 @@ def run_batch_operation(
                 }
             )
     return results, failures
+
+
+def update_project_time(project_ids: Union[str, Sequence[str]]):
+    if not project_ids:
+        return
+
+    if isinstance(project_ids, str):
+        project_ids = [project_ids]
+
+    return Project.objects(id__in=project_ids).update(last_update=datetime.utcnow())

apiserver/bll/workers/__init__.py

@@ -5,13 +5,13 @@ from typing import Sequence, Set, Optional
 
 import attr
 import elasticsearch.helpers
-from boltons.iterutils import partition
+from boltons.iterutils import partition, chunked_iter
+from pyhocon import ConfigTree
 
 from apiserver.es_factory import es_factory
 from apiserver.apierrors import APIError
 from apiserver.apierrors.errors import bad_request, server_error
 from apiserver.apimodels.workers import (
-    DEFAULT_TIMEOUT,
     IdNameEntry,
     WorkerEntry,
     StatusReportRequest,
@@ -30,12 +30,14 @@ from apiserver.redis_manager import redman
 from apiserver.tools import safe_get
 from .stats import WorkerStats
 
+
 log = config.logger(__file__)
 
 
 class WorkerBLL:
     def __init__(self, es=None, redis=None):
         self.es_client = es or es_factory.connect("workers")
+        self.config = config.get("services.workers", ConfigTree())
         self.redis = redis or redman.connection("workers")
         self._stats = WorkerStats(self.es_client)
 
@@ -68,7 +70,7 @@ class WorkerBLL:
         """
         key = WorkerBLL._get_worker_key(company_id, user_id, worker)
 
-        timeout = timeout or DEFAULT_TIMEOUT
+        timeout = timeout or int(self.config.get("default_worker_timeout_sec", 10 * 60))
         queues = queues or []
 
         with translate_errors_context():
@@ -141,8 +143,6 @@ class WorkerBLL:
 
         try:
             entry.ip = ip
-            now = datetime.utcnow()
-            entry.last_activity_time = now
 
             if tags is not None:
                 entry.tags = tags
@@ -150,15 +150,16 @@ class WorkerBLL:
                 entry.system_tags = system_tags
 
             if report.machine_stats:
-                self._log_stats_to_es(
+                self.log_stats_to_es(
                     company_id=company_id,
-                    company_name=entry.company.name,
-                    worker=entry.key,
+                    worker_id=report.worker,
                     timestamp=report.timestamp,
                     task=report.task,
                     machine_stats=report.machine_stats,
                 )
 
+            now = datetime.utcnow()
+            entry.last_activity_time = now
             entry.queue = report.queue
 
             if report.queues:
@@ -175,6 +176,7 @@ class WorkerBLL:
                         last_worker_report=now,
                         last_update=now,
                         last_change=now,
+                        last_changed_by=user_id,
                     )
                     # modify(new=True, ...) returns the modified object
                     task = Task.objects(**query).modify(new=True, **update)
@@ -200,6 +202,24 @@ class WorkerBLL:
         finally:
             self._save_worker(entry)
 
+    def get_count(
+        self,
+        company_id: str,
+        last_seen: Optional[int] = None,
+        tags: Sequence[str] = None,
+        system_tags: Sequence[str] = None,
+    ):
+        if not last_seen:
+            return len(
+                self._get_keys(company_id, user_tags=tags, system_tags=system_tags)
+            )
+
+        return len(
+            self.get_all(
+                company_id, last_seen=last_seen, tags=tags, system_tags=system_tags
+            )
+        )
+
     def get_all(
         self,
         company_id: str,
@@ -235,18 +255,15 @@ class WorkerBLL:
         tags: Sequence[str] = None,
         system_tags: Sequence[str] = None,
     ) -> Sequence[WorkerResponseEntry]:
-
-        helpers = list(
-            map(
-                WorkerConversionHelper.from_worker_entry,
-                self.get_all(
-                    company_id=company_id,
-                    last_seen=last_seen,
-                    tags=tags,
-                    system_tags=system_tags,
-                ),
+        helpers = [
+            WorkerConversionHelper.from_worker_entry(entry)
+            for entry in self.get_all(
+                company_id=company_id,
+                last_seen=last_seen,
+                tags=tags,
+                system_tags=system_tags,
             )
-        )
+        ]
 
         task_ids = set(filter(None, (helper.task_id for helper in helpers)))
         all_queues = set(
@@ -265,9 +282,7 @@ class WorkerBLL:
                     }
                 },
             ]
-            queues_info = {
-                res["_id"]: res for res in Queue.objects.aggregate(projection)
-            }
+            queues_info = {res["_id"]: res for res in Queue.aggregate(projection)}
             task_ids = task_ids.union(
                 filter(
                     None,
@@ -396,15 +411,16 @@ class WorkerBLL:
             msg = "Failed saving worker entry"
             log.exception(msg)
 
-    def _get(
+    def _get_keys(
         self,
         company: str,
         user: str = "*",
-        worker_id: str = "*",
         user_tags: Sequence[str] = None,
         system_tags: Sequence[str] = None,
-    ) -> Sequence[WorkerEntry]:
-        """Get worker entries matching the company and user, worker patterns"""
+    ) -> Sequence[bytes]:
+        if not (user_tags or system_tags):
+            match = self._get_worker_key(company, user, "*")
+            return list(self.redis.scan_iter(match))
 
         def filter_by_user(in_keys: Set[bytes]) -> Set[bytes]:
             if user == "*":
@@ -412,67 +428,79 @@ class WorkerBLL:
             user_bytes = user.encode()
             return {k for k in in_keys if user_bytes in k}
 
-        if user_tags or system_tags:
-            worker_keys = set()
-            for tags, tags_field in (
-                (user_tags, "tags"),
-                (system_tags, "systemtags"),
-            ):
-                if not tags:
-                    continue
-                timestamp = int(time())
-                include, exclude = partition(tags, key=lambda x: x[0] != "-")
-                if include:
-                    tagged_workers = set()
-                    for tag in include:
-                        tagged_workers_key = self._get_tagged_workers_key(
-                            company, tags_field, tag
-                        )
-                        self.redis.zremrangebyscore(
-                            tagged_workers_key, min=0, max=timestamp
-                        )
-                        tagged_workers.update(
-                            self.redis.zrange(tagged_workers_key, 0, -1)
-                        )
-                    tagged_workers = filter_by_user(tagged_workers)
-                    worker_keys = (
-                        worker_keys.intersection(tagged_workers)
-                        if worker_keys
-                        else tagged_workers
+        worker_keys = set()
+        for tags, tags_field in (
+            (user_tags, "tags"),
+            (system_tags, "systemtags"),
+        ):
+            if not tags:
+                continue
+
+            timestamp = int(time())
+            include, exclude = partition(tags, key=lambda x: x[0] != "-")
+            if include:
+                tagged_workers = set()
+                for tag in include:
+                    tagged_workers_key = self._get_tagged_workers_key(
+                        company, tags_field, tag
+                    )
+                    self.redis.zremrangebyscore(
+                        tagged_workers_key, min=0, max=timestamp
+                    )
+                    tagged_workers.update(self.redis.zrange(tagged_workers_key, 0, -1))
+
+                tagged_workers = filter_by_user(tagged_workers)
+                worker_keys = (
+                    worker_keys.intersection(tagged_workers)
+                    if worker_keys
+                    else tagged_workers
+                )
+                if not worker_keys:
+                    return []
+
+            if exclude:
+                if not worker_keys:
+                    all_workers_key = self._get_all_workers_key(company)
+                    self.redis.zremrangebyscore(all_workers_key, min=0, max=timestamp)
+                    worker_keys.update(self.redis.zrange(all_workers_key, 0, -1))
+                    worker_keys = filter_by_user(worker_keys)
+                    if not worker_keys:
+                        return []
+
+                for tag in exclude:
+                    tagged_workers_key = self._get_tagged_workers_key(
+                        company, tags_field, tag[1:]
+                    )
+                    self.redis.zremrangebyscore(
+                        tagged_workers_key, min=0, max=timestamp
+                    )
+                    worker_keys.difference_update(
+                        self.redis.zrange(tagged_workers_key, 0, -1)
                     )
                     if not worker_keys:
                         return []
-                if exclude:
-                    if not worker_keys:
-                        all_workers_key = self._get_all_workers_key(company)
-                        self.redis.zremrangebyscore(
-                            all_workers_key, min=0, max=timestamp
-                        )
-                        worker_keys.update(self.redis.zrange(all_workers_key, 0, -1))
-                        worker_keys = filter_by_user(worker_keys)
-                        if not worker_keys:
-                            return []
-                    for tag in exclude:
-                        tagged_workers_key = self._get_tagged_workers_key(
-                            company, tags_field, tag[1:]
-                        )
-                        self.redis.zremrangebyscore(
-                            tagged_workers_key, min=0, max=timestamp
-                        )
-                        worker_keys.difference_update(
-                            self.redis.zrange(tagged_workers_key, 0, -1)
-                        )
-                        if not worker_keys:
-                            return []
-        else:
-            match = self._get_worker_key(company, user, "*")
-            worker_keys = self.redis.scan_iter(match)
+
+        return list(worker_keys)
+
+    def _get(
+        self,
+        company: str,
+        user: str = "*",
+        user_tags: Sequence[str] = None,
+        system_tags: Sequence[str] = None,
+    ) -> Sequence[WorkerEntry]:
+        """Get worker entries matching the company and user, worker patterns"""
 
         entries = []
-        for key in worker_keys:
-            data = self.redis.get(key)
+        for keys in chunked_iter(
+            self._get_keys(
+                company, user=user, user_tags=user_tags, system_tags=system_tags
+            ),
+            1000,
+        ):
+            data = self.redis.mget(keys)
             if data:
-                entries.append(WorkerEntry.from_json(data))
+                entries.extend(WorkerEntry.from_json(d) for d in data if d)
 
         return entries
 
@@ -481,18 +509,17 @@ class WorkerBLL:
         """Get the index name suffix for storing current month data"""
         return datetime.utcnow().strftime("%Y-%m")
 
-    def _log_stats_to_es(
+    def log_stats_to_es(
         self,
         company_id: str,
-        company_name: str,
-        worker: str,
+        worker_id: str,
         timestamp: int,
         task: str,
         machine_stats: MachineStats,
-    ) -> bool:
+    ) -> int:
         """
         Actually writing the worker statistics to Elastic
-        :return: True if successful, False otherwise
+        :return: The amount of logged documents
         """
         es_index = (
             f"{self._stats.worker_stats_prefix_for_company(company_id)}"
@@ -504,8 +531,7 @@ class WorkerBLL:
                 _index=es_index,
                 _source=dict(
                     timestamp=timestamp,
-                    worker=worker,
-                    company=company_name,
+                    worker=worker_id,
                     task=task,
                     category=category,
                     metric=metric,
@@ -530,7 +556,7 @@ class WorkerBLL:
 
         es_res = elasticsearch.helpers.bulk(self.es_client, actions)
         added, errors = es_res[:2]
-        return (added == len(actions)) and not errors
+        return added
 
 
 @attr.s(auto_attribs=True)

apiserver/bll/workers/stats.py

@@ -215,6 +215,10 @@ class WorkerStats:
                     "date_histogram": {
                         "field": "timestamp",
                         "fixed_interval": f"{interval}s",
+                        "extended_bounds": {
+                          "min": int(from_date) * 1000,
+                          "max": int(to_date) * 1000,
+                        }
                     },
                     "aggs": {"workers_count": {"cardinality": {"field": "worker"}}},
                 }

apiserver/config/default/apiserver.conf

@@ -41,10 +41,6 @@
         # controls whether FieldDoesNotExist exception will be raised for any extra attribute existing in stored data
         # but not declared in a data model
         strict: false
-
-        aggregate {
-            allow_disk_use: true
-        }
     }
 
     elastic {
@@ -117,6 +113,10 @@
         # Timeout in seconds on task status update. If exceeded
         # then task can be stopped without communicating to the worker
         task_update_timeout: 600
+
+        # Timeout in seconds for worker registration (or status report). If a worker did not report for this long,
+        # it is discarded from the server's table
+        default_timeout: 600
     }
 
     check_for_updates {

apiserver/config/default/secure.conf

@@ -23,6 +23,11 @@
             user_secret: "yfc8KQo*GMXb*9p((qcYC7ByFIpF7I&4VH3BfUYXH%o9vX1ZUZQEEw1Inc)S"
             revoke_in_fixed_mode: true
         }
+        services_agent {
+            role: "admin"
+            user_key: "P4BMJA7RK3TKBXGSY8OAA1FA8TOD11"
+            user_secret: "9LsgSfa0SYz0zli1_c500ZcLqanre2xkWOpepyt1w-BKK3_DKPHrtoj3JSHvyy8bIi0"
+        }
         tests {
             role: "user"
             display_name: "Default User"

apiserver/config/default/services/_mongo.conf

@@ -2,3 +2,8 @@ max_page_size: 500
 
 # expiration time in seconds for the redis scroll states in get_many family of apis
 scroll_state_expiration_seconds: 600
+
+allow_disk_use {
+    sort: true
+    aggregate: true
+}

apiserver/config/default/services/async_urls_delete.conf

@@ -1,6 +1,6 @@
-# if set to True then on task delete/reset external file urls for know storage types are scheduled for async delete
+# if set to true then on task delete/reset external file urls for known storage types are scheduled for async delete
 # otherwise they are returned to a client for the client side delete
-enabled: false
+enabled: true
 max_retries: 3
 retry_timeout_sec: 60
 
@@ -9,4 +9,4 @@ fileserver {
     # Can be in the form <schema>://host:port/path or /path
     url_prefixes: ["https://files.community-master.hosted.allegro.ai/"]
     timeout_sec: 300
-}
+}

apiserver/config/default/services/events.conf

@@ -32,6 +32,8 @@ events_retrieval {
     max_raw_scalars_size: 200000
 
     scroll_id_key: "cTN5VEtWEC6QrHvUl0FTx9kNyO0CcCK1p57akxma"
+
+    multi_plots_batch_size: 1000
 }
 
 # if set then plot str will be checked for the valid json on plot add

apiserver/config/default/services/models.conf

@@ -1,7 +1,4 @@
 metadata_values {
-    # maximal amount of distinct model values to retrieve
-    max_count: 100
-
     # cache ttl sec
     cache_ttl_sec: 86400
 }

apiserver/config/default/services/organization.conf

@@ -1,3 +1,9 @@
 tags_cache {
   expiration_seconds: 3600
+}
+download {
+    redis_timeout_sec: 300
+    batch_size: 500
+    max_download_items: 50000
+    max_project_name_length: 60
 }

apiserver/config/default/services/storage_credentials.conf

@@ -0,0 +1,53 @@
+aws {
+    s3 {
+        # S3 credentials, used for read/write access by various SDK elements
+        # default, used for any bucket not specified below
+        key: ""
+        secret: ""
+        region: ""
+        use_credentials_chain: false
+        # Additional ExtraArgs passed to boto3 when uploading files. Can also be set per-bucket under "credentials".
+        extra_args: {}
+        credentials: [
+            # specifies key/secret credentials to use when handling s3 urls (read or write)
+            # {
+            #     bucket: "my-bucket-name"
+            #     key: "my-access-key"
+            #     secret: "my-secret-key"
+            # },
+            {
+                # This will apply to all buckets in this host (unless key/value is specifically provided for a given bucket)
+                host: "localhost:9000"
+                key: "evg_user"
+                secret: "evg_pass"
+                multipart: false
+                secure: false
+            }
+        ]
+    }
+}
+google.storage {
+    # Default project and credentials file
+    # Will be used when no bucket configuration is found
+//    project: "clearml"
+//    credentials_json: "/path/to/credentials.json"
+//
+//    # Specific credentials per bucket and sub directory
+//    credentials = [
+//        {
+//            bucket: "my-bucket"
+//            subdir: "path/in/bucket" # Not required
+//            project: "clearml"
+//            credentials_json: "/path/to/credentials.json"
+//        },
+//    ]
+}
+azure.storage {
+    # containers: [
+    #     {
+    #         account_name: "clearml"
+    #         account_key: "secret"
+    #         # container_name:
+    #     }
+    # ]
+}

apiserver/config/default/services/tasks.conf

@@ -11,9 +11,6 @@ non_responsive_tasks_watchdog {
 multi_task_histogram_limit: 100
 
 hyperparam_values {
-    # maximal amount of distinct hyperparam values to retrieve
-    max_count: 100
-
     # max allowed outdate time for the cashed result
     cache_allowed_outdate_sec: 60
 
@@ -26,4 +23,6 @@ hyperparam_values {
 max_last_metrics: 2000
 
 # if set then call to tasks.delete/cleanup does not wait for ES events deletion
-async_events_delete: false
+async_events_delete: true
+# do not use async_delete if the deleted task has amount of events lower than this threshold
+async_events_delete_threshold: 100000

apiserver/database/errors.py

@@ -16,7 +16,7 @@ from mongoengine.errors import (
     LookUpError,
     InvalidQueryError,
 )
-from pymongo.errors import PyMongoError, NotMasterError
+from pymongo.errors import PyMongoError, NotPrimaryError
 
 from apiserver.apierrors import errors
 
@@ -198,7 +198,7 @@ def translate_errors_context(message=None, **kwargs):
         MongoEngineErrorsHandler.invalid_query_error(e, message, **kwargs)
     except PyMongoError as e:
         raise errors.server_error.InternalError(message, err=str(e))
-    except NotMasterError as e:
+    except NotPrimaryError as e:
         raise errors.server_error.InternalError(message, err=str(e))
     except MakeGetAllQueryError as e:
         raise errors.bad_request.ValidationError(e.error, field=e.field)

apiserver/database/model/base.py

@@ -1,5 +1,6 @@
 import re
-from collections import namedtuple
+from collections import namedtuple, defaultdict
+from datetime import datetime
 from functools import reduce, partial
 from typing import (
     Collection,
@@ -11,17 +12,18 @@ from typing import (
     Mapping,
     Any,
     Callable,
-    Dict,
-    List,
+    Generator,
 )
 
+import attr
 from boltons.iterutils import first, partition
 from dateutil.parser import parse as parse_datetime
-from mongoengine import Q, Document, ListField, StringField, IntField
+from mongoengine import Q, Document, ListField, StringField, IntField, QuerySet
 from pymongo.command_cursor import CommandCursor
 
-from apiserver.apierrors import errors
+from apiserver.apierrors import errors, APIError
 from apiserver.apierrors.base import BaseError
+from apiserver.apierrors.errors.bad_request import FieldsValueError
 from apiserver.bll.redis_cache_manager import RedisCacheManager
 from apiserver.config_repo import config
 from apiserver.database import Database
@@ -39,7 +41,7 @@ from apiserver.redis_manager import redman
 from apiserver.utilities.dicts import project_dict, exclude_fields_from_dict
 
 log = config.logger("dbmodel")
-
+mongo_conf = config.get("services._mongo")
 ACCESS_REGEX = re.compile(r"^(?P<prefix>>=|>|<=|<)?(?P<value>.*)$")
 ACCESS_MODIFIER = {">=": "gte", ">": "gt", "<=": "lte", "<": "lt"}
 
@@ -132,84 +134,135 @@ class GetMixin(PropsMixin):
             self.range_fields = range_fields
             self.pattern_fields = pattern_fields
 
-    class ListFieldBucketHelper:
+    class NewListFieldBucketHelper:
         op_prefix = "__$"
-        _legacy_exclude_prefix = "-"
-        _legacy_exclude_mongo_op = "nin"
-
-        default_mongo_op = "in"
-        _ops = {
-            # op -> (mongo_op, sticky)
-            "not": ("nin", False),
-            "nop": (default_mongo_op, False),
-            "all": ("all", True),
-            "and": ("all", True),
-            "any": (default_mongo_op, True),
-            "or": (default_mongo_op, True),
+        _unary_operators = {
+            "__$not": False,
+        }
+        _reset_operator = "__$nop"
+        _operators = {
+            "__$all": Q.AND,
+            "__$and": Q.AND,
+            "__$any": Q.OR,
+            "__$or": Q.OR,
+        }
+        default_operator = Q.OR
+        mongo_modifiers = {
+            # not_all modifier currently not supported due to the backwards compatibility
+            Q.AND: {True: "all", False: "nin"},
+            Q.OR: {True: "in", False: "nin"},
         }
 
-        def __init__(self, legacy=False):
-            self._current_op = None
-            self._sticky = False
-            self._support_legacy = legacy
+        @attr.s(auto_attribs=True)
+        class Term:
+            operator: str = None
+            reset: bool = False
+            include: bool = True
+            value: str = None
+
+        def __init__(self, field: str, data: Sequence[str], legacy=False):
+            self._field = field
             self.allow_empty = False
+            self.global_operator = None
+            self.actions = defaultdict(list)
+            self.explicit_operator = False
 
-        def _get_op(self, v: str, translate: bool = False) -> Optional[str]:
-            op = (
-                v[len(self.op_prefix) :] if v and v.startswith(self.op_prefix) else None
-            )
-            if translate:
-                tup = self._ops.get(op, None)
-                return tup[0] if tup else None
-            return op
-
-        def _key(self, v) -> Optional[Union[str, bool]]:
-            if v is None:
-                self.allow_empty = True
-                return None
-
-            op = self._get_op(v)
-            if op is not None:
-                # operator - set state and return None
-                self._current_op, self._sticky = self._ops.get(
-                    op, (self.default_mongo_op, self._sticky)
-                )
-                return None
-            elif self._current_op:
-                current_op = self._current_op
-                if not self._sticky:
-                    self._current_op = None
-                return current_op
-            elif self._support_legacy and v.startswith(self._legacy_exclude_prefix):
-                self._current_op = None
-                return False
-
-            return self.default_mongo_op
-
-        def get_global_op(self, data: Sequence[str]) -> int:
-            op_to_res = {
-                "in": Q.OR,
-                "all": Q.AND,
-            }
-            data = (x for x in data if x is not None)
-            first_op = (
-                self._get_op(next(data, ""), translate=True) or self.default_mongo_op
-            )
-            return op_to_res.get(first_op, self.default_mongo_op)
-
-        def get_actions(self, data: Sequence[str]) -> Dict[str, List[Union[str, None]]]:
-            actions = {}
-
-            for val in data:
-                key = self._key(val)
-                if key is None:
+            self._support_legacy = legacy
+            current_context = self.default_operator
+            for d in self._get_next_term(data):
+                if d.operator is not None:
+                    current_context = d.operator
+                    self._support_legacy = False
+                    if self.global_operator is None:
+                        self.global_operator = d.operator
+                        self.explicit_operator = True
                     continue
-                elif self._support_legacy and key is False:
-                    key = self._legacy_exclude_mongo_op
-                    val = val[len(self._legacy_exclude_prefix) :]
-                actions.setdefault(key, []).append(val)
 
-            return actions
+                if self.global_operator is None:
+                    self.global_operator = self.default_operator
+
+                if d.reset:
+                    current_context = self.default_operator
+                    self._support_legacy = legacy
+                    continue
+
+                if d.value is None:
+                    self.allow_empty = True
+                    continue
+
+                self.actions[self.mongo_modifiers[current_context][d.include]].append(
+                    d.value
+                )
+
+            if self.global_operator is None:
+                self.global_operator = self.default_operator
+
+        def _get_next_term(self, data: Sequence[str]) -> Generator[Term, None, None]:
+            unary_operator = None
+            for value in data:
+                if value is None:
+                    unary_operator = None
+                    yield self.Term()
+                    continue
+
+                if not isinstance(value, str):
+                    raise FieldsValueError(
+                        "invalid value type, string expected",
+                        field=self._field,
+                        value=str(value),
+                    )
+
+                if value == self._reset_operator:
+                    unary_operator = None
+                    yield self.Term(reset=True)
+                    continue
+
+                if value.startswith(self.op_prefix):
+                    if unary_operator:
+                        raise FieldsValueError(
+                            "Value is expected after",
+                            field=self._field,
+                            operator=unary_operator,
+                        )
+                    if value in self._unary_operators:
+                        unary_operator = value
+                        continue
+
+                    operator = self._operators.get(value)
+                    if operator is None:
+                        raise FieldsValueError(
+                            "Unsupported operator",
+                            field=self._field,
+                            operator=value,
+                        )
+                    yield self.Term(operator=operator)
+                    continue
+
+                if (
+                    not unary_operator
+                    and self._support_legacy
+                    and value.startswith("-")
+                ):
+                    value = value[1:]
+                    if not value:
+                        raise FieldsValueError(
+                            "Missing value after the exclude prefix -",
+                            field=self._field,
+                            value=value,
+                        )
+                    yield self.Term(value=value, include=False)
+                    continue
+
+                term = self.Term(value=value)
+                if unary_operator:
+                    term.include = self._unary_operators[unary_operator]
+                    unary_operator = None
+                yield term
+
+            if unary_operator:
+                raise FieldsValueError(
+                    "Value is expected after", operator=unary_operator
+                )
 
     get_all_query_options = QueryParameterOptions()
 
@@ -227,8 +280,8 @@ class GetMixin(PropsMixin):
             cls._cache_manager = RedisCacheManager(
                 state_class=cls.GetManyScrollState,
                 redis=redman.connection("apiserver"),
-                expiration_interval=config.get(
-                    "services._mongo.scroll_state_expiration_seconds", 600
+                expiration_interval=mongo_conf.get(
+                    "scroll_state_expiration_seconds", 600
                 ),
             )
 
@@ -347,97 +400,121 @@ class GetMixin(PropsMixin):
         parameters_options = parameters_options or cls.get_all_query_options
         dict_query = {}
         query = RegexQ()
-        if parameters:
-            parameters = {
-                k: cls._get_fixed_field_value(k, v) for k, v in parameters.items()
-            }
-            opts = parameters_options
-            for field in opts.pattern_fields:
-                pattern = parameters.pop(field, None)
-                if pattern:
-                    dict_query[field] = RegexWrapper(pattern)
+        field = None
+        # noinspection PyBroadException
+        try:
+            if parameters:
+                parameters = {
+                    k: cls._get_fixed_field_value(k, v) for k, v in parameters.items()
+                }
+                filters = parameters.pop("filters", {})
+                if not isinstance(filters, dict):
+                    raise FieldsValueError(
+                        "invalid value type, string expected",
+                        field=filters,
+                        value=str(filters),
+                    )
+                opts = parameters_options
+                for field in opts.pattern_fields:
+                    pattern = parameters.pop(field, None)
+                    if pattern:
+                        dict_query[field] = RegexWrapper(pattern)
 
-            for field, data in cls._pop_matching_params(
-                patterns=opts.list_fields, parameters=parameters
-            ).items():
-                query &= cls.get_list_field_query(field, data)
+                for field, data in cls._pop_matching_params(
+                    patterns=opts.list_fields, parameters=filters
+                ).items():
+                    query &= cls.get_list_filter_query(field, data)
+                    parameters.pop(field, None)
 
-            for field, data in cls._pop_matching_params(
-                patterns=opts.range_fields, parameters=parameters
-            ).items():
-                query &= cls.get_range_field_query(field, data)
+                for field, data in cls._pop_matching_params(
+                    patterns=opts.list_fields, parameters=parameters
+                ).items():
+                    query &= cls.get_list_field_query(field, data)
 
-            for field, data in cls._pop_matching_params(
-                patterns=opts.fields or [], parameters=parameters
-            ).items():
-                if "._" in field or "_." in field:
-                    query &= RegexQ(__raw__={field: data})
-                else:
-                    dict_query[field.replace(".", "__")] = data
+                for field, data in cls._pop_matching_params(
+                    patterns=opts.range_fields, parameters=parameters
+                ).items():
+                    query &= cls.get_range_field_query(field, data)
 
-            for field in opts.datetime_fields or []:
-                data = parameters.pop(field, None)
-                if data is not None:
-                    if not isinstance(data, list):
-                        data = [data]
-                    # date time fields also support simplified range queries. Check if this is the case
-                    if len(data) == 2 and not any(
-                        d.startswith(mod)
-                        for d in data
-                        if d is not None
-                        for mod in ACCESS_MODIFIER
-                    ):
-                        query &= cls.get_range_field_query(field, data)
+                for field, data in cls._pop_matching_params(
+                    patterns=opts.fields or [], parameters=parameters
+                ).items():
+                    if "._" in field or "_." in field:
+                        query &= RegexQ(__raw__={field: data})
                     else:
-                        for d in data:  # type: str
-                            m = ACCESS_REGEX.match(d)
-                            if not m:
-                                continue
-                            try:
-                                value = parse_datetime(m.group("value"))
-                                prefix = m.group("prefix")
-                                modifier = ACCESS_MODIFIER.get(prefix)
-                                f = (
-                                    field
-                                    if not modifier
-                                    else "__".join((field, modifier))
-                                )
-                                dict_query[f] = value
-                            except (ValueError, OverflowError):
-                                pass
+                        dict_query[field.replace(".", "__")] = data
 
-            for field, value in parameters.items():
-                for keys, func in cls._multi_field_param_prefix.items():
-                    if field not in keys:
-                        continue
-                    try:
-                        data = cls.MultiFieldParameters(**value)
-                    except Exception:
-                        raise MakeGetAllQueryError("incorrect field format", field)
-                    if not data.fields:
-                        break
-                    if any("._" in f for f in data.fields):
-                        q = reduce(
-                            lambda a, x: func(
-                                a,
-                                RegexQ(
-                                    __raw__={
-                                        x: {"$regex": data.pattern, "$options": "i"}
-                                    }
+                for field in opts.datetime_fields or []:
+                    data = parameters.pop(field, None)
+                    if data is not None:
+                        if not isinstance(data, list):
+                            data = [data]
+                        # date time fields also support simplified range queries. Check if this is the case
+                        if len(data) == 2 and not any(
+                            d.startswith(mod)
+                            for d in data
+                            if d is not None
+                            for mod in ACCESS_MODIFIER
+                        ):
+                            query &= cls.get_range_field_query(field, data)
+                        else:
+                            for d in data:  # type: str
+                                m = ACCESS_REGEX.match(d)
+                                if not m:
+                                    continue
+                                try:
+                                    value = parse_datetime(m.group("value"))
+                                    prefix = m.group("prefix")
+                                    modifier = ACCESS_MODIFIER.get(prefix)
+                                    f = (
+                                        field
+                                        if not modifier
+                                        else "__".join((field, modifier))
+                                    )
+                                    dict_query[f] = value
+                                except (ValueError, OverflowError):
+                                    pass
+
+                for field, value in parameters.items():
+                    for keys, func in cls._multi_field_param_prefix.items():
+                        if field not in keys:
+                            continue
+                        try:
+                            data = cls.MultiFieldParameters(**value)
+                        except Exception:
+                            raise MakeGetAllQueryError("incorrect field format", field)
+                        if not data.fields:
+                            break
+                        if any("._" in f for f in data.fields):
+                            q = reduce(
+                                lambda a, x: func(
+                                    a,
+                                    RegexQ(
+                                        __raw__={
+                                            x: {"$regex": data.pattern, "$options": "i"}
+                                        }
+                                    ),
                                 ),
-                            ),
-                            data.fields,
-                            RegexQ(),
-                        )
-                    else:
-                        regex = RegexWrapper(data.pattern, flags=re.IGNORECASE)
-                        sep_fields = [f.replace(".", "__") for f in data.fields]
-                        q = reduce(
-                            lambda a, x: func(a, RegexQ(**{x: regex})),
-                            sep_fields,
-                            RegexQ(),
-                        )
-                    query = query & q
+                                data.fields,
+                                RegexQ(),
+                            )
+                        else:
+                            regex = RegexWrapper(data.pattern, flags=re.IGNORECASE)
+                            sep_fields = [f.replace(".", "__") for f in data.fields]
+                            q = reduce(
+                                lambda a, x: func(a, RegexQ(**{x: regex})),
+                                sep_fields,
+                                RegexQ(),
+                            )
+                        query = query & q
+        except APIError:
+            raise
+        except Exception as ex:
+            raise errors.bad_request.FieldsValueError(
+                "failed parsing query field",
+                error=str(ex),
+                **({"field": field} if field else {}),
+            )
 
         return query & RegexQ(**dict_query)
 
@@ -472,6 +549,135 @@ class GetMixin(PropsMixin):
 
         return q
 
+    @attr.s(auto_attribs=True)
+    class ListQueryFilter:
+        """
+        Deserialize filters data and build db_query object that represents it with the corresponding
+        mongo engine operations
+        Each part has include and exclude lists that map to mongoengine operations as following:
+        "any"
+          - include -> 'in'
+          - exclude -> 'not_all'
+          - combined by 'or' operation
+        "all"
+          - include -> 'all'
+          - exclude -> 'nin'
+          - combined by 'and' operation
+        "op" optional parameter for combining "and" and "all" parts. Can be "and" or "or". The default is "and"
+        """
+
+        _and_op = "and"
+        _or_op = "or"
+        _allowed_op = [_and_op, _or_op]
+        _db_modifiers: Mapping = {
+            (Q.OR, True): "in",
+            (Q.OR, False): "not__all",
+            (Q.AND, True): "all",
+            (Q.AND, False): "nin",
+        }
+
+        @attr.s(auto_attribs=True)
+        class ListFilter:
+            include: Sequence[str] = []
+            exclude: Sequence[str] = []
+
+            @classmethod
+            def from_dict(cls, d: Mapping):
+                if d is None:
+                    return None
+                return cls(**d)
+
+        any: ListFilter = attr.ib(converter=ListFilter.from_dict, default=None)
+        all: ListFilter = attr.ib(converter=ListFilter.from_dict, default=None)
+        op: str = attr.ib(default="and")
+        db_query: dict = attr.ib(init=False)
+
+        # noinspection PyUnresolvedReferences
+        @op.validator
+        def op_validator(self, _, value):
+            if value not in self._allowed_op:
+                raise ValueError(
+                    f"Invalid list query filter operator: {value}. "
+                    f"Should be one of {str(self._allowed_op)}"
+                )
+
+        @property
+        def and_op(self) -> bool:
+            return self.op == self._and_op
+
+        def __attrs_post_init__(self):
+            self.db_query = {}
+            for op, conditions in ((Q.OR, self.any), (Q.AND, self.all)):
+                if not conditions:
+                    continue
+
+                operations = {}
+                for vals, include in (
+                    (conditions.include, True),
+                    (conditions.exclude, False),
+                ):
+                    if not vals:
+                        continue
+                    operations[self._db_modifiers[(op, include)]] = list(set(vals))
+
+                self.db_query[op] = operations
+
+        @classmethod
+        def from_data(cls, field, data: Mapping):
+            if not isinstance(data, dict):
+                raise errors.bad_request.ValidationError(
+                    "invalid filter for field, dictionary expected",
+                    field=field,
+                    value=str(data),
+                )
+
+            try:
+                return cls(**data)
+            except Exception as ex:
+                raise errors.bad_request.ValidationError(
+                    field=field,
+                    value=str(ex),
+                )
+
+    @classmethod
+    def get_list_filter_query(
+        cls, field: str, data: Mapping
+    ) -> Union[RegexQ, RegexQCombination]:
+        if not data:
+            return RegexQ()
+
+        filter_ = cls.ListQueryFilter.from_data(field, data)
+
+        mongoengine_field = field.replace(".", "__")
+        queries = []
+        for op, actions in filter_.db_query.items():
+            if not actions:
+                continue
+
+            ops = []
+            for action, vals in actions.items():
+                if not vals:
+                    continue
+
+                ops.append(RegexQ(**{f"{mongoengine_field}__{action}": vals}))
+
+            if not ops:
+                continue
+
+            if len(ops) == 1:
+                queries.extend(ops)
+                continue
+
+            queries.append(RegexQCombination(operation=op, children=ops))
+
+        if not queries:
+            return RegexQ()
+        if len(queries) == 1:
+            return queries[0]
+
+        operation = Q.AND if filter_.and_op else Q.OR
+        return RegexQCombination(operation=operation, children=queries)
+
     @classmethod
     def get_list_field_query(cls, field: str, data: Sequence[Optional[str]]) -> RegexQ:
         """
@@ -486,15 +692,15 @@ class GetMixin(PropsMixin):
         if not isinstance(data, (list, tuple)):
             data = [data]
 
-        helper = cls.ListFieldBucketHelper(legacy=True)
-        global_op = helper.get_global_op(data)
-        actions = helper.get_actions(data)
+        helper = cls.NewListFieldBucketHelper(field, data=data, legacy=True)
+        global_op = helper.global_operator
+        actions = helper.actions
 
         mongoengine_field = field.replace(".", "__")
 
         queries = [
-            RegexQ(**{f"{mongoengine_field}__{action}": list(set(actions[action]))})
-            for action in filter(None, actions)
+            RegexQ(**{f"{mongoengine_field}__{action}": list(set(values))})
+            for action, values in actions.items()
         ]
 
         if not queries:
@@ -555,7 +761,7 @@ class GetMixin(PropsMixin):
         if start is not None:
             return start, cls.validate_scroll_size(parameters)
 
-        max_page_size = config.get("services._mongo.max_page_size", 500)
+        max_page_size = mongo_conf.get("max_page_size", 500)
         page = parameters.get("page", default_page)
         if page is not None and page < 0:
             raise errors.bad_request.ValidationError("page must be >=0", field="page")
@@ -580,7 +786,7 @@ class GetMixin(PropsMixin):
 
     @classmethod
     def get_projection(cls, parameters, override_projection=None, **__):
-        """ Extract a projection list from the provided dictionary. Supports an override projection. """
+        """Extract a projection list from the provided dictionary. Supports an override projection."""
         if override_projection is not None:
             return override_projection
         if not parameters:
@@ -594,7 +800,8 @@ class GetMixin(PropsMixin):
         """Return include and exclude lists based on passed projection and class definition"""
         if projection:
             include, exclude = partition(
-                projection, key=lambda x: x[0] != ProjectionHelper.exclusion_prefix,
+                projection,
+                key=lambda x: x[0] != ProjectionHelper.exclusion_prefix,
             )
         else:
             include, exclude = [], []
@@ -733,7 +940,9 @@ class GetMixin(PropsMixin):
     @classmethod
     def _get_collation_override(cls, field: str) -> Optional[dict]:
         return first(
-            v for k, v in cls._field_collation_overrides.items() if field.startswith(k)
+            v
+            for k, v in cls._field_collation_overrides.items()
+            if field.startswith(k) or field.startswith(f"-{k}")
         )
 
     @classmethod
@@ -839,7 +1048,9 @@ class GetMixin(PropsMixin):
                 projection_fields=projection_fields,
             )
             return cls.get_data_with_scroll_support(
-                query_dict=query_dict, data_getter=data_getter, ret_params=ret_params,
+                query_dict=query_dict,
+                data_getter=data_getter,
+                ret_params=ret_params,
             )
 
         return cls._get_many_no_company(
@@ -852,7 +1063,9 @@ class GetMixin(PropsMixin):
 
     @classmethod
     def get_many_public(
-        cls, query: Q = None, projection: Collection[str] = None,
+        cls,
+        query: Q = None,
+        projection: Collection[str] = None,
     ):
         """
         Fetch all public documents matching a provided query.
@@ -865,6 +1078,13 @@ class GetMixin(PropsMixin):
 
         return cls._get_many_no_company(query=_query, override_projection=projection)
 
+    @staticmethod
+    def _get_qs_with_ordering(qs: QuerySet, order_by: Sequence):
+        disk_use_setting = mongo_conf.get("allow_disk_use.sort", None)
+        if disk_use_setting is not None:
+            qs = qs.allow_disk_use(disk_use_setting)
+        return qs.order_by(*order_by)
+
     @classmethod
     def _get_many_no_company(
         cls: Union["GetMixin", Document],
@@ -906,7 +1126,7 @@ class GetMixin(PropsMixin):
             qs = qs.search_text(search_text)
         if order_by:
             # add ordering
-            qs = qs.order_by(*order_by)
+            qs = cls._get_qs_with_ordering(qs, order_by)
 
         if include:
             # add projection
@@ -998,7 +1218,7 @@ class GetMixin(PropsMixin):
             res = cls._get_queries_for_order_field(query, order_field)
             if res:
                 query_sets = [cls.objects(q) for q in res]
-            query_sets = [qs.order_by(*order_by) for qs in query_sets]
+            query_sets = [cls._get_qs_with_ordering(qs, order_by) for qs in query_sets]
             if order_field and not override_collation:
                 override_collation = cls._get_collation_override(order_field)
 
@@ -1138,7 +1358,7 @@ class UpdateMixin(object):
 
 
 class DbModelMixin(GetMixin, ProperDictMixin, UpdateMixin):
-    """ Provide convenience methods for a subclass of mongoengine.Document """
+    """Provide convenience methods for a subclass of mongoengine.Document"""
 
     @classmethod
     def aggregate(
@@ -1158,7 +1378,7 @@ class DbModelMixin(GetMixin, ProperDictMixin, UpdateMixin):
         kwargs.update(
             allowDiskUse=allow_disk_use
             if allow_disk_use is not None
-            else config.get("apiserver.mongo.aggregate.allow_disk_use", True)
+            else mongo_conf.get("allow_disk_use.aggregate", True)
         )
         return cls.objects.aggregate(pipeline, **kwargs)
 
@@ -1166,25 +1386,31 @@ class DbModelMixin(GetMixin, ProperDictMixin, UpdateMixin):
     def set_public(
         cls: Type[Document],
         company_id: str,
+        user_id: str,
         ids: Sequence[str],
         invalid_cls: Type[BaseError],
         enabled: bool = True,
     ):
         if enabled:
             items = list(cls.objects(id__in=ids, company=company_id).only("id"))
-            update = dict(set__company_origin=company_id, set__company="")
+            update: dict = dict(set__company_origin=company_id, set__company="")
         else:
             items = list(
                 cls.objects(
                     id__in=ids, company__in=(None, ""), company_origin=company_id
                 ).only("id")
             )
-            update = dict(set__company=company_id, unset__company_origin=1)
+            update: dict = dict(set__company=company_id, unset__company_origin=1)
 
         if len(items) < len(ids):
             missing = tuple(set(ids).difference(i.id for i in items))
             raise invalid_cls(ids=missing)
 
+        if hasattr(cls, "last_change"):
+            update["set__last_change"] = datetime.utcnow()
+        if hasattr(cls, "last_changed_by"):
+            update["set__last_changed_by"] = user_id
+
         return {"updated": cls.objects(id__in=ids).update(**update)}
 
 

apiserver/database/model/model.py

@@ -3,6 +3,8 @@ from mongoengine import (
     DateTimeField,
     BooleanField,
     EmbeddedDocumentField,
+    IntField,
+    ListField,
 )
 
 from apiserver.database import Database, strict
@@ -17,12 +19,14 @@ from apiserver.database.model.base import GetMixin
 from apiserver.database.model.metadata import MetadataItem
 from apiserver.database.model.model_labels import ModelLabels
 from apiserver.database.model.project import Project
+from apiserver.database.model.task.metrics import MetricEvent
 from apiserver.database.model.task.task import Task
 
 
 class Model(AttributedDocument):
     _field_collation_overrides = {
         "metadata.": AttributedDocument._numeric_locale,
+        "last_metrics.": AttributedDocument._numeric_locale,
     }
 
     meta = {
@@ -67,6 +71,7 @@ class Model(AttributedDocument):
             "parent",
             "metadata.*",
         ),
+        range_fields=("last_metrics.*", "last_iteration"),
         datetime_fields=("last_update",),
     )
 
@@ -85,6 +90,8 @@ class Model(AttributedDocument):
     labels = ModelLabels()
     ready = BooleanField(required=True)
     last_update = DateTimeField()
+    last_change = DateTimeField()
+    last_changed_by = StringField()
     ui_cache = SafeDictField(
         default=dict, user_set_allowed=True, exclude_by_default=True
     )
@@ -92,6 +99,9 @@ class Model(AttributedDocument):
     metadata = SafeMapField(
         field=EmbeddedDocumentField(MetadataItem), user_set_allowed=True
     )
+    last_iteration = IntField(default=0)
+    last_metrics = SafeMapField(field=SafeMapField(EmbeddedDocumentField(MetricEvent)))
+    unique_metrics = ListField(StringField(required=True), exclude_by_default=True)
 
     def get_index_company(self) -> str:
         return self.company or self.company_origin or ""

apiserver/database/model/task/metrics.py

@@ -4,6 +4,7 @@ from mongoengine import (
     DynamicField,
     LongField,
     EmbeddedDocumentField,
+    IntField,
 )
 
 from apiserver.database.fields import SafeMapField
@@ -19,7 +20,9 @@ class MetricEvent(EmbeddedDocument):
     variant = StringField(required=True)
     value = DynamicField(required=True)
     min_value = DynamicField()  # for backwards compatibility reasons
+    min_value_iteration = IntField()
     max_value = DynamicField()  # for backwards compatibility reasons
+    max_value_iteration = IntField()
 
 
 class EventStats(EmbeddedDocument):

apiserver/database/model/task/task.py

@@ -19,6 +19,7 @@ from apiserver.database.fields import (
     SafeSortedListField,
     EmbeddedDocumentListField,
     NullableStringField,
+    NoneType,
 )
 from apiserver.database.model import AttributedDocument
 from apiserver.database.model.base import ProperDictMixin, GetMixin
@@ -89,7 +90,9 @@ class Artifact(EmbeddedDocument):
     content_size = LongField()
     timestamp = LongField()
     type_data = EmbeddedDocumentField(ArtifactTypeData)
-    display_data = SafeSortedListField(ListField(UnionField((int, float, str))))
+    display_data = SafeSortedListField(
+        ListField(UnionField((int, float, str, NoneType)))
+    )
 
 
 class ParamsItem(EmbeddedDocument, ProperDictMixin):
@@ -149,6 +152,7 @@ class TaskType(object):
     application = "application"
     monitor = "monitor"
     controller = "controller"
+    report = "report"
     optimizer = "optimizer"
     service = "service"
     qc = "qc"
@@ -195,6 +199,7 @@ class Task(AttributedDocument):
                     "$name",
                     "$id",
                     "$comment",
+                    "$report",
                     "$models.input.model",
                     "$models.output.model",
                     "$script.repository",
@@ -205,6 +210,7 @@ class Task(AttributedDocument):
                     "name": 10,
                     "id": 10,
                     "comment": 10,
+                    "report": 10,
                     "models.output.model": 2,
                     "models.input.model": 2,
                     "script.repository": 1,
@@ -224,10 +230,12 @@ class Task(AttributedDocument):
             "project",
             "parent",
             "hyperparams.*",
+            "execution.queue",
         ),
         range_fields=("started", "active_duration", "last_metrics.*", "last_iteration"),
         datetime_fields=("status_changed", "last_update"),
-        pattern_fields=("name", "comment"),
+        pattern_fields=("name", "comment", "report"),
+        fields=("runtime.*", "models.input.model"),
     )
 
     id = StringField(primary_key=True)
@@ -241,6 +249,8 @@ class Task(AttributedDocument):
     status_message = StringField(user_set_allowed=True)
     status_changed = DateTimeField()
     comment = StringField(user_set_allowed=True)
+    report = StringField()
+    report_assets = ListField(StringField())
     created = DateTimeField(required=True, user_set_allowed=True)
     started = DateTimeField()
     completed = DateTimeField()
@@ -262,7 +272,7 @@ class Task(AttributedDocument):
     unique_metrics = ListField(StringField(required=True), exclude_by_default=True)
     metric_stats = SafeMapField(field=EmbeddedDocumentField(MetricEventStats))
     company_origin = StringField(exclude_by_default=True)
-    duration = IntField()  # task duration in seconds
+    duration = IntField()  # obsolete, do not use
     hyperparams = SafeMapField(field=SafeMapField(EmbeddedDocumentField(ParamsItem)))
     configuration = SafeMapField(field=EmbeddedDocumentField(ConfigurationItem))
     runtime = SafeDictField(default=dict)
@@ -271,6 +281,7 @@ class Task(AttributedDocument):
     enqueue_status = StringField(
         choices=get_options(TaskStatus), exclude_by_default=True
     )
+    last_changed_by = StringField()
 
     def get_index_company(self) -> str:
         """

apiserver/database/model/url_to_delete.py

@@ -8,6 +8,9 @@ from apiserver.database.model import AttributedDocument
 
 class StorageType(str, Enum):
     fileserver = "fileserver"
+    s3 = "s3"
+    azure = "azure"
+    gs = "gs"
     unknown = "unknown"
 
 
@@ -32,10 +35,8 @@ class UrlToDelete(AttributedDocument):
         "strict": strict,
         "indexes": [
             ("company", "user", "task"),
-            "storage_type",
-            "created",
-            "retry_count",
-            "type",
+            ("company", "storage_type", "url"),
+            ("status", "retry_count", "storage_type"),
         ],
     }
 

apiserver/database/projection.py

@@ -3,10 +3,14 @@ from concurrent.futures import ThreadPoolExecutor
 from itertools import groupby, chain
 from typing import Sequence, Dict, Callable
 
+from boltons import iterutils
+
 from apiserver.apierrors import errors
+from apiserver.config_repo import config
 from apiserver.database.props import PropsMixin
 
 SEP = "."
+max_items_per_fetch = config.get("services._mongo.max_page_size", 500)
 
 
 class _ReferenceProxy(dict):
@@ -278,10 +282,11 @@ class ProjectionHelper(object):
                     doc_only = list(filter(None, data["only"]))
                     doc_only = list({"id"} | set(doc_only)) if doc_only else None
 
-                    for res in projection_func(
-                        doc_type=doc_type, projection=doc_only, ids=ids
-                    ):
-                        self._proxy_manager.update(res)
+                    for ids_chunk in iterutils.chunked_iter(ids, max_items_per_fetch):
+                        for res in projection_func(
+                            doc_type=doc_type, projection=doc_only, ids=ids_chunk
+                        ):
+                            self._proxy_manager.update(res)
 
                 if len(ref_projection) == 1:
                     do_projection(items[0])

apiserver/documentation/api_versions.md

@@ -0,0 +1,19 @@
+### Supported api versions
+
+| Release | ApiVersion |
+|---------|------------|
+| v1.13   | 2.27       |
+| v1.12   | 2.26       |
+| v1.11   | 2.25       |
+| v1.10   | 2.24       |
+| v1.9    | 2.23       |
+| v1.8    | 2.22       |
+| v1.7    | 2.21       |
+| v1.6    | 2.20       |
+| v1.5    | 2.19       |
+| v1.4    | 2.18       |
+| v1.3    | 2.17       |
+| v1.2    | 2.16       |
+| v1.1    | 2.15       |
+| v1.0    | 2.14       |
+| v0.17   | 2.13       |

apiserver/elastic/apply_mappings.py

@@ -21,7 +21,7 @@ def apply_mappings_to_cluster(
         with f.open() as json_data:
             data = json.load(json_data)
             template_name = f.stem
-            res = es.indices.put_template(template_name, body=data)
+            res = es.indices.put_template(name=template_name, body=data)
             return {"mapping": template_name, "result": res}
 
     p = HERE / "mappings"

apiserver/elastic/initialize.py

@@ -85,7 +85,7 @@ def check_elastic_empty() -> bool:
                 es = Elasticsearch(
                     hosts=cluster_conf.get("hosts", None),
                     http_auth=es_factory.get_credentials("events", cluster_conf),
-                    **cluster_conf.get("args", {})
+                    **cluster_conf.get("args", {}),
                 )
                 return not es.indices.get_template(name="events*")
             except exceptions.NotFoundError as ex:
@@ -115,5 +115,7 @@ def init_es_data():
         args = cluster_conf.get("args", {})
         http_auth = es_factory.get_credentials(name)
 
-        res = apply_mappings_to_cluster(hosts_config, name, es_args=args, http_auth=http_auth)
+        res = apply_mappings_to_cluster(
+            hosts_config, name, es_args=args, http_auth=http_auth
+        )
         log.info(res)

apiserver/elastic/mappings/events/events.json

@@ -1,8 +1,8 @@
 {
   "index_patterns": "events-*",
   "settings": {
-    "number_of_shards": 1,
-    "number_of_replicas": 0
+    "number_of_replicas": 0,
+    "number_of_shards": 1
   },
   "mappings": {
     "_source": {

apiserver/elastic/mappings/workers/queue_metrics.json

@@ -1,8 +1,8 @@
 {
   "index_patterns": "queue_metrics_*",
   "settings": {
-    "number_of_shards": 1,
-    "number_of_replicas": 0
+    "number_of_replicas": 0,
+    "number_of_shards": 1
   },
   "mappings": {
     "_source": {
@@ -20,6 +20,9 @@
       },
       "queue_length": {
         "type": "integer"
+      },
+      "company_id": {
+        "type": "keyword"
       }
     }
   }

apiserver/elastic/mappings/workers/worker_stats.json

@@ -1,8 +1,8 @@
 {
   "index_patterns": "worker_stats_*",
   "settings": {
-    "number_of_shards": 1,
-    "number_of_replicas": 0
+    "number_of_replicas": 0,
+    "number_of_shards": 1
   },
   "mappings": {
     "_source": {
@@ -32,6 +32,9 @@
       },
       "task": {
         "type": "keyword"
+      },
+      "company_id": {
+        "type": "keyword"
       }
     }
   }

apiserver/es_factory.py

@@ -67,7 +67,7 @@ class MissingPasswordForElasticUser(Exception):
 
 class ESFactory:
     @classmethod
-    def connect(cls, cluster_name):
+    def connect(cls, cluster_name) -> Elasticsearch:
         """
         Returns the es client for the cluster.
         Connects to the cluster if did not connect previously

apiserver/jobs/async_urls_delete.py

@@ -1,3 +1,5 @@
+import os
+from abc import ABC, ABCMeta, abstractmethod
 from argparse import ArgumentParser
 from collections import defaultdict
 from datetime import datetime, timedelta
@@ -5,38 +7,28 @@ from functools import partial
 from itertools import chain
 from pathlib import Path
 from time import sleep
-from typing import Sequence, Tuple
+from typing import Sequence, Optional, Tuple, Mapping, TypeVar, Hashable, Generic
+from urllib.parse import urlparse
 
+import boto3
 import requests
+from azure.storage.blob import ContainerClient, PartialBatchErrorException
+from boltons.iterutils import bucketize, chunked_iter
 from furl import furl
+from google.cloud import storage as google_storage
 from mongoengine import Q
+from mypy_boto3_s3.service_resource import Bucket as AWSBucket
 
+from apiserver.bll.storage import StorageBLL
 from apiserver.config_repo import config
 from apiserver.database import db
-from apiserver.database.model.url_to_delete import (
-    UrlToDelete,
-    DeletionStatus,
-    StorageType,
-)
+from apiserver.database.model.url_to_delete import UrlToDelete, StorageType, DeletionStatus
 
 log = config.logger(f"JOB-{Path(__file__).name}")
 conf = config.get("services.async_urls_delete")
 max_retries = conf.get("max_retries", 3)
 retry_timeout = timedelta(seconds=conf.get("retry_timeout_sec", 60))
-fileserver_timeout = conf.get("fileserver.timeout_sec", 300)
-UrlPrefix = Tuple[str, str]
-
-
-def validate_fileserver_access(fileserver_host: str) -> str:
-    fileserver_host = fileserver_host or config.get("hosts.fileserver", None)
-    if not fileserver_host:
-        log.error(f"Fileserver host not configured")
-        exit(1)
-
-    res = requests.get(url=fileserver_host)
-    res.raise_for_status()
-
-    return fileserver_host
+storage_bll = StorageBLL()
 
 
 def mark_retry_failed(ids: Sequence[str], reason: str):
@@ -58,112 +50,517 @@ def mark_failed(query: Q, reason: str):
     )
 
 
-def delete_fileserver_urls(
-    urls_query: Q, fileserver_host: str, url_prefixes: Sequence[UrlPrefix]
-):
-    to_delete = list(UrlToDelete.objects(urls_query).limit(10000))
+def scheme_prefix(scheme: str) -> str:
+    return str(furl(scheme=scheme, netloc=""))
+
+
+T = TypeVar("T", bound=Hashable)
+
+
+class Storage(Generic[T], metaclass=ABCMeta):
+    class Client(ABC):
+        @property
+        @abstractmethod
+        def chunk_size(self) -> int:
+            pass
+
+        def get_path(self, url: UrlToDelete) -> str:
+            pass
+
+        def delete_many(
+            self, paths: Sequence[str]
+        ) -> Tuple[Sequence[str], Mapping[str, Sequence[str]]]:
+            pass
+
+    @property
+    @abstractmethod
+    def name(self) -> str:
+        pass
+
+    def group_urls(
+        self, urls: Sequence[UrlToDelete]
+    ) -> Mapping[T, Sequence[UrlToDelete]]:
+        pass
+
+    def get_client(self, base: T, urls: Sequence[UrlToDelete]) -> Client:
+        pass
+
+
+def delete_urls(urls_query: Q, storage: Storage):
+    to_delete = list(UrlToDelete.objects(urls_query).order_by("url").limit(10000))
     if not to_delete:
         return
 
-    def resolve_path(url_: UrlToDelete) -> str:
-        parsed = furl(url_.url)
-        url_host = f"{parsed.scheme}://{parsed.netloc}" if parsed.scheme else None
-        url_path = str(parsed.path)
+    grouped_urls = storage.group_urls(to_delete)
+    for base, urls in grouped_urls.items():
+        if not base:
+            msg = f"Invalid {storage.name} url or missing {storage.name} configuration for account"
+            mark_failed(
+                Q(id__in=[url.id for url in urls]), msg,
+            )
+            log.warning(
+                f"Failed to delete {len(urls)} files from {storage.name} due to: {msg}"
+            )
+            continue
 
-        for host, path_prefix in url_prefixes:
+        try:
+            client = storage.get_client(base, urls)
+        except Exception as ex:
+            failed = [url.id for url in urls]
+            mark_retry_failed(failed, reason=str(ex))
+            log.warning(
+                f"Failed to delete {len(failed)} files from {storage.name} due to: {str(ex)}"
+            )
+            continue
+
+        for chunk in chunked_iter(urls, client.chunk_size):
+            paths = []
+            path_to_id_mapping = defaultdict(list)
+            ids_to_delete = set()
+            for url in chunk:
+                try:
+                    path = client.get_path(url)
+                except Exception as ex:
+                    err = str(ex)
+                    mark_failed(Q(id=url.id), err)
+                    log.warning(f"Error getting path for {url.url}: {err}")
+                    continue
+
+                paths.append(path)
+                path_to_id_mapping[path].append(url.id)
+                ids_to_delete.add(url.id)
+
+            if not paths:
+                continue
+
+            try:
+                deleted_paths, errors = client.delete_many(paths)
+            except Exception as ex:
+                mark_retry_failed([url.id for url in urls], str(ex))
+                log.warning(
+                    f"Error deleting {len(paths)} files from {storage.name}: {str(ex)}"
+                )
+                continue
+
+            failed_ids = set()
+            for reason, err_paths in errors.items():
+                error_ids = set(
+                    chain.from_iterable(
+                        path_to_id_mapping.get(p, []) for p in err_paths
+                    )
+                )
+                mark_retry_failed(list(error_ids), reason)
+                log.warning(
+                    f"Failed to delete {len(error_ids)} files from {storage.name} storage due to: {reason}"
+                )
+                failed_ids.update(error_ids)
+
+            deleted_ids = set(
+                chain.from_iterable(
+                    path_to_id_mapping.get(p, []) for p in deleted_paths
+                )
+            )
+            if deleted_ids:
+                UrlToDelete.objects(id__in=list(deleted_ids)).delete()
+                log.info(
+                    f"{len(deleted_ids)} files deleted from {storage.name} storage"
+                )
+
+            missing_ids = ids_to_delete - deleted_ids - failed_ids
+            if missing_ids:
+                mark_retry_failed(list(missing_ids), "Not succeeded")
+
+
+class FileserverStorage(Storage):
+    class Client(Storage.Client):
+        timeout = conf.get("fileserver.timeout_sec", 300)
+
+        def __init__(self, session: requests.Session, host: str):
+            self.session = session
+            self.delete_url = furl(host).add(path="delete_many").url
+
+        @property
+        def chunk_size(self) -> int:
+            return 10000
+
+        def get_path(self, url: UrlToDelete) -> str:
+            path = url.url.strip("/")
+            if not path:
+                raise ValueError("Empty path")
+
+            return path
+
+        def delete_many(
+            self, paths: Sequence[str]
+        ) -> Tuple[Sequence[str], Mapping[str, Sequence[str]]]:
+            res = self.session.post(
+                url=self.delete_url, json={"files": list(paths)}, timeout=self.timeout
+            )
+            res.raise_for_status()
+            res_data = res.json()
+            return list(res_data.get("deleted", {})), res_data.get("errors", {})
+
+    def __init__(self, company: str, fileserver_host: str = None):
+        fileserver_host = fileserver_host or config.get("hosts.fileserver", None)
+        self.host = fileserver_host.rstrip("/")
+        if not self.host:
+            log.warning(f"Fileserver host not configured")
+
+        def _parse_url_prefix(prefix) -> Tuple[str, str]:
+            url = furl(prefix)
+            host = f"{url.scheme}://{url.netloc}" if url.scheme else None
+            return host, str(url.path).rstrip("/")
+
+        url_prefixes = [
+            _parse_url_prefix(p) for p in conf.get("fileserver.url_prefixes", [])
+        ]
+        if not any(self.host == host for host, _ in url_prefixes):
+            url_prefixes.append((self.host, ""))
+        self.url_prefixes = url_prefixes
+
+        self.company = company
+
+    # @classmethod
+    # def validate_fileserver_access(cls, fileserver_host: str):
+    #     res = requests.get(
+    #         url=fileserver_host
+    #     )
+    #     res.raise_for_status()
+
+    @property
+    def name(self) -> str:
+        return "Fileserver"
+
+    def _resolve_base_url(self, url: UrlToDelete) -> Optional[str]:
+        """
+        For the url return the base_url containing schema, optional host and bucket name
+        """
+        if not url.url:
+            return None
+
+        try:
+            parsed = furl(url.url)
+            url_host = f"{parsed.scheme}://{parsed.netloc}" if parsed.scheme else None
+            url_path = str(parsed.path)
+        except Exception:
+            return None
+
+        for host, path_prefix in self.url_prefixes:
             if host and url_host != host:
                 continue
             if path_prefix and not url_path.startswith(path_prefix + "/"):
                 continue
-            return url_path[len(path_prefix or ""):]
+            url.url = url_path[len(path_prefix or "") :]
+            return self.host
 
-        raise ValueError("could not map path")
+    def group_urls(
+        self, urls: Sequence[UrlToDelete]
+    ) -> Mapping[str, Sequence[UrlToDelete]]:
+        return bucketize(urls, key=self._resolve_base_url)
 
-    paths = set()
-    path_to_id_mapping = defaultdict(list)
-    for url in to_delete:
-        try:
-            path = resolve_path(url)
-            path = path.strip("/")
-            if not path:
-                raise ValueError("Empty path")
-        except Exception as ex:
-            err = str(ex)
-            log.warn(f"Error getting path for {url.url}: {err}")
-            mark_failed(Q(id=url.id), err)
-            continue
-
-        paths.add(path)
-        path_to_id_mapping[path].append(url.id)
-
-    if not paths:
-        return
-
-    ids_to_delete = set(chain.from_iterable(path_to_id_mapping.values()))
-    try:
-        res = requests.post(
-            url=furl(fileserver_host).add(path="delete_many").url,
-            json={"files": list(paths)},
-            timeout=fileserver_timeout,
-        )
+    def get_client(self, base: str, urls: Sequence[UrlToDelete]) -> Client:
+        host = base
+        session = requests.session()
+        res = session.get(url=host, timeout=self.Client.timeout)
         res.raise_for_status()
-    except Exception as ex:
-        err = str(ex)
-        log.warn(f"Error deleting {len(paths)} files from fileserver: {err}")
-        mark_retry_failed(list(ids_to_delete), err)
-        return
 
-    res_data = res.json()
-    deleted_ids = set(
-        chain.from_iterable(
-            path_to_id_mapping.get(path, [])
-            for path in list(res_data.get("deleted", {}))
+        return self.Client(session, host)
+
+
+class AzureStorage(Storage):
+    class Client(Storage.Client):
+        def __init__(self, container: ContainerClient):
+            self.container = container
+
+        @property
+        def chunk_size(self) -> int:
+            return 256
+
+        def get_path(self, url: UrlToDelete) -> str:
+            parsed = furl(url.url)
+            if (
+                not parsed.path
+                or not parsed.path.segments
+                or len(parsed.path.segments) <= 1
+            ):
+                raise ValueError("No path found following container name")
+
+            return os.path.join(*parsed.path.segments[1:])
+
+        @staticmethod
+        def _path_from_request_url(request_url: str) -> str:
+            try:
+                return furl(request_url).path.segments[-1]
+            except Exception:
+                return request_url
+
+        def delete_many(
+            self, paths: Sequence[str]
+        ) -> Tuple[Sequence[str], Mapping[str, Sequence[str]]]:
+            try:
+                res = self.container.delete_blobs(*paths)
+            except PartialBatchErrorException as pex:
+                deleted = []
+                errors = defaultdict(list)
+                for part in pex.parts:
+                    if 300 >= part.status_code >= 200:
+                        deleted.append(self._path_from_request_url(part.request.url))
+                    else:
+                        errors[part.reason].append(
+                            self._path_from_request_url(part.request.url)
+                        )
+                return deleted, errors
+
+            return [self._path_from_request_url(part.request.url) for part in res], {}
+
+    def __init__(self, company: str):
+        self.configs = storage_bll.get_azure_settings_for_company(company)
+        self.scheme = "azure"
+
+    @property
+    def name(self) -> str:
+        return "Azure"
+
+    def _resolve_base_url(self, url: UrlToDelete) -> Optional[Tuple]:
+        """
+        For the url return the base_url containing schema, optional host and bucket name
+        """
+        try:
+            parsed = urlparse(url.url)
+            if parsed.scheme != self.scheme:
+                return None
+
+            azure_conf = self.configs.get_config_by_uri(url.url)
+            if azure_conf is None:
+                return None
+
+            account_url = parsed.netloc
+            return account_url, azure_conf.container_name
+        except Exception as ex:
+            log.warning(f"Error resolving base url for {url.url}: " + str(ex))
+            return None
+
+    def group_urls(
+        self, urls: Sequence[UrlToDelete]
+    ) -> Mapping[Tuple, Sequence[UrlToDelete]]:
+        return bucketize(urls, key=self._resolve_base_url)
+
+    def get_client(self, base: Tuple, urls: Sequence[UrlToDelete]) -> Client:
+        account_url, container_name = base
+        sample_url = urls[0].url
+        cfg = self.configs.get_config_by_uri(sample_url)
+        if not cfg or not cfg.account_name or not cfg.account_key:
+            raise ValueError(
+                f"Missing account name or key for Azure Blob Storage "
+                f"account: {account_url}, container: {container_name}"
+            )
+
+        return self.Client(
+            ContainerClient(
+                account_url=account_url,
+                container_name=cfg.container_name,
+                credential={
+                    "account_name": cfg.account_name,
+                    "account_key": cfg.account_key,
+                },
+            )
         )
-    )
-    if deleted_ids:
-        UrlToDelete.objects(id__in=list(deleted_ids)).delete()
-        log.info(f"{len(deleted_ids)} files deleted from the fileserver")
 
-    failed_ids = set()
-    for err, error_ids in res_data.get("errors", {}).items():
-        error_ids = list(
-            chain.from_iterable(path_to_id_mapping.get(path, []) for path in error_ids)
+
+class AWSStorage(Storage):
+    class Client(Storage.Client):
+        def __init__(self, base_url: str, container: AWSBucket):
+            self.container = container
+            self.base_url = base_url
+
+        @property
+        def chunk_size(self) -> int:
+            return 1000
+
+        def get_path(self, url: UrlToDelete) -> str:
+            """ Normalize remote path. Remove any prefix that is already handled by the container """
+            path = url.url
+            if path.startswith(self.base_url):
+                path = path[len(self.base_url) :]
+            path = path.lstrip("/")
+            return path
+
+        @staticmethod
+        def _path_from_request_url(request_url: str) -> str:
+            try:
+                return furl(request_url).path.segments[-1]
+            except Exception:
+                return request_url
+
+        def delete_many(
+            self, paths: Sequence[str]
+        ) -> Tuple[Sequence[str], Mapping[str, Sequence[str]]]:
+            res = self.container.delete_objects(
+                Delete={"Objects": [{"Key": p} for p in paths]}
+            )
+            errors = defaultdict(list)
+            for err in res.get("Errors", []):
+                msg = err.get("Message", "")
+                errors[msg].append(err.get("Key"))
+
+            return [d.get("Key") for d in res.get("Deleted", [])], errors
+
+    def __init__(self, company: str):
+        self.configs = storage_bll.get_aws_settings_for_company(company)
+        self.scheme = "s3"
+
+    @property
+    def name(self) -> str:
+        return "AWS"
+
+    def _resolve_base_url(self, url: UrlToDelete) -> Optional[str]:
+        """
+        For the url return the base_url containing schema, optional host and bucket name
+        """
+        try:
+            parsed = urlparse(url.url)
+            if parsed.scheme != self.scheme:
+                return None
+
+            s3_conf = self.configs.get_config_by_uri(url.url)
+            if s3_conf is None:
+                return None
+
+            s3_bucket = s3_conf.bucket
+            if not s3_bucket:
+                parts = Path(parsed.path.strip("/")).parts
+                if parts:
+                    s3_bucket = parts[0]
+            return "/".join(filter(None, ("s3:/", s3_conf.host, s3_bucket)))
+        except Exception as ex:
+            log.warning(f"Error resolving base url for {url.url}: " + str(ex))
+            return None
+
+    def group_urls(
+        self, urls: Sequence[UrlToDelete]
+    ) -> Mapping[str, Sequence[UrlToDelete]]:
+        return bucketize(urls, key=self._resolve_base_url)
+
+    def get_client(self, base: str, urls: Sequence[UrlToDelete]) -> Client:
+        sample_url = urls[0].url
+        cfg = self.configs.get_config_by_uri(sample_url)
+        boto_kwargs = {
+            "endpoint_url": (("https://" if cfg.secure else "http://") + cfg.host)
+            if cfg.host
+            else None,
+            "use_ssl": cfg.secure,
+            "verify": cfg.verify,
+        }
+        name = base[len(scheme_prefix(self.scheme)) :]
+        bucket_name = name[len(cfg.host) + 1 :] if cfg.host else name
+        if not cfg.use_credentials_chain:
+            if not cfg.key or not cfg.secret:
+                raise ValueError(
+                    f"Missing key or secret for AWS S3 host: {cfg.host}, bucket: {str(bucket_name)}"
+                )
+
+            boto_kwargs["aws_access_key_id"] = cfg.key
+            boto_kwargs["aws_secret_access_key"] = cfg.secret
+            if cfg.token:
+                boto_kwargs["aws_session_token"] = cfg.token
+
+        return self.Client(
+            base, boto3.resource("s3", **boto_kwargs).Bucket(bucket_name)
         )
-        mark_retry_failed(error_ids, err)
-        log.warning(
-            f"Failed to delete {len(error_ids)} files from the fileserver due to: {err}"
+
+
+class GoogleCloudStorage(Storage):
+    class Client(Storage.Client):
+        def __init__(self, base_url: str, container: google_storage.Bucket):
+            self.container = container
+            self.base_url = base_url
+
+        @property
+        def chunk_size(self) -> int:
+            return 100
+
+        def get_path(self, url: UrlToDelete) -> str:
+            """ Normalize remote path. Remove any prefix that is already handled by the container """
+            path = url.url
+            if path.startswith(self.base_url):
+                path = path[len(self.base_url) :]
+            path = path.lstrip("/")
+            return path
+
+        def delete_many(
+            self, paths: Sequence[str]
+        ) -> Tuple[Sequence[str], Mapping[str, Sequence[str]]]:
+            not_found = set()
+
+            def error_callback(blob: google_storage.Blob):
+                not_found.add(blob.name)
+
+            self.container.delete_blobs(
+                [self.container.blob(p) for p in paths], on_error=error_callback,
+            )
+            errors = {"Not found": list(not_found)} if not_found else {}
+            return list(set(paths) - not_found), errors
+
+    def __init__(self, company: str):
+        self.configs = storage_bll.get_gs_settings_for_company(company)
+        self.scheme = "gs"
+
+    @property
+    def name(self) -> str:
+        return "Google Storage"
+
+    def _resolve_base_url(self, url: UrlToDelete) -> Optional[str]:
+        """
+        For the url return the base_url containing schema, optional host and bucket name
+        """
+        try:
+            parsed = urlparse(url.url)
+            if parsed.scheme != self.scheme:
+                return None
+
+            gs_conf = self.configs.get_config_by_uri(url.url)
+            if gs_conf is None:
+                return None
+
+            return str(furl(scheme=parsed.scheme, netloc=gs_conf.bucket))
+        except Exception as ex:
+            log.warning(f"Error resolving base url for {url.url}: " + str(ex))
+            return None
+
+    def group_urls(
+        self, urls: Sequence[UrlToDelete]
+    ) -> Mapping[str, Sequence[UrlToDelete]]:
+        return bucketize(urls, key=self._resolve_base_url)
+
+    def get_client(self, base: str, urls: Sequence[UrlToDelete]) -> Client:
+        sample_url = urls[0].url
+        cfg = self.configs.get_config_by_uri(sample_url)
+        if cfg.credentials_json:
+            from google.oauth2 import service_account
+
+            credentials = service_account.Credentials.from_service_account_file(
+                cfg.credentials_json
+            )
+        else:
+            credentials = None
+
+        bucket_name = base[len(scheme_prefix(self.scheme)) :]
+        return self.Client(
+            base,
+            google_storage.Client(project=cfg.project, credentials=credentials).bucket(
+                bucket_name
+            ),
         )
-        failed_ids.update(error_ids)
-
-    missing_ids = ids_to_delete - deleted_ids - failed_ids
-    if missing_ids:
-        mark_retry_failed(list(missing_ids), "Not succeeded")
-
-
-def _get_fileserver_url_prefixes(fileserver_host: str) -> Sequence[UrlPrefix]:
-    def _parse_url_prefix(prefix) -> UrlPrefix:
-        url = furl(prefix)
-        host = f"{url.scheme}://{url.netloc}" if url.scheme else None
-        return host, str(url.path).rstrip("/")
-
-    url_prefixes = [
-        _parse_url_prefix(p) for p in conf.get("fileserver.url_prefixes", [])
-    ]
-    if not any(fileserver_host == host for host, _ in url_prefixes):
-        url_prefixes.append((fileserver_host, ""))
-
-    return url_prefixes
 
 
 def run_delete_loop(fileserver_host: str):
-    fileserver_host = validate_fileserver_access(fileserver_host)
-
-    storage_delete_funcs = {
+    storage_helpers = {
         StorageType.fileserver: partial(
-            delete_fileserver_urls,
-            fileserver_host=fileserver_host,
-            url_prefixes=_get_fileserver_url_prefixes(fileserver_host),
+            FileserverStorage, fileserver_host=fileserver_host
         ),
+        StorageType.s3: AWSStorage,
+        StorageType.azure: AzureStorage,
+        StorageType.gs: GoogleCloudStorage,
     }
     while True:
         now = datetime.utcnow()
@@ -177,7 +574,7 @@ def run_delete_loop(fileserver_host: str):
         )
 
         url_to_delete: UrlToDelete = UrlToDelete.objects(
-            urls_query & Q(storage_type__in=list(storage_delete_funcs))
+            urls_query & Q(storage_type__in=list(storage_helpers))
         ).order_by("retry_count").limit(1).first()
         if not url_to_delete:
             sleep(10)
@@ -192,7 +589,10 @@ def run_delete_loop(fileserver_host: str):
         company_storage_urls_query = urls_query & Q(
             company=company, storage_type=storage_type,
         )
-        storage_delete_funcs[storage_type](urls_query=company_storage_urls_query)
+        delete_urls(
+            urls_query=company_storage_urls_query,
+            storage=storage_helpers[storage_type](company=company),
+        )
 
 
 def main():

apiserver/mongo/initialize/migration.py

@@ -18,9 +18,15 @@ _migration_dir = _parent_dir / _migrations
 
 
 def check_mongo_empty() -> bool:
-    return not all(
-        get_db(alias).collection_names() for alias in utils.get_options(Database)
-    )
+    for alias in utils.get_options(Database):
+        collection_names = get_db(alias).list_collection_names()
+        if collection_names and any(
+            name in collection_names
+            for name in ["company", "user", "versions"]
+        ):
+            return False
+
+    return True
 
 
 def get_last_server_version() -> Version:

apiserver/mongo/initialize/pre_populate.py

@@ -960,13 +960,17 @@ class PrePopulate:
             return tasks
 
     @classmethod
-    def _import_events(cls, f: IO[bytes], company_id: str, _, task_id: str):
+    def _import_events(cls, f: IO[bytes], company_id: str, user_id: str, task_id: str):
         print(f"Writing events for task {task_id} into database")
         for events_chunk in chunked_iter(cls.json_lines(f), 1000):
             events = [json.loads(item) for item in events_chunk]
             for ev in events:
                 ev["task"] = task_id
                 ev["company_id"] = company_id
+                ev["allow_locked"] = True
             cls.event_bll.add_events(
-                company_id, events=events, worker="", allow_locked=True
+                company_id=company_id,
+                user_id=user_id,
+                events=events,
+                worker="",
             )

apiserver/mongo/migrations/1_9_0.py

@@ -0,0 +1,17 @@
+import logging as log
+
+from pymongo.collection import Collection
+from pymongo.database import Database
+from pymongo.errors import OperationFailure
+
+
+def migrate_backend(db: Database):
+    """
+    Drop task text index so that the new one including reports field is created
+    """
+    tasks: Collection = db["task"]
+    try:
+        tasks.drop_index("backend-db.task.main_text_index")
+    except OperationFailure as ex:
+        log.warning(f"Could not delete task text index due to: {str(ex)}")
+        pass

apiserver/redis_manager.py

@@ -2,7 +2,7 @@ from os import getenv
 
 from boltons.iterutils import first
 from redis import StrictRedis
-from rediscluster import RedisCluster
+from redis.cluster import RedisCluster
 
 from apiserver.apierrors.errors.server_error import ConfigError, GeneralError
 from apiserver.config_repo import config
@@ -83,7 +83,7 @@ class RedisManager(object):
     def host(self, alias):
         r = self.connection(alias)
         if isinstance(r, RedisCluster):
-            connections = first(r.connection_pool._available_connections.values())
+            connections = r.get_default_node().redis_connection.connection_pool._available_connections
         else:
             connections = r.connection_pool._available_connections
 

apiserver/requirements.txt

@@ -1,33 +1,36 @@
-attrs>=22.1.0
+attrs>=22.1.0,<23
+azure-storage-blob>=12.13.1
 bcrypt>=3.1.4
 boltons>=19.1.0
-boto3==1.14.13
+boto3>=1.26
+boto3-stubs[s3]>=1.26
+clearml>=1.10.3
 dpath>=1.4.2,<2.0
-elasticsearch==7.13.3
+elasticsearch==7.17.9
 fastjsonschema>=2.8
 flask-compress>=1.4.0
 flask-cors>=3.0.5
-flask>=0.12.2
-funcsigs==1.0.2
+flask>=2.3.2
 furl>=2.0.0
-gunicorn>=19.7.1
-humanfriendly==4.18
-jinja2==2.11.3
+google-cloud-storage>=2.8.0
+gunicorn>=20.1.0
+humanfriendly>=4.17
+jinja2
 jsonmodels>=2.3
 jsonschema>=2.6.0
 luqum>=0.10.0
-mongoengine==0.23.1
+mongoengine==0.27.0
 nested_dict>=1.61
 packaging==20.3
 psutil>=5.6.5
-pyhocon>=0.3.35
+pyhocon>=0.3.35r
 pyjwt>=2.4.0
-pymongo[srv]==3.12.0
+pymongo==4.4.0
 python-rapidjson>=0.6.3
-redis==3.5.3
-redis-py-cluster>=2.1.3
+redis>=4.5.4,<5
 requests>=2.13.0
 semantic_version>=2.8.3,<3
+setuptools>=65.5.1
 six
-tqdm
-validators>=0.12.4
+validators>=0.12.4
+urllib3>=1.26.18

apiserver/schema/services/_common.conf

@@ -1,3 +1,43 @@
+field_filter {
+    type: object
+    description: Filter on a field that includes combination of 'any' or 'all' included and excluded terms
+    properties {
+        any {
+            type: object
+            description: All the terms in 'any' condition are combined with 'or' operation
+            properties {
+                "include" {
+                    type: array
+                    items {type: string}
+                }
+                exclude {
+                    type: array
+                    items {type: string}
+                }
+            }
+        }
+        all {
+            type: object
+            description: All the terms in 'all' condition are combined with 'and' operation
+            properties {
+                "include" {
+                    type: array
+                    items {type: string}
+                }
+                exclude {
+                    type: array
+                    items {type: string}
+                }
+            }
+        }
+        op {
+            type: string
+            description: The operation between 'any' and 'all' parts of the filter if both are provided
+            default: and
+            enum: [and, or]
+        }
+    }
+}
 metadata_item {
     type: object
     properties {
@@ -15,6 +55,35 @@ metadata_item {
         }
     }
 }
+task_status_enum {
+    type: string
+    enum: [
+        created
+        queued
+        in_progress
+        stopped
+        published
+        publishing
+        closed
+        failed
+        completed
+        unknown
+    ]
+}
+multi_field_pattern_data {
+    type: object
+    properties {
+        pattern {
+            description: "Pattern string (regex)"
+            type: string
+        }
+        fields {
+            description: "List of field names"
+            type: array
+            items { type: string }
+        }
+    }
+}
 credentials {
     type: object
     properties {

apiserver/schema/services/_events_common.conf

@@ -0,0 +1,141 @@
+scalar_key_enum {
+    type: string
+    enum: [
+        iter
+        timestamp
+        iso_time
+    ]
+}
+metric_variants {
+    type: object
+    properties {
+        metric {
+            description: The metric name
+            type: string
+        }
+        variants {
+            type: array
+            description: The names of the metric variants
+            items {type: string}
+        }
+    }
+}
+debug_images_response_task_metrics {
+    type: object
+    properties {
+        task {
+            type: string
+            description: Task ID
+        }
+        iterations {
+            type: array
+            items {
+                type: object
+                properties {
+                    iter {
+                        type: integer
+                        description: Iteration number
+                    }
+                    events {
+                        type: array
+                        items {
+                            type: object
+                            description: Debug image event
+                        }
+                    }
+                }
+            }
+        }
+    }
+}
+debug_images_response {
+    type: object
+    properties {
+        scroll_id {
+            type: string
+            description: "Scroll ID for getting more results"
+        }
+        metrics {
+            type: array
+            description: "Debug image events grouped by tasks and iterations"
+            items {"$ref": "#/definitions/debug_images_response_task_metrics"}
+        }
+    }
+}
+plots_response_task_metrics {
+    type: object
+    properties {
+        task {
+            type: string
+            description: Task ID
+        }
+        iterations {
+            type: array
+            items {
+                type: object
+                properties {
+                    iter {
+                        type: integer
+                        description: Iteration number
+                    }
+                    events {
+                        type: array
+                        items {
+                            type: object
+                            description: Plot event
+                        }
+                    }
+                }
+            }
+        }
+    }
+}
+plots_response {
+    type: object
+    properties {
+        scroll_id {
+            type: string
+            description: "Scroll ID for getting more results"
+        }
+        metrics {
+            type: array
+            description: "Plot events grouped by tasks and iterations"
+            items {"$ref": "#/definitions/plots_response_task_metrics"}
+        }
+    }
+}
+single_value_task_metrics {
+    type: object
+    properties {
+        task {
+            type: string
+            description: Task ID
+        }
+        task_name {
+            type: string
+            description: Task name
+        }
+        values {
+            type: array
+            items {
+                type: object
+                properties {
+                    metric { type: string }
+                    variant { type: string}
+                    value { type: number }
+                    timestamp { type: number }
+                }
+            }
+        }
+    }
+}
+single_value_metrics_response {
+    type: object
+    properties {
+        tasks {
+            description: Single value metrics grouped by task
+            type: array
+            items {"$ref": "#/definitions/single_value_task_metrics"}
+        }
+    }
+}

apiserver/schema/services/_tasks_common.conf

@@ -0,0 +1,506 @@
+include "_common.conf"
+task_type_enum {
+    type: string
+    enum: [
+        training
+        testing
+        inference
+        data_processing
+        application
+        monitor
+        controller
+        optimizer
+        service
+        qc
+        custom
+    ]
+}
+script {
+    type: object
+    properties {
+        binary {
+            description: "Binary to use when running the script"
+            type: string
+            default: python
+        }
+        repository {
+            description: "Name of the repository where the script is located"
+            type: string
+        }
+        tag {
+            description: "Repository tag"
+            type: string
+        }
+        branch  {
+            description: "Repository branch id If not provided and tag not provided, default repository branch is used."
+            type: string
+        }
+        version_num {
+            description: "Version (changeset) number. Optional (default is head version) Unused if tag is provided."
+            type: string
+        }
+        entry_point {
+            description: "Path to execute within the repository"
+            type: string
+        }
+        working_dir {
+            description: "Path to the folder from which to run the script Default - root folder of repository"
+            type: string
+        }
+        requirements {
+            description: "A JSON object containing requirements strings by key"
+            type: object
+        }
+        diff {
+            description: "Uncommitted changes found in the repository when task was run"
+            type: string
+        }
+    }
+}
+model_type_enum {
+    type: string
+    enum: ["input", "output"]
+}
+task_model_item {
+    type: object
+    required: [ name, model]
+    properties {
+        name {
+            description: "The task model name"
+            type: string
+        }
+        model {
+            description: "The model ID"
+            type: string
+        }
+    }
+}
+output {
+        type: object
+        properties {
+            destination {
+                description: "Storage id. This is where output files will be stored."
+                type: string
+            }
+            model {
+                description: "Model id."
+                type: string
+            }
+            result {
+                description: "Task result. Values: 'success', 'failure'"
+                type: string
+            }
+            error {
+                description: "Last error text"
+                type: string
+            }
+        }
+    }
+task_execution_progress_enum {
+    type: string
+    enum: [
+        unknown
+        running
+        stopping
+        stopped
+    ]
+}
+artifact_type_data {
+    type: object
+    properties {
+        preview {
+            description: "Description or textual data"
+            type: string
+        }
+        content_type {
+            description: "System defined raw data content type"
+            type: string
+        }
+        data_hash {
+            description: "Hash of raw data, without any headers or descriptive parts"
+            type: string
+        }
+    }
+}
+artifact_mode_enum {
+    type: string
+    enum: [
+        input
+        output
+    ]
+    default: output
+}
+artifact {
+    type: object
+    required: [key, type]
+    properties {
+        key {
+            description: "Entry key"
+            type: string
+        }
+        type {
+            description: "System defined type"
+            type: string
+        }
+        mode {
+            description: "System defined input/output indication"
+            "$ref": "#/definitions/artifact_mode_enum"
+        }
+        uri {
+            description: "Raw data location"
+            type: string
+        }
+        content_size {
+            description: "Raw data length in bytes"
+            type: integer
+        }
+        hash {
+            description: "Hash of entire raw data"
+            type: string
+        }
+        timestamp {
+            description: "Epoch time when artifact was created"
+            type: integer
+        }
+        type_data {
+            description: "Additional fields defined by the system"
+            "$ref": "#/definitions/artifact_type_data"
+        }
+        display_data {
+            description: "User-defined list of key/value pairs, sorted"
+            type: array
+            items {
+                type: array
+                items {
+                    type: string  # can also be a number... TODO: upgrade the generator
+                }
+            }
+        }
+    }
+}
+artifact_id {
+    type: object
+    required: [key]
+    properties {
+        key {
+            description: "Entry key"
+            type: string
+        }
+        mode {
+            description: "System defined input/output indication"
+            "$ref": "#/definitions/artifact_mode_enum"
+        }
+    }
+}
+task_models {
+    type: object
+    properties {
+        input {
+            description: "The list of task input models"
+            type: array
+            items {"$ref": "#/definitions/task_model_item"}
+
+        }
+        output {
+            description: "The list of task output models"
+            type: array
+            items {"$ref": "#/definitions/task_model_item"}
+        }
+    }
+}
+execution {
+    type: object
+    properties {
+        queue {
+            description: "Queue ID where task was queued."
+            type: string
+        }
+        parameters {
+            description: "Json object containing the Task parameters"
+            type: object
+            additionalProperties: true
+        }
+        model {
+            description: "Execution input model ID Not applicable for Register (Import) tasks"
+            type: string
+        }
+        model_desc {
+            description: "Json object representing the Model descriptors"
+            type: object
+            additionalProperties: true
+        }
+        model_labels {
+            description: """Json object representing the ids of the labels in the model.
+            The keys are the layers' names and the values are the IDs.
+            Not applicable for Register (Import) tasks.
+            Mandatory for Training tasks"""
+            type: object
+            additionalProperties: { type: integer }
+        }
+        framework {
+            description: """Framework related to the task. Case insensitive. Mandatory for Training tasks. """
+            type: string
+        }
+        docker_cmd {
+            description: "Command for running docker script for the execution of the task"
+            type: string
+        }
+        artifacts {
+            description: "Task artifacts"
+            type: array
+            items { "$ref": "#/definitions/artifact" }
+        }
+    }
+}
+last_metrics_event {
+    type: object
+    properties {
+        metric {
+            description: "Metric name"
+            type: string
+        }
+        variant {
+            description: "Variant name"
+            type: string
+        }
+        value {
+            description: "Last value reported"
+            type: number
+        }
+        min_value {
+            description: "Minimum value reported"
+            type: number
+        }
+        min_value_iteration {
+            description: "The iteration at which the minimum value was reported"
+            type: integer
+        }
+        max_value {
+            description: "Maximum value reported"
+            type: number
+        }
+        max_value_iteration {
+            description: "The iteration at which the maximum value was reported"
+            type: integer
+        }
+    }
+}
+last_metrics_variants {
+    type: object
+    description: "Last metric events, one for each variant hash"
+    additionalProperties {
+        "$ref": "#/definitions/last_metrics_event"
+    }
+}
+params_item {
+    type: object
+    properties {
+        section {
+            description: "Section that the parameter belongs to"
+            type: string
+        }
+        name {
+            description: "Name of the parameter. The combination of section and name should be unique"
+            type: string
+        }
+        value {
+            description: "Value of the parameter"
+            type: string
+        }
+        type {
+            description: "Type of the parameter. Optional"
+            type: string
+        }
+        description {
+            description: "The parameter description. Optional"
+            type: string
+        }
+    }
+}
+configuration_item {
+    type: object
+    properties {
+        name {
+            description: "Name of the parameter. Should be unique"
+            type: string
+        }
+        value {
+            description: "Value of the parameter"
+            type: string
+        }
+        type {
+            description: "Type of the parameter. Optional"
+            type: string
+        }
+        description {
+            description: "The parameter description. Optional"
+            type: string
+        }
+    }
+}
+section_params {
+    description: "Task section params"
+    type: object
+    additionalProperties {
+        "$ref": "#/definitions/params_item"
+    }
+}
+task {
+    type: object
+    properties {
+        id {
+            description: "Task id"
+            type: string
+        }
+        name {
+            description: "Task Name"
+            type: string
+        }
+        user {
+            description: "Associated user id"
+            type: string
+        }
+        company {
+            description: "Company ID"
+            type: string
+        }
+        type {
+            description: "Type of task. Values: 'training', 'testing'"
+            "$ref": "#/definitions/task_type_enum"
+        }
+        status {
+            description: ""
+            "$ref": "#/definitions/task_status_enum"
+        }
+        comment {
+            description: "Free text comment"
+            type: string
+        }
+        created {
+            description: "Task creation time (UTC) "
+            type: string
+            format: "date-time"
+        }
+        started {
+            description: "Task start time (UTC)"
+            type: string
+            format: "date-time"
+        }
+        completed {
+            description: "Task end time (UTC)"
+            type: string
+            format: "date-time"
+        }
+        active_duration {
+            description: "Task duration time (seconds)"
+            type: integer
+        }
+        parent {
+            description: "Parent task id"
+            type: string
+        }
+        project {
+            description: "Project ID of the project to which this task is assigned"
+            type: string
+        }
+        output {
+            description: "Task output params"
+            "$ref": "#/definitions/output"
+        }
+        execution {
+            description: "Task execution params"
+            "$ref": "#/definitions/execution"
+        }
+        container {
+            description: "Docker container parameters"
+            type: object
+            additionalProperties { type: string }
+        }
+        models {
+            description: "Task models"
+            "$ref": "#/definitions/task_models"
+        }
+        // TODO: will be removed
+        script {
+            description: "Script info"
+            "$ref": "#/definitions/script"
+        }
+        tags {
+            description: "User-defined tags list"
+            type: array
+            items { type: string }
+        }
+        system_tags {
+            description: "System tags list. This field is reserved for system use, please don't use it."
+            type: array
+            items { type: string }
+        }
+        status_changed {
+            description: "Last status change time"
+            type: string
+            format: "date-time"
+        }
+        status_message {
+            description: "free text string representing info about the status"
+            type: string
+        }
+        status_reason {
+            description: "Reason for last status change"
+            type: string
+        }
+        published {
+            description: "Task publish time"
+            type: string
+            format: "date-time"
+        }
+        last_worker {
+            description: "ID of last worker that handled the task"
+            type: string
+        }
+        last_worker_report {
+            description: "Last time a worker reported while working on this task"
+            type: string
+            format: "date-time"
+        }
+        last_update {
+            description: "Last time this task was created, edited, changed or events for this task were reported"
+            type: string
+            format: "date-time"
+        }
+        last_change {
+            description: "Last time any update was done to the task"
+            type: string
+            format: "date-time"
+        }
+        last_iteration {
+            description: "Last iteration reported for this task"
+            type: integer
+        }
+        last_metrics {
+            description: "Last metric variants (hash to events), one for each metric hash"
+            type: object
+            additionalProperties {
+                "$ref": "#/definitions/last_metrics_variants"
+            }
+        }
+        hyperparams {
+            description: "Task hyper params per section"
+            type: object
+            additionalProperties {
+                "$ref": "#/definitions/section_params"
+            }
+        }
+        configuration {
+            description: "Task configuration params"
+            type: object
+            additionalProperties {
+                "$ref": "#/definitions/configuration_item"
+            }
+        }
+        runtime {
+            description: "Task runtime mapping"
+            type: object
+            additionalProperties: true
+        }
+    }
+}

apiserver/schema/services/events.conf

@@ -1,17 +1,6 @@
 _description : "Provides an API for running tasks to report events collected by the system."
 _definitions {
-    metric_variants {
-        type: object
-        metric {
-            description: The metric name
-            type: string
-        }
-        variants {
-            type: array
-            description: The names of the metric variants
-            items {type: string}
-        }
-    }
+    include "_events_common.conf"
     metrics_scalar_event {
         description: "Used for reporting scalar metrics during training task"
         type: object
@@ -22,8 +11,8 @@ _definitions {
                 type: number
             }
             type {
-                description: "training_stats_vector"
-                const: "training_stats_scalar"
+                description: "'training_stats_scalar'"
+                type: string
             }
             task {
                 description: "Task ID (required)"
@@ -57,8 +46,8 @@ _definitions {
                 type: number
             }
             type {
-                description: "training_stats_vector"
-                const: "training_stats_vector"
+                description: "'training_stats_vector'"
+                type: string
             }
             task {
                 description: "Task ID (required)"
@@ -93,8 +82,8 @@ _definitions {
                 type: number
             }
             type {
-                description: ""
-                const: "training_debug_image"
+                description: "'training_debug_image'"
+                type: string
             }
             task {
                 description: "Task ID (required)"
@@ -134,7 +123,7 @@ _definitions {
             }
             type {
                 description: "'plot'"
-                const: "plot"
+                type: string
             }
             task {
                 description: "Task ID (required)"
@@ -164,14 +153,6 @@ _definitions {
             }
         }
     }
-    scalar_key_enum {
-        type: string
-        enum: [
-            iter
-            timestamp
-            iso_time
-        ]
-    }
     log_level_enum {
         type: string
         enum: [
@@ -240,7 +221,7 @@ _definitions {
             }
             type {
                 description: "'log'"
-                const: "log"
+                type: string
             }
             task {
                 description: "Task ID (required)"
@@ -260,90 +241,6 @@ _definitions {
             }
         }
     }
-    debug_images_response_task_metrics {
-        type: object
-        properties {
-            task {
-                type: string
-                description: Task ID
-            }
-            iterations {
-                type: array
-                items {
-                    type: object
-                    properties {
-                        iter {
-                            type: integer
-                            description: Iteration number
-                        }
-                        events {
-                            type: array
-                            items {
-                                type: object
-                                description: Debug image event
-                            }
-                        }
-                    }
-                }
-            }
-        }
-    }
-    debug_images_response {
-        type: object
-        properties {
-            scroll_id {
-                type: string
-                description: "Scroll ID for getting more results"
-            }
-            metrics {
-                type: array
-                description: "Debug image events grouped by tasks and iterations"
-                items {"$ref": "#/definitions/debug_images_response_task_metrics"}
-            }
-        }
-    }
-    plots_response_task_metrics {
-        type: object
-        properties {
-            task {
-                type: string
-                description: Task ID
-            }
-            iterations {
-                type: array
-                items {
-                    type: object
-                    properties {
-                        iter {
-                            type: integer
-                            description: Iteration number
-                        }
-                        events {
-                            type: array
-                            items {
-                                type: object
-                                description: Plot event
-                            }
-                        }
-                    }
-                }
-            }
-        }
-    }
-    plots_response {
-        type: object
-        properties {
-            scroll_id {
-                type: string
-                description: "Scroll ID for getting more results"
-            }
-            metrics {
-                type: array
-                description: "Plot events grouped by tasks and iterations"
-                items {"$ref": "#/definitions/plots_response_task_metrics"}
-            }
-        }
-    }
     debug_image_sample_response {
         type: object
         properties {
@@ -547,7 +444,7 @@ debug_images {
                 }
                 total {
                     type: number
-                    description: "Total number of results available for this query"
+                    description: "Total number of results available for this query. In case there are more than 10000 results it is set to 10000"
                 }
                 scroll_id {
                     type: string
@@ -601,7 +498,7 @@ debug_images {
         }
     }
     "2.22": ${debug_images."2.14"} {
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then the retrieving model events. Otherwise task events
             default: false
@@ -624,7 +521,7 @@ plots {
                 }
                 iters {
                     type: integer
-                    description: "Max number of latest iterations for which to return debug images"
+                    description: "Max number of latest iterations for which to return plots"
                 }
                 navigate_earlier {
                     type: boolean
@@ -643,7 +540,7 @@ plots {
         response {"$ref": "#/definitions/plots_response"}
     }
     "2.22": ${plots."2.20"} {
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then the retrieving model plots. Otherwise task plots
             default: false
@@ -693,7 +590,7 @@ get_debug_image_sample {
         }
     }
     "2.22": ${get_debug_image_sample."2.20"} {
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then the retrieving model debug images. Otherwise task debug images
             default: false
@@ -730,7 +627,7 @@ next_debug_image_sample {
             default: false
             description: If set then navigate to the next/previous iteration
         }
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then the retrieving model debug images. Otherwise task debug images
             default: false
@@ -774,7 +671,7 @@ get_plot_sample {
         response {"$ref": "#/definitions/plot_sample_response"}
     }
     "2.22": ${get_plot_sample."2.20"} {
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then the retrieving model plots. Otherwise task plots
             default: false
@@ -811,7 +708,7 @@ next_plot_sample {
             default: false
             description: If set then navigate to the next/previous iteration
         }
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then the retrieving model plots. Otherwise task plots
             default: false
@@ -850,7 +747,7 @@ get_task_metrics{
         }
     }
     "2.22": ${get_task_metrics."2.7"} {
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then get metrics from model events. Otherwise from task events
             default: false
@@ -955,7 +852,7 @@ get_task_log {
                 }
                 total {
                     type: number
-                    description: "Total number of results available for this query"
+                    description: "Total number of results available for this query. In case there are more than 10000 results it is set to 10000"
                 }
                 scroll_id {
                     type: string
@@ -1009,7 +906,7 @@ get_task_log {
                 }
                 total {
                     type: number
-                    description: "Total number of log events available for this query"
+                    description: "Total number of log events available for this query. In case there are more than 10000 events it is set to 10000"
                 }
             }
         }
@@ -1064,7 +961,7 @@ get_task_events {
                 }
                 total {
                     type: number
-                    description: "Total number of results available for this query"
+                    description: "Total number of results available for this query. In case there are more than 10000 results it is set to 10000"
                 }
                 scroll_id {
                     type: string
@@ -1074,7 +971,7 @@ get_task_events {
         }
     }
     "2.22": ${get_task_events."2.1"} {
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then get retrieving model events. Otherwise task events
             default: false
@@ -1130,7 +1027,7 @@ get_task_plots {
                 }
                 iters {
                     type: integer
-                    description: "Max number of latest iterations for which to return debug images"
+                    description: "Max number of latest iterations for which to return plots"
                 }
                 scroll_id {
                     type: string
@@ -1154,7 +1051,7 @@ get_task_plots {
                 }
                 total {
                     type: number
-                    description: "Total number of results available for this query"
+                    description: "Total number of results available for this query. In case there are more than 10000 results it is set to 10000"
                 }
                 scroll_id {
                     type: string
@@ -1182,7 +1079,7 @@ get_task_plots {
         }
     }
     "2.22": ${get_task_plots."2.16"} {
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then the retrieving model events. Otherwise task events
             default: false
@@ -1208,7 +1105,7 @@ get_multi_task_plots {
                 }
                 iters {
                     type: integer
-                    description: "Max number of latest iterations for which to return debug images"
+                    description: "Max number of latest iterations for which to return plots"
                 }
                 scroll_id {
                     type: string
@@ -1229,7 +1126,7 @@ get_multi_task_plots {
                 }
                 total {
                     type: number
-                    description: "Total number of results available for this query"
+                    description: "Total number of results available for this query. In case there are more than 10000 results it is set to 10000"
                 }
                 scroll_id {
                     type: string
@@ -1246,12 +1143,19 @@ get_multi_task_plots {
         }
     }
     "2.22": ${get_multi_task_plots."2.16"} {
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then the retrieving model events. Otherwise task events
             default: false
         }
     }
+    "2.26": ${get_multi_task_plots."2.22"} {
+        request.properties.last_iters_per_task_metric {
+            type: boolean
+            description: If set to 'true' and iters passed then last iterations for each task metrics are retrieved. Otherwise last iterations for the whole task are retrieved
+            default: true
+        }
+    }
 }
 get_vector_metrics_and_variants {
     "2.1" {
@@ -1281,7 +1185,7 @@ get_vector_metrics_and_variants {
         }
     }
     "2.22": ${get_vector_metrics_and_variants."2.1"} {
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then the retrieving model events. Otherwise task events
             default: false
@@ -1326,7 +1230,7 @@ vector_metrics_iter_histogram {
         }
     }
     "2.22": ${vector_metrics_iter_histogram."2.1"} {
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then the retrieving model events. Otherwise task events
             default: false
@@ -1386,7 +1290,7 @@ scalar_metrics_iter_histogram {
         }
     }
     "2.22": ${scalar_metrics_iter_histogram."2.14"} {
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then the retrieving model events. Otherwise task events
             default: false
@@ -1403,7 +1307,7 @@ multi_task_scalar_metrics_iter_histogram {
             ]
             properties {
                 tasks {
-                    description: "List of task Task IDs. Maximum amount of tasks is 10"
+                    description: "List of task Task IDs. Maximum amount of tasks is 100"
                     type: array
                     items {
                         type: string
@@ -1432,7 +1336,7 @@ multi_task_scalar_metrics_iter_histogram {
         }
     }
     "2.22": ${multi_task_scalar_metrics_iter_histogram."2.1"} {
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then the retrieving model events. Otherwise task events
             default: false
@@ -1456,39 +1360,10 @@ get_task_single_value_metrics {
                 }
             }
         }
-        response {
-            type: object
-            properties {
-                tasks {
-                    description: Single value metrics grouped by task
-                    type: array
-                    items {
-                        type: object
-                        properties {
-                            task {
-                                type: string
-                                description: Task ID
-                            }
-                            values {
-                                type: array
-                                items {
-                                    type: object
-                                    properties {
-                                        metric { type: string }
-                                        variant { type: string}
-                                        value { type: number }
-                                        timestamp { type: number }
-                                    }
-                                }
-                            }
-                        }
-                    }
-                }
-            }
-        }
+        response {"$ref": "#/definitions/single_value_metrics_response"}
     }
     "2.22": ${get_task_single_value_metrics."2.20"} {
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then the retrieving model events. Otherwise task events
             default: false
@@ -1574,7 +1449,7 @@ get_scalar_metrics_and_variants {
         }
     }
     "2.22": ${get_scalar_metrics_and_variants."2.1"} {
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then the retrieving model events. Otherwise task events
             default: false
@@ -1595,6 +1470,10 @@ get_scalar_metric_data {
                     type: string
                     description: type of metric
                 }
+                scroll_id {
+                    type: string
+                    description: "Scroll ID of previous call (used for getting more results)"
+                }
             }
         }
         response {
@@ -1617,7 +1496,7 @@ get_scalar_metric_data {
                 }
                 scroll_id {
                     type: string
-                    description: "Scroll ID of previous call (used for getting more results)"
+                    description: "Scroll ID for getting more results"
                 }
             }
         }
@@ -1630,7 +1509,7 @@ get_scalar_metric_data {
         }
     }
     "2.22": ${get_scalar_metric_data."2.16"} {
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then the retrieving model events. Otherwise task events
             default: false
@@ -1701,7 +1580,7 @@ scalar_metrics_iter_raw {
         }
     }
     "2.22": ${scalar_metrics_iter_raw."2.16"} {
-        model_events {
+        request.properties.model_events {
             type: boolean
             description: If set then the retrieving model events. Otherwise task events
             default: false
@@ -1762,4 +1641,4 @@ clear_task_log {
             }
         }
     }
-}
+}

apiserver/schema/services/login.conf

@@ -6,7 +6,7 @@ _default {
 }
 
 supported_modes {
-    authorize: false
+    authorize: null
     "2.9" {
         description:  """ Return supported login modes."""
         request {
@@ -59,7 +59,7 @@ supported_modes {
                     description: "SSO authentication providers"
                     type: object
                     additionalProperties {
-                        desctiprion: "Provider redirect URL"
+                        description: "Provider redirect URL"
                         type: string
                     }
                 }
@@ -95,7 +95,7 @@ supported_modes {
 }
 
 logout {
-    authorize: false
+    authorize: null
     allow_roles = [ "*" ]
     "2.13" {
         description: """ Logout (including SSO, if used)) """

apiserver/schema/services/models.conf

@@ -1,6 +1,6 @@
 _description: """This service provides a management interface for models (results of training tasks) stored in the system."""
 _definitions {
-    include "_common.conf"
+    include "_tasks_common.conf"
     multi_field_pattern_data {
         type: object
         properties {
@@ -104,6 +104,17 @@ _definitions {
                     "$ref": "#/definitions/metadata_item"
                 }
             }
+            last_iteration {
+                description: "Last iteration reported for this model"
+                type: integer
+            }
+            last_metrics {
+                description: "Last metric variants (hash to events), one for each metric hash"
+                type: object
+                additionalProperties {
+                    "$ref": "#/definitions/last_metrics_variants"
+                }
+            }
             stats {
                 description: "Model statistics"
                 type: object
@@ -241,6 +252,23 @@ get_all_ex {
             default: false
         }
     }
+    "2.23": ${get_all_ex."2.20"} {
+        request.properties {
+            allow_public {
+                description: "Allow public models to be returned in the results"
+                type: boolean
+                default: true
+            }
+        }
+    }
+    "2.27": ${get_all_ex."2.23"} {
+        request.properties {
+            filters {
+                type: object
+                additionalProperties: ${_definitions.field_filter}
+            }
+        }
+    }
 }
 get_all {
     "2.1" {
@@ -337,9 +365,6 @@ get_all {
                     "$ref": "#/definitions/multi_field_pattern_data"
                 }
             }
-            dependencies {
-                page: [ page_size ]
-            }
         }
         response {
             type: object
@@ -375,6 +400,17 @@ get_all {
             description: "Scroll ID that can be used with the next calls to get_all to retrieve more data"
         }
     }
+    "2.26": ${get_all."2.15"} {
+        request {
+            properties {
+                include_subprojects {
+                    description: "If set to 'true' and project field is set then models from the subprojects are searched too"
+                    type: boolean
+                    default: false
+                }
+            }
+        }
+    }
 }
 get_frameworks {
     "2.8" {
@@ -976,7 +1012,7 @@ move {
                     items { type: string }
                 }
                 project {
-                    description: "Target project ID. If not provided, `project_name` must be provided."
+                    description: "Target project ID. If not provided, `project_name` must be provided. Use null for the root project"
                     type: string
                 }
                 project_name {
@@ -1055,4 +1091,38 @@ delete_metadata {
             }
         }
     }
+}
+update_tags {
+    "2.27" {
+        description: Add or remove tags from multiple models
+        request {
+            type: object
+            properties {
+                ids {
+                    type: array
+                    description: IDs of the models to update
+                    items {type: string}
+                }
+                add_tags {
+                    type: array
+                    description: User tags to add
+                    items {type: string}
+                }
+                remove_tags {
+                    type: array
+                    description: User tags to remove
+                    items {type: string}
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                updated {
+                    type: integer
+                    description: The number of updated models
+                }
+            }
+        }
+    }
 }

apiserver/schema/services/organization.conf

@@ -1,5 +1,38 @@
 _description: "This service provides organization level operations"
-
+_definitions {
+    value_mapping {
+        type: object
+        required: [key, value]
+        properties {
+            key {
+                description: Original value
+                type: object
+            }
+            value {
+                description: Translated value
+                type: object
+            }
+        }
+    }
+    field_mapping {
+        type: object
+        required: [field]
+        properties {
+            field {
+                description: The source field name as specified in the only_fields
+                type: string
+            }
+            name {
+                description: The column name in the exported csv file
+                type: string
+            }
+            values {
+                type: array
+                items { "$ref": "#/definitions/value_mapping"}
+            }
+        }
+    }
+}
 get_tags {
     "2.8" {
         description: "Get all the user and system tags used for the company tasks and models"
@@ -170,10 +203,96 @@ get_entities_count {
                 default: false
             }
             active_users {
-                descritpion: "The list of users that were active in the project. If passes then the resulting projects are filtered to the ones that have tasks created by these users"
+                description: "The list of users that were active in the project. If passes then the resulting projects are filtered to the ones that have tasks created by these users"
                 type: array
                 items: {type: string}
             }
         }
     }
+    "2.23": ${get_entities_count."2.22"} {
+        request.properties {
+            reports {
+                type: object
+                additionalProperties: true
+                description: Search criteria for reports
+            }
+            allow_public {
+                description: "Allow public entities to be counted in the results"
+                type: boolean
+                default: true
+            }
+        }
+        response.properties {
+            reports {
+                type: integer
+                description: The number of reports matching the criteria
+            }
+        }
+    }
 }
+prepare_download_for_get_all {
+    "2.26": {
+        description: Prepares download from get_all_ex parameters
+        request {
+            type: object
+            required: [ entity_type,  only_fields, field_mappings]
+            properties {
+                only_fields {
+                    description: "List of task field names (nesting is supported using '.', e.g. execution.model_labels). If provided, this list defines the query's projection (only these fields will be returned for each result entry)"
+                    type: array
+                    items {type: string}
+                }
+                allow_public {
+                    description: "Allow public entities to be returned in the results"
+                    type: boolean
+                    default: true
+                }
+                search_hidden {
+                    description: "If set to 'true' then hidden entities are included in the search results"
+                    type: boolean
+                    default: false
+                }
+                entity_type {
+                    description: "The type of the entity to retrieve"
+                    type: string
+                    enum: [
+                        task
+                        model
+                    ]
+                }
+                field_mappings {
+                    description: The name and value mappings for the exported fields. The fields that are not in the mappings will not be exported
+                    type: array
+                    items { "$ref": "#/definitions/field_mapping"}
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                prepare_id {
+                    description: "Prepare ID (use when calling 'download_for_get_all')"
+                    type: string
+                }
+            }
+        }
+    }
+}
+download_for_get_all {
+    "2.26": {
+        description: Generates a file for the download
+        request {
+            type: object
+            required: [ prepare_id ]
+            properties {
+                prepare_id {
+                    description: "Call ID returned by a call to prepare_download_for_get_all"
+                    type: string
+                }
+            }
+        }
+        response {
+            type: string
+        }
+    }
+}

apiserver/schema/services/pipelines.conf

@@ -1,7 +1,42 @@
 _description: "Provides a management API for pipelines in the system."
 _definitions {
+    include "_common.conf"
+}
+delete_runs {
+    "2.26": ${_definitions.batch_operation} {
+        description: Delete pipeline runs
+        request {
+            required: [ids, project]
+            properties {
+                ids.description: "IDs of the pipeline runs to delete. Should be the ids of pipeline controller tasks"
+                project {
+                    description: "Pipeline project ids. When deleting at least one run should be left"
+                    type: string
+                }
+            }
+        }
+        response {
+            properties {
+                succeeded.items.properties.deleted {
+                    description: "Indicates whether the task was deleted"
+                    type: boolean
+                }
+                succeeded.items.properties.updated_children {
+                    description: "Number of child tasks whose parent property was updated"
+                    type: integer
+                }
+                succeeded.items.properties.updated_models {
+                    description: "Number of models whose task property was updated"
+                    type: integer
+                }
+                succeeded.items.properties.deleted_models {
+                    description: "Number of deleted output models"
+                    type: integer
+                }
+            }
+        }
+    }
 }
-
 start_pipeline {
     "2.17" {
         description: "Start a pipeline"
@@ -24,7 +59,7 @@ start_pipeline {
                         type: object
                         properties {
                             name: { type: string }
-                            value: { type: [string, null] }
+                            value: { type: string }
                         }
                     }
                 }

apiserver/schema/services/projects.conf

@@ -1,5 +1,6 @@
 _description: "Provides support for defining Projects containing Tasks, Models and Dataset Versions."
 _definitions {
+    include "_common.conf"
     multi_field_pattern_data {
         type: object
         properties {
@@ -61,12 +62,26 @@ _definitions {
                 type: string
             }
             last_update {
-                description: """Last project update time. Reflects the last time the project metadata was changed or a task in this project has changed status"""
+                description: "Last project update time. Reflects the last time the project metadata was changed or a task in this project has changed status"
                 type: string
                 format: "date-time"
             }
         }
     }
+    stats_datasets {
+        type: object
+        properties {
+            count {
+                description: Number of datasets
+                type: integer
+            }
+            tags {
+                description: Dataset tags
+                type: array
+                items {type: string}
+            }
+        }
+    }
     stats_status_count {
         type: object
         properties {
@@ -141,6 +156,10 @@ _definitions {
                 description: "Stats for archived tasks"
                 "$ref": "#/definitions/stats_status_count"
             }
+            datasets {
+                description: "Stats for childrent datasets"
+                "$ref": "#/definitions/stats_datasets"
+            }
         }
     }
     projects_get_all_response_single {
@@ -191,11 +210,15 @@ _definitions {
                 type: string
             }
             last_update {
-                description: """Last project update time. Reflects the last time the project metadata was changed or a task in this project has changed status"""
+                description: "Last project update time. Reflects the last time the project metadata was changed or a task in this project has changed status"
                 type: string
                 format: "date-time"
             }
             // extra properties
+            hidden {
+                description: "Returned if the search_hidden flag was specified in the get_all_ex call and the project is hidden"
+                type: boolean
+            }
             stats {
                 description: "Additional project stats"
                 "$ref": "#/definitions/stats"
@@ -217,6 +240,10 @@ _definitions {
                     }
                 }
             }
+            own_datasets {
+                description: "The amount of datasets/hyperdatasers under this project (without children projects). Returned if 'check_own_contents' flag is set in the request and children_type is set to 'dataset' or 'hyperdataset'"
+                type: integer
+            }
             own_tasks {
                 description: "The amount of tasks under this project (without children projects). Returned if 'check_own_contents' flag is set in the request"
                 type: integer
@@ -543,7 +570,7 @@ get_all_ex {
         request {
             properties {
                 active_users {
-                    descritpion: "The list of users that were active in the project. If passes then the resulting projects are filtered to the ones that have tasks created by these users"
+                    description: "The list of users that were active in the project. If passes then the resulting projects are filtered to the ones that have tasks created by these users"
                     type: array
                     items: {type: string}
                 }
@@ -611,6 +638,38 @@ get_all_ex {
             default: false
         }
     }
+    "2.23": ${get_all_ex."2.20"} {
+        request.properties {
+            allow_public {
+                description: "Allow public projects to be returned in the results"
+                type: boolean
+                default: true
+            }
+        }
+    }
+    "2.24": ${get_all_ex."2.23"} {
+        request.properties.children_type {
+            description: If specified that only the projects under which the entities of this type can be found will be returned
+            type: string
+            enum: [pipeline, report, dataset]
+        }
+    }
+    "2.25": ${get_all_ex."2.24"} {
+        request.properties.children_tags {
+            description: "The list of tag values to filter children by. Takes effect only if children_type is set. Use 'null' value to specify empty tags. Use '__Snot' value to specify that the following value should be excluded"
+            type: array
+            items {type: string}
+        }
+    }
+    "2.27": ${get_all_ex."2.25"} {
+        request.properties {
+            filters {
+                type: object
+                additionalProperties: ${_definitions.field_filter}
+            }
+            children_tags_filter: ${_definitions.field_filter}
+        }
+    }
 }
 update {
     "2.1" {
@@ -759,6 +818,26 @@ validate_delete {
             }
         }
     }
+    "2.26": ${validate_delete."2.14"} {
+        response.properties {
+            reports {
+                description: "The total number of reports under the project and all its children"
+                type: integer
+            }
+            non_archived_reports {
+                description: "The total number of non-archived reports under the project and all its children"
+                type: integer
+            }
+            pipelines {
+                description: "The total number of pipelines with active controllers under the project and all its children"
+                type: integer
+            }
+            datasets {
+                description: "The total number of non-empty datasets under the project and all its children"
+                type: integer
+            }
+        }
+    }
 }
 delete {
     "2.1" {
@@ -819,6 +898,13 @@ delete {
             }
         }
     }
+    "2.26": ${delete."2.13"} {
+        request.properties.delete_external_artifacts {
+            description: "If set to 'true' then BE will try to delete the extenal artifacts associated with the project tasks and models from the fileserver (if configured to do so)"
+            type: boolean
+            default: true
+        }
+    }
 }
 get_unique_metric_variants {
     "2.1" {
@@ -856,6 +942,13 @@ get_unique_metric_variants {
             }
         }
     }
+    "2.25": ${get_unique_metric_variants."2.13"} {
+        request.properties.model_metrics {
+            description: If set to true then bring unique metric and variant names from the project models otherwise from the project tasks
+            type: boolean
+            default: false
+        }
+    }
 }
 get_hyperparam_values {
     "2.13" {
@@ -903,6 +996,26 @@ get_hyperparam_values {
             }
         }
     }
+    "2.26": ${get_hyperparam_values."2.13"} {
+        request.properties {
+            page {
+                description: "Page number"
+                default: 0
+                type: integer
+            }
+            page_size {
+                description: "Page size"
+                default: 500
+                type: integer
+            }
+        }
+    }
+    "2.27": ${get_hyperparam_values."2.26"} {
+        request.properties.pattern {
+            type: string
+            description: The search pattern regex
+        }
+    }
 }
 get_hyper_parameters {
     "2.9" {
@@ -1000,6 +1113,20 @@ get_model_metadata_values {
             }
         }
     }
+    "2.26": ${get_model_metadata_values."2.17"} {
+        request.properties {
+            page {
+                description: "Page number"
+                default: 0
+                type: integer
+            }
+            page_size {
+                description: "Page size"
+                default: 500
+                type: integer
+            }
+        }
+    }
 }
 get_model_metadata_keys {
     "2.17" {
@@ -1159,13 +1286,15 @@ get_task_parents {
                             }
                             project {
                                 type: object
-                                id {
-                                    description: "The ID of the parent task project"
-                                    type: string
-                                }
-                                name {
-                                    description: "The name of the parent task project"
-                                    type: string
+                                properties {
+                                    id {
+                                        description: "The ID of the parent task project"
+                                        type: string
+                                    }
+                                    name {
+                                        description: "The name of the parent task project"
+                                        type: string
+                                    }
                                 }
                             }
                         }
@@ -1185,4 +1314,58 @@ get_task_parents {
             }
         }
     }
+    "2.25": ${get_task_parents."2.13"} {
+        request.properties.task_name {
+            description: Task name pattern for the returned parent tasks
+            type: string
+        }
+    }
 }
+get_user_names {
+    "2.26" {
+        description: "Get names and ids of the users who created child entitites under the passed projects"
+        request {
+            type: object
+            properties {
+                projects {
+                    description: "The list of projects. If not passed or empty then all the projects are searched"
+                    type: array
+                    items { type: string }
+                }
+                include_subprojects {
+                    description: "If set to 'true' and the projects field is not empty then the result includes user name from the subprojects children"
+                    type: boolean
+                    default: true
+                }
+                entity {
+                    description: The type of the child entity to look for
+                    type: string
+                    enum: [task, model]
+                    default: task
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                users {
+                    description: "The list of users sorted by their names"
+                    type: array
+                    items {
+                        type: object
+                        properties {
+                            id {
+                                description: "The ID of the user"
+                                type: string
+                            }
+                            name {
+                                description: "The name of the user"
+                                type: string
+                            }
+                        }
+                    }
+                }
+            }
+        }
+    }
+}

apiserver/schema/services/queues.conf

@@ -159,6 +159,14 @@ get_all_ex {
             default: false
         }
     }
+    "2.27": ${get_all_ex."2.21"} {
+        request.properties {
+            filters {
+                type: object
+                additionalProperties: ${_definitions.field_filter}
+            }
+        }
+    }
 }
 get_all {
     "2.4" {

apiserver/schema/services/reports.conf

@@ -0,0 +1,747 @@
+_description: "Provides a management API for reports in the system."
+_definitions {
+    include "_tasks_common.conf"
+    include "_events_common.conf"
+    update_response {
+        type: object
+        properties {
+            updated {
+                description: "Number of reports updated (0 or 1)"
+                type: integer
+                enum: [ 0, 1 ]
+            }
+            fields {
+                description: "Updated fields names and values"
+                type: object
+                additionalProperties: true
+            }
+        }
+    }
+    report_status_enum {
+        type: string
+        enum: [
+            created
+            published
+        ]
+    }
+    report {
+        type: object
+        properties {
+            id {
+                description: "Report id"
+                type: string
+            }
+            name {
+                description: "Report Name"
+                type: string
+            }
+            user {
+                description: "Associated user id"
+                type: string
+            }
+            company {
+                description: "Company ID"
+                type: string
+            }
+            status {
+                description: ""
+                "$ref": "#/definitions/report_status_enum"
+            }
+            comment {
+                description: "Free text comment"
+                type: string
+            }
+            report {
+                description: "Report template"
+                type: string
+            }
+            report_assets {
+                description: "List of the external report assets"
+                type: array
+                items { type: string }
+            }
+            created {
+                description: "Report creation time (UTC) "
+                type: string
+                format: "date-time"
+            }
+            project {
+                description: "Project ID of the project to which this report is assigned"
+                type: string
+            }
+            tags {
+                description: "User-defined tags list"
+                type: array
+                items { type: string }
+            }
+            system_tags {
+                description: "System tags list. This field is reserved for system use, please don't use it."
+                type: array
+                items { type: string }
+            }
+            status_changed {
+                description: "Last status change time"
+                type: string
+                format: "date-time"
+            }
+            status_message {
+                description: "free text string representing info about the status"
+                type: string
+            }
+            status_reason {
+                description: "Reason for last status change"
+                type: string
+            }
+            published {
+                description: "Report publish time"
+                type: string
+                format: "date-time"
+            }
+            last_update {
+                description: "Last time this report was created, edited, changed"
+                type: string
+                format: "date-time"
+            }
+        }
+    }
+}
+create {
+    "2.23" {
+        description: "Create a new report"
+        request {
+            type: object
+            required: [
+                name
+            ]
+            properties {
+                name {
+                    description: "Report name. Unique within the company."
+                    type: string
+                }
+                tags {
+                    description: "User-defined tags list"
+                    type: array
+                    items { type: string }
+                }
+                comment {
+                    description: "Free text comment "
+                    type: string
+                }
+                report {
+                    description: "Report template"
+                    type: string
+                }
+                project {
+                    description: "Project ID of the project to which this report is assigned Must exist[ab]"
+                    type: string
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                id {
+                    description: "ID of the report"
+                    type: string
+                }
+                project_id {
+                    description: "ID of the project that the report belongs to"
+                    type: string
+                }
+            }
+        }
+    }
+    "2.24": ${create."2.23"} {
+        request.properties.report_assets {
+            description: "List of the external report assets"
+            type: array
+            items { type: string }
+        }
+    }
+}
+update {
+    "2.23" {
+        description: "Create a new report"
+        request {
+            type: object
+            required: [
+                task
+            ]
+            properties {
+                task {
+                    description: "The ID of the report task to update"
+                    type: string
+                }
+                name {
+                    description: "Report name. Unique within the company."
+                    type: string
+                }
+                tags {
+                    description: "User-defined tags list"
+                    type: array
+                    items { type: string }
+                }
+                comment {
+                    description: "Free text comment "
+                    type: string
+                }
+                report {
+                    description: "Report template"
+                    type: string
+                }
+            }
+        }
+        response: ${_definitions.update_response}
+    }
+    "2.24": ${update."2.23"} {
+        request.properties.report_assets {
+            description: "List of the external report assets"
+            type: array
+            items { type: string }
+        }
+    }
+}
+move {
+    "2.23" {
+        description: "Move reports to a project"
+        request {
+            type: object
+            required: [task]
+            properties {
+                task {
+                    description: "ID of the report to move"
+                    type: string
+                }
+                project {
+                    description: "Target project ID. If not provided, `project_name` must be provided. Use null for the root project"
+                    type: string
+                }
+                project_name {
+                    description: "Target project name. If provided and a project with this name does not exist, a new project will be created. If not provided, `project` must be provided."
+                    type: string
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                project_id: {
+                    description: The ID of the target project
+                    type: string
+                }
+            }
+        }
+    }
+}
+publish {
+    "2.23" {
+        description: "Publish report"
+        request {
+            type: object
+            required: [
+                task
+            ]
+            properties {
+                task {
+                    description: "The ID of the report task to publish"
+                    type: string
+                }
+                comment {
+                    description: "The client message"
+                    type: string
+                }
+            }
+        }
+        response: ${_definitions.update_response}
+    }
+}
+archive {
+    "2.23" {
+        description: "Archive report"
+        request {
+            type: object
+            required: [
+                task
+            ]
+            properties {
+                task {
+                    description: "The ID of the report task to archive"
+                    type: string
+                }
+                comment {
+                    description: "The client message"
+                    type: string
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                archived {
+                    description: "Number of reports archived (0 or 1)"
+                    type: integer
+                    enum: [0, 1]
+                }
+            }
+        }
+    }
+}
+unarchive {
+    "2.23" {
+        description: "Unarchive report"
+        request {
+            type: object
+            required: [
+                task
+            ]
+            properties {
+                task {
+                    description: "The ID of the report task to unarchive"
+                    type: string
+                }
+                comment {
+                    description: "The client message"
+                    type: string
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                unarchived {
+                    description: "Number of reports unarchived (0 or 1)"
+                    type: integer
+                    enum: [0, 1]
+                }
+            }
+        }
+    }
+}
+//share {
+//    "999.0" {
+//        description: "Share or unshare report"
+//        request {
+//            type: object
+//            required: [
+//                task
+//            ]
+//            properties {
+//                task {
+//                    description: "The ID of the report task to share/unshare"
+//                    type: string
+//                }
+//                share {
+//                    description: "If set to 'true' then the report will be shared. Otherwise unshared."
+//                    type: boolean
+//                    default: true
+//                }
+//            }
+//        }
+//        response {
+//            type: object
+//            properties {
+//                changed {
+//                    description: "Number of changed reports (0 or 1)"
+//                    type: integer
+//                    enum: [0, 1]
+//                }
+//            }
+//        }
+//    }
+//}
+delete {
+    "2.23" {
+        description: "Delete report"
+        request {
+            type: object
+            required: [
+                task
+            ]
+            properties {
+                task {
+                    description: "The ID of the report task to delete"
+                    type: string
+                }
+                force {
+                    description: "If not set then published or unarchived reports cannot be deleted"
+                    type: boolean
+                    default: false
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                deleted {
+                    description: "Number of deleted reports (0 or 1)"
+                    type: integer
+                    enum: [0, 1]
+                }
+            }
+        }
+    }
+}
+get_task_data {
+    "2.23" {
+        description: "Get the tasks data according the passed search criteria + requested events"
+        request {
+            type: object
+            properties {
+                id {
+                    description: "List of IDs to filter by"
+                    type: array
+                    items { type: string }
+                }
+                name {
+                    description: "Get only tasks whose name matches this pattern (python regular expression syntax)"
+                    type: string
+                }
+                user {
+                    description: "List of user IDs used to filter results by the task's creating user"
+                    type: array
+                    items { type: string }
+                }
+                size {
+                    type: integer
+                    minimum: 1
+                    description: "The number of tasks to retrieve"
+                }
+                order_by {
+                    description: "List of field names to order by. When search_text is used, '@text_score' can be used as a field representing the text score of returned documents. Use '-' prefix to specify descending order. Optional, recommended when using page"
+                    type: array
+                    items { type: string }
+                }
+                type {
+                    description: "List of task types. One or more of: 'import', 'annotation', 'training' or 'testing' (case insensitive)"
+                    type: array
+                    items { type: string }
+                }
+                tags {
+                    description: "List of task user-defined tags. Use '-' prefix to exclude tags"
+                    type: array
+                    items { type: string }
+                }
+                system_tags {
+                    description: "List of task system tags. Use '-' prefix to exclude system tags"
+                    type: array
+                    items { type: string }
+                }
+                status {
+                    description: "List of task status."
+                    type: array
+                    items { "$ref": "#/definitions/task_status_enum" }
+                }
+                project {
+                    description: "List of project IDs"
+                    type: array
+                    items { type: string }
+                }
+                only_fields {
+                    description: "List of task field names (nesting is supported using '.', e.g. execution.model_labels). If provided, this list defines the query's projection (only these fields will be returned for each result entry)"
+                    type: array
+                    items { type: string }
+                }
+                parent {
+                    description: "Parent ID"
+                    type: string
+                }
+                status_changed {
+                    description: "List of status changed constraint strings (utcformat, epoch) with an optional prefix modifier (>, >=, <, <=)"
+                    type: array
+                    items {
+                        type: string
+                        pattern: "^(>=|>|<=|<)?.*$"
+                    }
+                }
+                search_text {
+                    description: "Free text search query"
+                    type: string
+                }
+                allow_public {
+                    description: "Allow public tasks to be returned in the results"
+                    type: boolean
+                    default: true
+                }
+                _all_ {
+                    description: "Multi-field pattern condition (all fields match pattern)"
+                    "$ref": "#/definitions/multi_field_pattern_data"
+                }
+                _any_ {
+                    description: "Multi-field pattern condition (any field matches pattern)"
+                    "$ref": "#/definitions/multi_field_pattern_data"
+                }
+                "input.view.entries.dataset" {
+                    description: "List of input dataset IDs"
+                    type: array
+                    items { type: string }
+                }
+                "input.view.entries.version" {
+                    description: "List of input dataset version IDs"
+                    type: array
+                    items { type: string }
+                }
+                search_hidden {
+                    description: "If set to 'true' then hidden tasks are included in the search results"
+                    type: boolean
+                    default: false
+                }
+                include_subprojects {
+                    description: "If set to 'true' and project field is set then tasks from the subprojects are searched too"
+                    type: boolean
+                    default: false
+                }
+                plots {
+                    type: object
+                    properties {
+                        iters {
+                            type: integer
+                            description: "Max number of latest iterations for which to return plots"
+                        }
+                        metrics {
+                            type: array
+                            description: List of metrics and variants
+                            items { "$ref": "#/definitions/metric_variants" }
+                        }
+                    }
+                }
+                debug_images {
+                    type: object
+                    properties {
+                        iters {
+                            type: integer
+                            description: "Max number of latest iterations for which to return debug images"
+                        }
+                        metrics {
+                            type: array
+                            description: List of metrics and variants
+                            items { "$ref": "#/definitions/metric_variants" }
+                        }
+                    }
+                }
+                scalar_metrics_iter_histogram {
+                    type: object
+                    properties {
+                        samples {
+                            description: "The amount of histogram points to return (0 to return all the points). Optional, the default value is 6000."
+                            type: integer
+                        }
+                        key {
+                            description: """
+                            Histogram x axis to use:
+                            iter - iteration number
+                            iso_time - event time as ISO formatted string
+                            timestamp - event timestamp as milliseconds since epoch
+                            """
+                            "$ref": "#/definitions/scalar_key_enum"
+                        }
+                        metrics {
+                            type: array
+                            description: List of metrics and variants
+                            items { "$ref": "#/definitions/metric_variants" }
+                        }
+                    }
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                tasks {
+                    description: "List of tasks"
+                    type: array
+                    items { "$ref": "#/definitions/task" }
+                }
+                plots {
+                    type: object
+                    description: "Plots mapped by metric, variant, task and iteration"
+                    additionalProperties: true
+                }
+                debug_images {
+                    type: array
+                    description: "Debug image events grouped by tasks and iterations"
+                    items {"$ref": "#/definitions/debug_images_response_task_metrics"}
+                }
+                scalar_metrics_iter_histogram {
+                    type: object
+                    additionalProperties: true
+                }
+            }
+        }
+    }
+    "2.25": ${get_task_data."2.23"} {
+        request.properties {
+            model_events {
+                type: boolean
+                description: If set then the retrieving model events. Otherwise task events
+                default: false
+            }
+            single_value_metrics {
+                type: object
+                description: If passed then task single value metrics are returned
+                additionalProperties: false
+            }
+        }
+        response.properties.single_value_metrics {
+            type: array
+            description: Single value metrics grouped by task
+            items {"$ref": "#/definitions/single_value_task_metrics"}
+        }
+    }
+    "2.26": ${get_task_data."2.25"} {
+        request.properties.plots.properties.last_iters_per_task_metric {
+            type: boolean
+            description: If set to 'true' and iters passed then last iterations for each task metrics are retrieved. Otherwise last iterations for the whole task are retrieved
+            default: true
+        }
+    }
+}
+get_all_ex {
+    "2.23" {
+        description: "Get all the company's and public report tasks"
+        request {
+            type: object
+            properties {
+                id {
+                    description: "List of IDs to filter by"
+                    type: array
+                    items { type: string }
+                }
+                name {
+                    description: "Get only reports whose name matches this pattern (python regular expression syntax)"
+                    type: string
+                }
+                user {
+                    description: "List of user IDs used to filter results by the reports's creating user"
+                    type: array
+                    items { type: string }
+                }
+                page {
+                    description: "Page number, returns a specific page out of the resulting list of reports"
+                    type: integer
+                    minimum: 0
+                }
+                page_size {
+                    description: "Page size, specifies the number of results returned in each page (last page may contain fewer results)"
+                    type: integer
+                    minimum: 1
+                }
+                order_by {
+                    description: "List of field names to order by. When search_text is used, '@text_score' can be used as a field representing the text score of returned documents. Use '-' prefix to specify descending order. Optional, recommended when using page"
+                    type: array
+                    items { type: string }
+                }
+                tags {
+                    description: "List of report user-defined tags. Use '-' prefix to exclude tags"
+                    type: array
+                    items { type: string }
+                }
+                system_tags {
+                    description: "List of report system tags. Use '-' prefix to exclude system tags"
+                    type: array
+                    items { type: string }
+                }
+                status {
+                    description: "List of report status."
+                    type: array
+                    items { "$ref": "#/definitions/report_status_enum" }
+                }
+                project {
+                    description: "List of project IDs"
+                    type: array
+                    items { type: string }
+                }
+                only_fields {
+                    description: "List of report field names (nesting is supported using '.'). If provided, this list defines the query's projection (only these fields will be returned for each result entry)"
+                    type: array
+                    items { type: string }
+                }
+                status_changed {
+                    description: "List of status changed constraint strings (utcformat, epoch) with an optional prefix modifier (>, >=, <, <=)"
+                    type: array
+                    items {
+                        type: string
+                        pattern: "^(>=|>|<=|<)?.*$"
+                    }
+                }
+                search_text {
+                    description: "Free text search query"
+                    type: string
+                }
+                scroll_id {
+                    type: string
+                    description: "Scroll ID returned from the previos calls to get_all"
+                }
+                refresh_scroll {
+                    type: boolean
+                    description: "If set then all the data received with this scroll will be requeried"
+                }
+                size {
+                    type: integer
+                    minimum: 1
+                    description: "The number of tasks to retrieve"
+                }
+                allow_public {
+                    description: "Allow public reports to be returned in the results"
+                    type: boolean
+                    default: true
+                }
+                _all_ {
+                    description: "Multi-field pattern condition (all fields match pattern)"
+                    "$ref": "#/definitions/multi_field_pattern_data"
+                }
+                _any_ {
+                    description: "Multi-field pattern condition (any field matches pattern)"
+                    "$ref": "#/definitions/multi_field_pattern_data"
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                tasks {
+                    description: "List of report tasks"
+                    type: array
+                    items { "$ref": "#/definitions/report" }
+                }
+                scroll_id {
+                    type: string
+                    description: "Scroll ID that can be used with the next calls to get_all to retrieve more data"
+                }
+            }
+        }
+    }
+    "2.26": ${get_all_ex."2.23"} {
+        request.properties.include_subprojects {
+            description: "If set to 'true' and project field is set then reports from the subprojects are searched too"
+            type: boolean
+            default: false
+        }
+    }
+    "2.27": ${get_all_ex."2.26"} {
+        request.properties {
+            filters {
+                type: object
+                additionalProperties: ${_definitions.field_filter}
+            }
+        }
+    }
+}
+get_tags {
+    "2.23" {
+        description: "Get all the user tags used for the company reports"
+        request {
+            type: object
+            additionalProperties: false
+        }
+        response {
+            type: object
+            properties {
+                tags {
+                    description: "The list of unique tag values"
+                    type: array
+                    items {type: string}
+                }
+            }
+        }
+    }
+}

apiserver/schema/services/tasks.conf

@@ -25,7 +25,7 @@ _references {
     }
 }
 _definitions {
-    include "_common.conf"
+    include "_tasks_common.conf"
     change_many_request: ${_definitions.batch_operation} {
         request {
             properties {
@@ -69,374 +69,6 @@ _definitions {
             }
         }
     }
-    multi_field_pattern_data {
-        type: object
-        properties {
-            pattern {
-                description: "Pattern string (regex)"
-                type: string
-            }
-            fields {
-                description: "List of field names"
-                type: array
-                items { type: string }
-            }
-        }
-    }
-    model_type_enum {
-        type: string
-        enum: ["input", "output"]
-    }
-    task_model_item {
-        type: object
-        required: [ name, model]
-        properties {
-            name {
-                description: "The task model name"
-                type: string
-            }
-            model {
-                description: "The model ID"
-                type: string
-            }
-        }
-    }
-    script {
-        type: object
-        properties {
-            binary {
-                description: "Binary to use when running the script"
-                type: string
-                default: python
-            }
-            repository {
-                description: "Name of the repository where the script is located"
-                type: string
-            }
-            tag {
-                description: "Repository tag"
-                type: string
-            }
-            branch  {
-                description: "Repository branch id If not provided and tag not provided, default repository branch is used."
-                type: string
-            }
-            version_num {
-                description: "Version (changeset) number. Optional (default is head version) Unused if tag is provided."
-                type: string
-            }
-            entry_point {
-                description: "Path to execute within the repository"
-                type: string
-            }
-            working_dir {
-                description: "Path to the folder from which to run the script Default - root folder of repository"
-                type: string
-            }
-            requirements {
-                description: "A JSON object containing requirements strings by key"
-                type: object
-            }
-            diff {
-                description: "Uncommitted changes found in the repository when task was run"
-                type: string
-            }
-        }
-    }
-    output {
-        type: object
-        properties {
-            destination {
-                description: "Storage id. This is where output files will be stored."
-                type: string
-            }
-            model {
-                description: "Model id."
-                type: string
-            }
-            result {
-                description: "Task result. Values: 'success', 'failure'"
-                type: string
-            }
-            error {
-                description: "Last error text"
-                type: string
-            }
-        }
-    }
-    task_execution_progress_enum {
-        type: string
-        enum: [
-            unknown
-            running
-            stopping
-            stopped
-        ]
-    }
-    output_rois_enum {
-        type: string
-        enum: [
-            all_in_frame
-            only_filtered
-            frame_per_roi
-        ]
-    }
-    artifact_type_data {
-        type: object
-        properties {
-            preview {
-                description: "Description or textual data"
-                type: string
-            }
-            content_type {
-                description: "System defined raw data content type"
-                type: string
-            }
-            data_hash {
-                description: "Hash of raw data, without any headers or descriptive parts"
-                type: string
-            }
-        }
-    }
-    artifact_mode_enum {
-        type: string
-        enum: [
-            input
-            output
-        ]
-        default: output
-    }
-    artifact {
-        type: object
-        required: [key, type]
-        properties {
-            key {
-                description: "Entry key"
-                type: string
-            }
-            type {
-                description: "System defined type"
-                type: string
-            }
-            mode {
-                description: "System defined input/output indication"
-                "$ref": "#/definitions/artifact_mode_enum"
-            }
-            uri {
-                description: "Raw data location"
-                type: string
-            }
-            content_size {
-                description: "Raw data length in bytes"
-                type: integer
-            }
-            hash {
-                description: "Hash of entire raw data"
-                type: string
-            }
-            timestamp {
-                description: "Epoch time when artifact was created"
-                type: integer
-            }
-            type_data {
-                description: "Additional fields defined by the system"
-                "$ref": "#/definitions/artifact_type_data"
-            }
-            display_data {
-                description: "User-defined list of key/value pairs, sorted"
-                type: array
-                items {
-                    type: array
-                    items {
-                        type: string  # can also be a number... TODO: upgrade the generator
-                    }
-                }
-            }
-        }
-    }
-    artifact_id {
-        type: object
-        required: [key]
-        properties {
-            key {
-                description: "Entry key"
-                type: string
-            }
-            mode {
-                description: "System defined input/output indication"
-                "$ref": "#/definitions/artifact_mode_enum"
-            }
-        }
-    }
-    task_models {
-        type: object
-        properties {
-            input {
-                description: "The list of task input models"
-                type: array
-                items {"$ref": "#/definitions/task_model_item"}
-
-            }
-            output {
-                description: "The list of task output models"
-                type: array
-                items {"$ref": "#/definitions/task_model_item"}
-            }
-        }
-    }
-    execution {
-        type: object
-        properties {
-            queue {
-                description: "Queue ID where task was queued."
-                type: string
-            }
-            parameters {
-                description: "Json object containing the Task parameters"
-                type: object
-                additionalProperties: true
-            }
-            model {
-                description: "Execution input model ID Not applicable for Register (Import) tasks"
-                type: string
-            }
-            model_desc {
-                description: "Json object representing the Model descriptors"
-                type: object
-                additionalProperties: true
-            }
-            model_labels {
-                description: """Json object representing the ids of the labels in the model.
-                The keys are the layers' names and the values are the IDs.
-                Not applicable for Register (Import) tasks.
-                Mandatory for Training tasks"""
-                type: object
-                additionalProperties: { type: integer }
-            }
-            framework {
-                description: """Framework related to the task. Case insensitive. Mandatory for Training tasks. """
-                type: string
-            }
-            docker_cmd {
-                description: "Command for running docker script for the execution of the task"
-                type: string
-            }
-            artifacts {
-                description: "Task artifacts"
-                type: array
-                items { "$ref": "#/definitions/artifact" }
-            }
-        }
-    }
-    task_status_enum {
-        type: string
-        enum: [
-            created
-            queued
-            in_progress
-            stopped
-            published
-            publishing
-            closed
-            failed
-            completed
-            unknown
-        ]
-    }
-    task_type_enum {
-        type: string
-        enum: [
-            training
-            testing
-            inference
-            data_processing
-            application
-            monitor
-            controller
-            optimizer
-            service
-            qc
-            custom
-        ]
-    }
-    last_metrics_event {
-        type: object
-        properties {
-            metric {
-                description: "Metric name"
-                type: string
-            }
-            variant {
-                description: "Variant name"
-                type: string
-            }
-            value {
-                description: "Last value reported"
-                type: number
-            }
-            min_value {
-                description: "Minimum value reported"
-                type: number
-            }
-            max_value {
-                description: "Maximum value reported"
-                type: number
-            }
-        }
-    }
-    last_metrics_variants {
-        type: object
-        description: "Last metric events, one for each variant hash"
-        additionalProperties {
-            "$ref": "#/definitions/last_metrics_event"
-        }
-    }
-    params_item {
-        type: object
-        properties {
-            section {
-                description: "Section that the parameter belongs to"
-                type: string
-            }
-            name {
-                description: "Name of the parameter. The combination of section and name should be unique"
-                type: string
-            }
-            value {
-                description: "Value of the parameter"
-                type: string
-            }
-            type {
-                description: "Type of the parameter. Optional"
-                type: string
-            }
-            description {
-                description: "The parameter description. Optional"
-                type: string
-            }
-        }
-    }
-    configuration_item {
-        type: object
-        properties {
-            name {
-                description: "Name of the parameter. Should be unique"
-                type: string
-            }
-            value {
-                description: "Value of the parameter"
-                type: string
-            }
-            type {
-                description: "Type of the parameter. Optional"
-                type: string
-            }
-            description {
-                description: "The parameter description. Optional"
-                type: string
-            }
-        }
-    }
     param_key {
         type: object
         properties {
@@ -450,13 +82,6 @@ _definitions {
             }
         }
     }
-    section_params {
-        description: "Task section params"
-        type: object
-        additionalProperties {
-            "$ref": "#/definitions/params_item"
-        }
-    }
     replace_hyperparams_enum {
         type: string
         enum: [
@@ -465,165 +90,6 @@ _definitions {
             all
         ]
     }
-    task {
-        type: object
-        properties {
-            id {
-                description: "Task id"
-                type: string
-            }
-            name {
-                description: "Task Name"
-                type: string
-            }
-            user {
-                description: "Associated user id"
-                type: string
-            }
-            company {
-                description: "Company ID"
-                type: string
-            }
-            type {
-                description: "Type of task. Values: 'training', 'testing'"
-                "$ref": "#/definitions/task_type_enum"
-            }
-            status {
-                description: ""
-                "$ref": "#/definitions/task_status_enum"
-            }
-            comment {
-                description: "Free text comment"
-                type: string
-            }
-            created {
-                description: "Task creation time (UTC) "
-                type: string
-                format: "date-time"
-            }
-            started {
-                description: "Task start time (UTC)"
-                type: string
-                format: "date-time"
-            }
-            completed {
-                description: "Task end time (UTC)"
-                type: string
-                format: "date-time"
-            }
-            active_duration {
-                description: "Task duration time (seconds)"
-                type: integer
-            }
-            parent {
-                description: "Parent task id"
-                type: string
-            }
-            project {
-                description: "Project ID of the project to which this task is assigned"
-                type: string
-            }
-            output {
-                description: "Task output params"
-                "$ref": "#/definitions/output"
-            }
-            execution {
-                description: "Task execution params"
-                "$ref": "#/definitions/execution"
-            }
-            container {
-                description: "Docker container parameters"
-                type: object
-                additionalProperties { type: [string, null] }
-            }
-            models {
-                description: "Task models"
-                "$ref": "#/definitions/task_models"
-            }
-            // TODO: will be removed
-            script {
-                description: "Script info"
-                "$ref": "#/definitions/script"
-            }
-            tags {
-                description: "User-defined tags list"
-                type: array
-                items { type: string }
-            }
-            system_tags {
-                description: "System tags list. This field is reserved for system use, please don't use it."
-                type: array
-                items { type: string }
-            }
-            status_changed {
-                description: "Last status change time"
-                type: string
-                format: "date-time"
-            }
-            status_message {
-                description: "free text string representing info about the status"
-                type: string
-            }
-            status_reason {
-                description: "Reason for last status change"
-                type: string
-            }
-            published {
-                description: "Last status change time"
-                type: string
-                format: "date-time"
-            }
-            last_worker {
-                description: "ID of last worker that handled the task"
-                type: string
-            }
-            last_worker_report {
-                description: "Last time a worker reported while working on this task"
-                type: string
-                format: "date-time"
-            }
-            last_update {
-                description: "Last time this task was created, updated, changed or events for this task were reported"
-                type: string
-                format: "date-time"
-            }
-            last_change {
-                description: "Last time any update was done to the task"
-                type: string
-                format: "date-time"
-            }
-            last_iteration {
-                description: "Last iteration reported for this task"
-                type: integer
-            }
-            last_metrics {
-                description: "Last metric variants (hash to events), one for each metric hash"
-                type: object
-                additionalProperties {
-                    "$ref": "#/definitions/last_metrics_variants"
-                }
-            }
-            hyperparams {
-                description: "Task hyper params per section"
-                type: object
-                additionalProperties {
-                    "$ref": "#/definitions/section_params"
-                }
-            }
-            configuration {
-                description: "Task configuration params"
-                type: object
-                additionalProperties {
-                    "$ref": "#/definitions/configuration_item"
-                }
-            }
-            runtime {
-                description: "Task runtime mapping"
-                type: object
-                additionalProperties: true
-            }
-        }
-    }
     task_urls {
         type: object
         properties {
@@ -715,6 +181,23 @@ get_all_ex {
             description: "Scroll ID that can be used with the next calls to get_all_ex to retrieve more data"
         }
     }
+    "2.23": ${get_all_ex."2.15"} {
+        request.properties {
+            allow_public {
+                description: "Allow public tasks to be returned in the results"
+                type: boolean
+                default: true
+            }
+        }
+    }
+    "2.27": ${get_all_ex."2.23"} {
+        request.properties {
+            filters {
+                type: object
+                additionalProperties: ${_definitions.field_filter}
+            }
+        }
+    }
 }
 get_all {
     "2.1" {
@@ -814,9 +297,6 @@ get_all {
                     "$ref": "#/definitions/multi_field_pattern_data"
                 }
             }
-            dependencies {
-                page: [ page_size ]
-            }
         }
         response {
             type: object
@@ -859,6 +339,17 @@ get_all {
             description: "Scroll ID that can be used with the next calls to get_all to retrieve more data"
         }
     }
+    "2.26": ${get_all."2.15"} {
+        request {
+            properties {
+                include_subprojects {
+                    description: "If set to 'true' and project field is set then tasks from the subprojects are searched too"
+                    type: boolean
+                    default: false
+                }
+            }
+        }
+    }
 }
 get_types {
     "2.8" {
@@ -995,7 +486,7 @@ clone {
                 new_task_container {
                     description: "The docker container properties for the new task. If not provided then taken from the original task"
                     type: object
-                    additionalProperties { type: [string, null] }
+                    additionalProperties { type: string }
                 }
             }
         }
@@ -1173,7 +664,7 @@ create {
                 container {
                     description: "Docker container parameters"
                     type: object
-                    additionalProperties { type: [string, null] }
+                    additionalProperties { type: string }
                 }
             }
         }
@@ -1227,6 +718,10 @@ validate {
                     description: "Task execution params"
                     "$ref": "#/definitions/execution"
                 }
+                script {
+                    description: "Script info"
+                    "$ref": "#/definitions/script"
+                }
                 hyperparams {
                     description: "Task hyper params per section"
                     type: object
@@ -1241,10 +736,6 @@ validate {
                         "$ref": "#/definitions/configuration_item"
                     }
                 }
-                script {
-                    description: "Script info"
-                    "$ref": "#/definitions/script"
-                }
             }
         }
         response {
@@ -1262,7 +753,7 @@ validate {
                 container {
                     description: "Docker container parameters"
                     type: object
-                    additionalProperties { type: [string, null] }
+                    additionalProperties { type: string }
                 }
             }
         }
@@ -1392,6 +883,10 @@ edit {
                     description: "Task execution params"
                     "$ref": "#/definitions/execution"
                 }
+                script {
+                    description: "Script info"
+                    "$ref": "#/definitions/script"
+                }
                 hyperparams {
                     description: "Task hyper params per section"
                     type: object
@@ -1406,10 +901,6 @@ edit {
                         "$ref": "#/definitions/configuration_item"
                     }
                 }
-                script {
-                    description: "Script info"
-                    "$ref": "#/definitions/script"
-                }
             }
         }
         response: ${_definitions.update_response}
@@ -1424,7 +915,7 @@ edit {
                 container {
                     description: "Docker container parameters"
                     type: object
-                    additionalProperties { type: [string, null] }
+                    additionalProperties { type: string }
                 }
                 runtime {
                     description: "Task runtime mapping"
@@ -1478,6 +969,10 @@ reset {
                     description: "If set to 'true' then return the urls of the files that were uploaded by this task. Default value is 'false'"
                     type: boolean
                 }
+                delete_output_models {
+                    description: "If set to 'true' then delete output models of this task that are not referenced by other tasks. Default value is 'true'"
+                    type: boolean
+                }
             }
         }
         response {
@@ -1665,6 +1160,10 @@ delete {
                     description: "If set to 'true' then return the urls of the files that were uploaded by this task. Default value is 'false'"
                     type: boolean
                 }
+                delete_output_models {
+                    description: "If set to 'true' then delete output models of this task that are not referenced by other tasks. Default value is 'true'"
+                    type: boolean
+                }
             }
         }
         response {
@@ -1736,12 +1235,12 @@ archive_many {
                     type: string
                 }
             }
-            response {
-                properties {
-                    succeeded.items.properties.archived {
-                        description: "Indicates whether the task was archived"
-                        type: boolean
-                    }
+        }
+        response {
+            properties {
+                succeeded.items.properties.archived {
+                    description: "Indicates whether the task was archived"
+                    type: boolean
                 }
             }
         }
@@ -2024,6 +1523,19 @@ dequeue {
             }
         }
     }
+    "2.25": ${dequeue."1.5"} {
+        request.properties.remove_from_all_queues {
+            type: boolean
+            description: If set to 'true' then the task is searched and removed from all the queues. Otherwise only from the queue stored in the task execution parameters
+            default: false
+        }
+    }
+    "2.26": ${dequeue."2.25"} {
+        request.properties.new_status {
+            type: string
+            description: The new status to assign to the task after the dequeue instead of the default one
+        }
+    }
 }
 dequeue_many {
     "2.13": ${_definitions.change_many_request} {
@@ -2042,6 +1554,19 @@ dequeue_many {
             }
         }
     }
+    "2.25": ${dequeue_many."2.13"} {
+        request.properties.remove_from_all_queues {
+            type: boolean
+            description: If set to 'true' then the tasks are searched and removed from all the queues. Otherwise only from the queue stored in the task execution parameters
+            default: false
+        }
+    }
+    "2.26": ${dequeue_many."2.25"} {
+        request.properties.new_status {
+            type: string
+            description: The new status to assign to the task after the dequeue instead of the default one
+        }
+    }
 }
 set_requirements {
     "2.1" {
@@ -2515,7 +2040,7 @@ move {
                     items { type: string }
                 }
                 project {
-                    description: "Target project ID. If not provided, `project_name` must be provided."
+                    description: "Target project ID. If not provided, `project_name` must be provided. Use null for the root project"
                     type: string
                 }
                 project_name {
@@ -2530,3 +2055,37 @@ move {
         }
     }
 }
+update_tags {
+    "2.27" {
+        description: Add or remove tags from multiple tasks
+        request {
+            type: object
+            properties {
+                ids {
+                    type: array
+                    description: IDs of the tasks to update
+                    items {type: string}
+                }
+                add_tags {
+                    type: array
+                    description: User tags to add
+                    items {type: string}
+                }
+                remove_tags {
+                    type: array
+                    description: User tags to remove
+                    items {type: string}
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                updated {
+                    type: integer
+                    description: The number of updated tasks
+                }
+            }
+        }
+    }
+}

apiserver/schema/services/users.conf

@@ -155,6 +155,17 @@ get_current_user {
             }
         }
     }
+    "2.26": ${get_current_user."2.20"} {
+        response.properties.settings {
+            type: object
+            properties {
+                max_download_items {
+                    type: string
+                    description: The maximum items downloaded for this user in csv file downloads
+                }
+            }
+        }
+    }
 }
 
 get_all_ex {

apiserver/schema/services/workers.conf

@@ -311,6 +311,41 @@ get_all {
         }
     }
 }
+get_count {
+    "2.26": {
+        description: "Returns the number of registered workers."
+        request {
+            type: object
+            properties {
+                last_seen {
+                    description: """Filter out workers not active for more than last_seen seconds.
+                        A value or 0 or 'none' will disable the filter."""
+                    type: integer
+                    default: 0
+                }
+                tags {
+                    description: The list of allowed worker tags. Prepend tag value with '-' in order to exclude
+                    type: array
+                    items { type: string }
+                }
+                system_tags {
+                    description: The list of allowed worker system tags. Prepend tag value with '-' in order to exclude
+                    type: array
+                    items { type: string }
+                }
+            }
+        }
+        response {
+            type: object
+            properties {
+                count {
+                    description: Workers count
+                    type: integer
+                }
+            }
+        }
+    }
+}
 register {
     "2.4" {
         description: "Register a worker in the system. Called by the Worker Daemon."

apiserver/server_init/app_sequence.py

@@ -46,7 +46,6 @@ class AppSequence:
         self._attach_request_handlers(request_handlers)
 
     def _attach_request_handlers(self, request_handlers: RequestHandlers):
-        self.app.before_first_request(request_handlers.before_app_first_request)
         self.app.before_request(request_handlers.before_request)
         self.app.after_request(request_handlers.after_request)
 

apiserver/server_init/request_handlers.py

@@ -1,3 +1,5 @@
+import unicodedata
+import urllib.parse
 from functools import partial
 
 from flask import request, Response, redirect
@@ -20,9 +22,6 @@ class RequestHandlers:
     _request_strip_prefix = config.get("apiserver.request.strip_prefix", None)
     _server_header = config.get("apiserver.response.headers.server", "clearml")
 
-    def before_app_first_request(self):
-        pass
-
     def before_request(self):
         if request.method == "OPTIONS":
             return "", 200
@@ -44,9 +43,17 @@ class RequestHandlers:
             else:
                 headers = None
                 if call.result.filename:
-                    headers = {
-                        "Content-Disposition": f"attachment; filename={call.result.filename}"
-                    }
+                    try:
+                        call.result.filename.encode("ascii")
+                    except UnicodeEncodeError:
+                        simple = unicodedata.normalize("NFKD", call.result.filename)
+                        simple = simple.encode("ascii", "ignore").decode("ascii")
+                        # safe = RFC 5987 attr-char
+                        quoted = urllib.parse.quote(call.result.filename, safe="")
+                        filenames = f"filename={simple}; filename*=UTF-8''{quoted}"
+                    else:
+                        filenames = f"filename={call.result.filename}"
+                    headers = {"Content-Disposition": "attachment; " + filenames}
 
                 response = Response(
                     content,

apiserver/service_repo/apicall.py

@@ -655,7 +655,11 @@ class APICall(DataContainer):
             }
             if self.content_type.lower() == JSON_CONTENT_TYPE:
                 try:
-                    func = json.dumps if self._json_flags.pop("ensure_ascii", True) else json.dumps_notascii
+                    func = (
+                        json.dumps
+                        if self._json_flags.pop("ensure_ascii", True)
+                        else json.dumps_notascii
+                    )
                     res = func(res, **(self._json_flags or {}))
                 except Exception as ex:
                     # JSON serialization may fail, probably problem with data or error_data so pop it and try again
@@ -685,8 +689,12 @@ class APICall(DataContainer):
             cookies=self._result.cookies,
         )
 
-    def get_redacted_headers(self):
-        headers = self.headers.copy()
+    def get_redacted_headers(self, fields=None):
+        headers = (
+            {k: v for k, v in self._headers.items() if k in fields}
+            if fields
+            else self.headers
+        )
         if not self.requires_authorization or self.auth:
             # We won't log the authorization header if call shouldn't be authorized, or if it was successfully
             #  authorized. This means we'll only log authorization header for calls that failed to authorize (hopefully

apiserver/service_repo/auth/auth.py

@@ -30,24 +30,35 @@ def get_auth_func(auth_type):
     raise errors.unauthorized.BadAuthType()
 
 
-def authorize_token(jwt_token, *_, **__):
+def authorize_token(jwt_token, service, action, call):
     """Validate token against service/endpoint and requests data (dicts).
     Returns a parsed token object (auth payload)
     """
+    call_info = {"ip": call.real_ip}
+
+    def log_error(msg):
+        info = ", ".join(f"{k}={v}" for k, v in call_info.items())
+        log.error(f"{msg} Call info: {info}")
+
     try:
         return Token.from_encoded_token(jwt_token)
 
     except jwt.exceptions.InvalidKeyError as ex:
+        log_error("Failed parsing token.")
         raise errors.unauthorized.InvalidToken(
             "jwt invalid key error", reason=ex.args[0]
         )
     except jwt.InvalidTokenError as ex:
+        log_error("Failed parsing token.")
         raise errors.unauthorized.InvalidToken("invalid jwt token", reason=ex.args[0])
     except ValueError as ex:
-        log.exception("Failed while processing token: %s" % ex.args[0])
+        log_error(f"Failed while processing token: {str(ex.args[0])}.")
         raise errors.unauthorized.InvalidToken(
             "failed processing token", reason=ex.args[0]
         )
+    except Exception:
+        log_error("Failed processing token.")
+        raise
 
 
 def authorize_credentials(auth_data, service, action, call):

apiserver/service_repo/auth/payload/token.py

@@ -90,7 +90,7 @@ class Token(Payload):
             return token
         except Exception as e:
             raise errors.unauthorized.InvalidToken(
-                "failed parsing token, %s" % e.args[0]
+                "failed parsing token", reason=e.args[0]
             )
 
     @classmethod

apiserver/service_repo/service_repo.py

@@ -39,7 +39,7 @@ class ServiceRepo(object):
     """If the check is set, parsing will fail for endpoint request with the version that is grater than the current 
     maximum """
 
-    _max_version = PartialVersion("2.22")
+    _max_version = PartialVersion("2.27")
     """ Maximum version number (the highest min_version value across all endpoints) """
 
     _endpoint_exp = (

apiserver/service_repo/validators.py

@@ -17,7 +17,7 @@ log = config.logger(__file__)
 def validate_data(call: APICall, endpoint: Endpoint):
     """ Perform all required call/endpoint validation, update call result appropriately """
     try:
-        # todo: remove vaildate_required_fields once all endpoints have json schema
+        # todo: remove validate_required_fields once all endpoints have json schema
         validate_required_fields(endpoint, call)
 
         # set models. models will be validated automatically
@@ -50,10 +50,17 @@ def validate_role(endpoint, call):
         pass
 
 
-def validate_auth(endpoint, call):
-    """ Validate authorization for this endpoint and call.
-        If authentication has occurred, the call is updated with the authentication results.
+def validate_auth(endpoint: Endpoint, call: "APICall"):
     """
+    Validate authorization for this endpoint and call.
+    If authentication has occurred, the call is updated with the authentication results.
+    For the endpoints with authorize==False the validation is not performed to improve performance
+    For the endpoints with authorize==True the validation should pass otherwise exception will be thrown
+    For the endpoints with authorize==None the validation will be tried, but it does not have to succeed
+    """
+    if endpoint.authorize is not None and not endpoint.authorize:
+        return
+
     if not call.authorization:
         # No auth data. Invalid if we need to authorize and valid otherwise
         if endpoint.authorize:
@@ -63,10 +70,9 @@ def validate_auth(endpoint, call):
     # prepare arguments for validation
     service, _, action = endpoint.name.partition(".")
 
-    # If we have auth data, we'll try to validate anyway (just so we'll have auth-based permissions whenever possible,
-    # even if endpoint did not require authorization)
+    # noinspection PyBroadException
     try:
-        auth = call.authorization or ""
+        auth = call.authorization
         auth_type, _, auth_data = auth.partition(" ")
         authorize_func = get_auth_func(auth_type)
         call.auth = authorize_func(auth_data, service, action, call)
@@ -78,7 +84,7 @@ def validate_auth(endpoint, call):
 
 def validate_impersonation(endpoint, call):
     """ Validate impersonation headers and set impersonated identity and authorization data accordingly.
-        :returns True if impersonating, False otherwise
+        :return: True if impersonating, False otherwise
     """
     try:
         act_as = call.act_as

apiserver/services/debug.py

@@ -3,4 +3,7 @@ from apiserver.service_repo import APICall, endpoint
 
 @endpoint("debug.ping")
 def ping(call: APICall, _, __):
-    call.result.data = {"msg": "ClearML server"}
+    res = {"msg": "ClearML server"}
+    if call.data:
+        res.update(call.data)
+    call.result.data = res

apiserver/services/events.py

@@ -2,7 +2,7 @@ import itertools
 import math
 from collections import defaultdict
 from operator import itemgetter
-from typing import Sequence, Optional, Union, Tuple
+from typing import Sequence, Optional, Union, Tuple, Mapping
 
 import attr
 import jsonmodels.fields
@@ -27,10 +27,13 @@ from apiserver.apimodels.events import (
     ClearScrollRequest,
     ClearTaskLogRequest,
     SingleValueMetricsRequest,
-    GetVariantSampleRequest, GetMetricSamplesRequest,
+    GetVariantSampleRequest,
+    GetMetricSamplesRequest,
+    TaskMetric,
+    MultiTaskPlotsRequest,
 )
 from apiserver.bll.event import EventBLL
-from apiserver.bll.event.event_common import EventType, MetricVariants
+from apiserver.bll.event.event_common import EventType, MetricVariants, TaskCompanies
 from apiserver.bll.event.events_iterator import Scroll
 from apiserver.bll.event.scalar_key import ScalarKeyEnum, ScalarKey
 from apiserver.bll.model import ModelBLL
@@ -68,9 +71,11 @@ def _assert_task_or_model_exists(
 @endpoint("events.add")
 def add(call: APICall, company_id, _):
     data = call.data.copy()
-    allow_locked = data.pop("allow_locked", False)
     added, err_count, err_info = event_bll.add_events(
-        company_id, [data], call.worker, allow_locked=allow_locked
+        company_id=company_id,
+        user_id=call.identity.user,
+        events=[data],
+        worker=call.worker,
     )
     call.result.data = dict(added=added, errors=err_count, errors_info=err_info)
 
@@ -82,10 +87,10 @@ def add_batch(call: APICall, company_id, _):
         raise errors.bad_request.BatchContainsNoItems()
 
     added, err_count, err_info = event_bll.add_events(
-        company_id,
-        events,
-        call.worker,
-        allow_locked=events[0].get("allow_locked", False),
+        company_id=company_id,
+        user_id=call.identity.user,
+        events=events,
+        worker=call.worker,
     )
     call.result.data = dict(added=added, errors=err_count, errors_info=err_info)
 
@@ -255,7 +260,9 @@ def get_vector_metrics_and_variants(call, company_id, _):
     task_id = call.data["task"]
     model_events = call.data["model_events"]
     task_or_model = _assert_task_or_model_exists(
-        company_id, task_id, model_events=model_events,
+        company_id,
+        task_id,
+        model_events=model_events,
     )[0]
     call.result.data = dict(
         metrics=event_bll.get_metrics_and_variants(
@@ -269,7 +276,9 @@ def get_scalar_metrics_and_variants(call, company_id, _):
     task_id = call.data["task"]
     model_events = call.data["model_events"]
     task_or_model = _assert_task_or_model_exists(
-        company_id, task_id, model_events=model_events,
+        company_id,
+        task_id,
+        model_events=model_events,
     )[0]
     call.result.data = dict(
         metrics=event_bll.get_metrics_and_variants(
@@ -287,7 +296,9 @@ def vector_metrics_iter_histogram(call, company_id, _):
     task_id = call.data["task"]
     model_events = call.data["model_events"]
     task_or_model = _assert_task_or_model_exists(
-        company_id, task_id, model_events=model_events,
+        company_id,
+        task_id,
+        model_events=model_events,
     )[0]
     metric = call.data["metric"]
     variant = call.data["variant"]
@@ -320,7 +331,9 @@ def make_response(
 def get_task_events(_, company_id, request: TaskEventsRequest):
     task_id = request.task
     task_or_model = _assert_task_or_model_exists(
-        company_id, task_id, model_events=request.model_events,
+        company_id,
+        task_id,
+        model_events=request.model_events,
     )[0]
 
     key = ScalarKeyEnum.iter
@@ -353,7 +366,7 @@ def get_task_events(_, company_id, request: TaskEventsRequest):
         total = event_bll.events_iterator.count_task_events(
             event_type=request.event_type,
             company_id=task_or_model.get_index_company(),
-            task_id=task_id,
+            task_ids=[task_id],
             metric_variants=metric_variants,
         )
 
@@ -398,17 +411,18 @@ def get_scalar_metric_data(call, company_id, _):
     model_events = call.data.get("model_events", False)
 
     task_or_model = _assert_task_or_model_exists(
-        company_id, task_id, model_events=model_events,
+        company_id,
+        task_id,
+        model_events=model_events,
     )[0]
     result = event_bll.get_task_events(
         task_or_model.get_index_company(),
         task_id,
         event_type=EventType.metrics_scalar,
         sort=[{"iter": {"order": "desc"}}],
-        metric=metric,
+        metrics={metric: []},
         scroll_id=scroll_id,
         no_scroll=no_scroll,
-        model_events=model_events,
     )
 
     call.result.data = dict(
@@ -457,19 +471,23 @@ def scalar_metrics_iter_histogram(
         task_id=request.task,
         samples=request.samples,
         key=request.key,
+        metric_variants=_get_metric_variants_from_request(request.metrics),
     )
     call.result.data = metrics
 
 
-def _get_task_or_model_index_company(
-    company_id: str, task_ids: Sequence[str], model_events=False,
-) -> Tuple[str, Sequence[Task]]:
+def _get_task_or_model_index_companies(
+    company_id: str,
+    task_ids: Sequence[str],
+    model_events=False,
+) -> TaskCompanies:
     """
-    Verify that all tasks exists and belong to store data in the same company index
-    Return company and tasks
+    Returns lists of tasks grouped by company
     """
     tasks_or_models = _assert_task_or_model_exists(
-        company_id, task_ids, model_events=model_events,
+        company_id,
+        task_ids,
+        model_events=model_events,
     )
 
     unique_ids = set(task_ids)
@@ -482,13 +500,7 @@ def _get_task_or_model_index_company(
         )
         raise error_cls(company=company_id, ids=invalid)
 
-    companies = {t.get_index_company() for t in tasks_or_models}
-    if len(companies) > 1:
-        raise errors.bad_request.InvalidTaskId(
-            "only tasks from the same company are supported"
-        )
-
-    return companies.pop(), tasks_or_models
+    return bucketize(tasks_or_models, key=lambda t: t.get_index_company())
 
 
 @endpoint(
@@ -501,30 +513,44 @@ def multi_task_scalar_metrics_iter_histogram(
     task_ids = request.tasks
     if isinstance(task_ids, str):
         task_ids = [s.strip() for s in task_ids.split(",")]
-    company, tasks_or_models = _get_task_or_model_index_company(
-        company_id, task_ids, request.model_events
-    )
+
     call.result.data = dict(
         metrics=event_bll.metrics.compare_scalar_metrics_average_per_iter(
-            company_id=company,
-            tasks=tasks_or_models,
+            companies=_get_task_or_model_index_companies(
+                company_id, task_ids, request.model_events
+            ),
             samples=request.samples,
             key=request.key,
         )
     )
 
 
+def _get_single_value_metrics_response(
+    companies: TaskCompanies, value_metrics: Mapping[str, dict]
+) -> Sequence[dict]:
+    task_names = {
+        task.id: task.name for task in itertools.chain.from_iterable(companies.values())
+    }
+    return [
+        {"task": task_id, "task_name": task_names.get(task_id), "values": values}
+        for task_id, values in value_metrics.items()
+    ]
+
+
 @endpoint("events.get_task_single_value_metrics")
 def get_task_single_value_metrics(
     call, company_id: str, request: SingleValueMetricsRequest
 ):
-    company, tasks_or_models = _get_task_or_model_index_company(
+    companies = _get_task_or_model_index_companies(
         company_id, request.tasks, request.model_events
     )
-
-    res = event_bll.metrics.get_task_single_value_metrics(company, tasks_or_models)
     call.result.data = dict(
-        tasks=[{"task": task, "values": values} for task, values in res.items()]
+        tasks=_get_single_value_metrics_response(
+            companies=companies,
+            value_metrics=event_bll.metrics.get_task_single_value_metrics(
+                companies=companies
+            ),
+        )
     )
 
 
@@ -534,22 +560,23 @@ def get_multi_task_plots_v1_7(call, company_id, _):
     iters = call.data.get("iters", 1)
     scroll_id = call.data.get("scroll_id")
 
-    company, tasks_or_models = _get_task_or_model_index_company(company_id, task_ids)
+    companies = _get_task_or_model_index_companies(company_id, task_ids)
 
     # Get last 10K events by iteration and group them by unique metric+variant, returning top events for combination
     result = event_bll.get_task_events(
-        company,
-        task_ids,
+        company_id=list(companies),
+        task_id=task_ids,
         event_type=EventType.metrics_plot,
         sort=[{"iter": {"order": "desc"}}],
         size=10000,
         scroll_id=scroll_id,
     )
 
-    tasks = {t.id: t.name for t in tasks_or_models}
-
+    task_names = {
+        t.id: t.name for t in itertools.chain.from_iterable(companies.values())
+    }
     return_events = _get_top_iter_unique_events_per_task(
-        result.events, max_iters=iters, tasks=tasks
+        result.events, max_iters=iters, task_names=task_names
     )
 
     call.result.data = dict(
@@ -560,40 +587,54 @@ def get_multi_task_plots_v1_7(call, company_id, _):
     )
 
 
-@endpoint("events.get_multi_task_plots", min_version="1.8", required_fields=["tasks"])
-def get_multi_task_plots(call, company_id, _):
-    task_ids = call.data["tasks"]
-    iters = call.data.get("iters", 1)
-    scroll_id = call.data.get("scroll_id")
-    no_scroll = call.data.get("no_scroll", False)
-    model_events = call.data.get("model_events", False)
-
-    company, tasks_or_models = _get_task_or_model_index_company(
-        company_id, task_ids, model_events
-    )
-
+def _get_multitask_plots(
+    companies: TaskCompanies,
+    last_iters: int,
+    last_iters_per_task_metric: bool,
+    metrics: MetricVariants = None,
+    scroll_id=None,
+    no_scroll=True,
+) -> Tuple[dict, int, str]:
+    task_names = {
+        t.id: t.name for t in itertools.chain.from_iterable(companies.values())
+    }
     result = event_bll.get_task_events(
-        company,
-        task_ids,
+        company_id=list(companies),
+        task_id=list(task_names),
         event_type=EventType.metrics_plot,
+        metrics=metrics,
+        last_iter_count=last_iters,
         sort=[{"iter": {"order": "desc"}}],
-        last_iter_count=iters,
         scroll_id=scroll_id,
         no_scroll=no_scroll,
-        model_events=model_events,
+        size=config.get(
+            "services.events.events_retrieval.multi_plots_batch_size", 1000
+        ),
+        last_iters_per_task_metric=last_iters_per_task_metric,
     )
-
-    tasks = {t.id: t.name for t in tasks_or_models}
-
     return_events = _get_top_iter_unique_events_per_task(
-        result.events, max_iters=iters, tasks=tasks
+        result.events, max_iters=last_iters, task_names=task_names
     )
+    return return_events, result.total_events, result.next_scroll_id
 
+
+@endpoint("events.get_multi_task_plots", min_version="1.8")
+def get_multi_task_plots(call, company_id, request: MultiTaskPlotsRequest):
+    companies = _get_task_or_model_index_companies(
+        company_id, request.tasks, model_events=request.model_events
+    )
+    return_events, total_events, next_scroll_id = _get_multitask_plots(
+        companies=companies,
+        last_iters=request.iters,
+        scroll_id=request.scroll_id,
+        no_scroll=request.no_scroll,
+        last_iters_per_task_metric=request.last_iters_per_task_metric,
+    )
     call.result.data = dict(
         plots=return_events,
         returned=len(return_events),
-        total=result.total_events,
-        scroll_id=result.next_scroll_id,
+        total=total_events,
+        scroll_id=next_scroll_id,
     )
 
 
@@ -648,20 +689,15 @@ def _get_metric_variants_from_request(
 def get_task_plots(call, company_id, request: TaskPlotsRequest):
     task_id = request.task
     iters = request.iters
-    scroll_id = request.scroll_id
 
     task_or_model = _assert_task_or_model_exists(
         company_id, task_id, model_events=request.model_events
     )[0]
     result = event_bll.get_task_plots(
         task_or_model.get_index_company(),
-        tasks=[task_id],
-        sort=[{"iter": {"order": "desc"}}],
+        task_id=task_id,
         last_iterations_per_plot=iters,
-        scroll_id=scroll_id,
-        no_scroll=request.no_scroll,
         metric_variants=_get_metric_variants_from_request(request.metrics),
-        model_events=request.model_events,
     )
 
     return_events = result.events
@@ -674,24 +710,43 @@ def get_task_plots(call, company_id, request: TaskPlotsRequest):
     )
 
 
+def _task_metrics_dict_from_request(req_metrics: Sequence[TaskMetric]) -> dict:
+    task_metrics = defaultdict(dict)
+    for tm in req_metrics:
+        task_metrics[tm.task][tm.metric] = tm.variants
+    for metrics in task_metrics.values():
+        if None in metrics:
+            metrics.clear()
+
+    return task_metrics
+
+
+def _get_metrics_response(metric_events: Sequence[tuple]) -> Sequence[MetricEvents]:
+    return [
+        MetricEvents(
+            task=task,
+            iterations=[
+                IterationEvents(iter=iteration["iter"], events=iteration["events"])
+                for iteration in iterations
+            ],
+        )
+        for (task, iterations) in metric_events
+    ]
+
+
 @endpoint(
     "events.plots",
     request_data_model=MetricEventsRequest,
     response_data_model=MetricEventsResponse,
 )
 def task_plots(call, company_id, request: MetricEventsRequest):
-    task_metrics = defaultdict(dict)
-    for tm in request.metrics:
-        task_metrics[tm.task][tm.metric] = tm.variants
-    for metrics in task_metrics.values():
-        if None in metrics:
-            metrics.clear()
-
-    company, _ = _get_task_or_model_index_company(
-        company_id, task_ids=list(task_metrics), model_events=request.model_events
+    task_metrics = _task_metrics_dict_from_request(request.metrics)
+    task_ids = list(task_metrics)
+    task_or_models = _assert_task_or_model_exists(
+        company_id, task_ids=task_ids, model_events=request.model_events
     )
     result = event_bll.plots_iterator.get_task_events(
-        company_id=company,
+        companies={t.id: t.get_index_company() for t in task_or_models},
         task_metrics=task_metrics,
         iter_count=request.iters,
         navigate_earlier=request.navigate_earlier,
@@ -701,16 +756,7 @@ def task_plots(call, company_id, request: MetricEventsRequest):
 
     call.result.data_model = MetricEventsResponse(
         scroll_id=result.next_scroll_id,
-        metrics=[
-            MetricEvents(
-                task=task,
-                iterations=[
-                    IterationEvents(iter=iteration["iter"], events=iteration["events"])
-                    for iteration in iterations
-                ],
-            )
-            for (task, iterations) in result.metric_events
-        ],
+        metrics=_get_metrics_response(result.metric_events),
     )
 
 
@@ -759,7 +805,9 @@ def get_debug_images_v1_8(call, company_id, _):
     model_events = call.data.get("model_events", False)
 
     tasks_or_model = _assert_task_or_model_exists(
-        company_id, task_id, model_events=model_events,
+        company_id,
+        task_id,
+        model_events=model_events,
     )[0]
     result = event_bll.get_task_events(
         tasks_or_model.get_index_company(),
@@ -768,7 +816,6 @@ def get_debug_images_v1_8(call, company_id, _):
         sort=[{"iter": {"order": "desc"}}],
         last_iter_count=iters,
         scroll_id=scroll_id,
-        model_events=model_events,
     )
 
     return_events = result.events
@@ -789,19 +836,13 @@ def get_debug_images_v1_8(call, company_id, _):
     response_data_model=MetricEventsResponse,
 )
 def get_debug_images(call, company_id, request: MetricEventsRequest):
-    task_metrics = defaultdict(dict)
-    for tm in request.metrics:
-        task_metrics[tm.task][tm.metric] = tm.variants
-    for metrics in task_metrics.values():
-        if None in metrics:
-            metrics.clear()
-
-    company, _ = _get_task_or_model_index_company(
-        company_id, task_ids=list(task_metrics), model_events=request.model_events
+    task_metrics = _task_metrics_dict_from_request(request.metrics)
+    task_ids = list(task_metrics)
+    task_or_models = _assert_task_or_model_exists(
+        company_id, task_ids=task_ids, model_events=request.model_events
     )
-
     result = event_bll.debug_images_iterator.get_task_events(
-        company_id=company,
+        companies={t.id: t.get_index_company() for t in task_or_models},
         task_metrics=task_metrics,
         iter_count=request.iters,
         navigate_earlier=request.navigate_earlier,
@@ -811,16 +852,7 @@ def get_debug_images(call, company_id, request: MetricEventsRequest):
 
     call.result.data_model = MetricEventsResponse(
         scroll_id=result.next_scroll_id,
-        metrics=[
-            MetricEvents(
-                task=task,
-                iterations=[
-                    IterationEvents(iter=iteration["iter"], events=iteration["events"])
-                    for iteration in iterations
-                ],
-            )
-            for (task, iterations) in result.metric_events
-        ],
+        metrics=_get_metrics_response(result.metric_events),
     )
 
 
@@ -831,7 +863,9 @@ def get_debug_images(call, company_id, request: MetricEventsRequest):
 )
 def get_debug_image_sample(call, company_id, request: GetVariantSampleRequest):
     task_or_model = _assert_task_or_model_exists(
-        company_id, request.task, model_events=request.model_events,
+        company_id,
+        request.task,
+        model_events=request.model_events,
     )[0]
     res = event_bll.debug_image_sample_history.get_sample_for_variant(
         company_id=task_or_model.get_index_company(),
@@ -853,7 +887,9 @@ def get_debug_image_sample(call, company_id, request: GetVariantSampleRequest):
 )
 def next_debug_image_sample(call, company_id, request: NextHistorySampleRequest):
     task_or_model = _assert_task_or_model_exists(
-        company_id, request.task, model_events=request.model_events,
+        company_id,
+        request.task,
+        model_events=request.model_events,
     )[0]
     res = event_bll.debug_image_sample_history.get_next_sample(
         company_id=task_or_model.get_index_company(),
@@ -866,11 +902,14 @@ def next_debug_image_sample(call, company_id, request: NextHistorySampleRequest)
 
 
 @endpoint(
-    "events.get_plot_sample", request_data_model=GetMetricSamplesRequest,
+    "events.get_plot_sample",
+    request_data_model=GetMetricSamplesRequest,
 )
 def get_plot_sample(call, company_id, request: GetMetricSamplesRequest):
     task_or_model = _assert_task_or_model_exists(
-        company_id, request.task, model_events=request.model_events,
+        company_id,
+        request.task,
+        model_events=request.model_events,
     )[0]
     res = event_bll.plot_sample_history.get_samples_for_metric(
         company_id=task_or_model.get_index_company(),
@@ -885,11 +924,14 @@ def get_plot_sample(call, company_id, request: GetMetricSamplesRequest):
 
 
 @endpoint(
-    "events.next_plot_sample", request_data_model=NextHistorySampleRequest,
+    "events.next_plot_sample",
+    request_data_model=NextHistorySampleRequest,
 )
 def next_plot_sample(call, company_id, request: NextHistorySampleRequest):
     task_or_model = _assert_task_or_model_exists(
-        company_id, request.task, model_events=request.model_events,
+        company_id,
+        request.task,
+        model_events=request.model_events,
     )[0]
     res = event_bll.plot_sample_history.get_next_sample(
         company_id=task_or_model.get_index_company(),
@@ -903,11 +945,15 @@ def next_plot_sample(call, company_id, request: NextHistorySampleRequest):
 
 @endpoint("events.get_task_metrics", request_data_model=TaskMetricsRequest)
 def get_task_metrics(call: APICall, company_id, request: TaskMetricsRequest):
-    company, _ = _get_task_or_model_index_company(
-        company_id, request.tasks, model_events=request.model_events
+    task_or_models = _assert_task_or_model_exists(
+        company_id,
+        request.tasks,
+        model_events=request.model_events,
     )
     res = event_bll.metrics.get_task_metrics(
-        company, task_ids=request.tasks, event_type=request.event_type
+        task_or_models[0].get_index_company(),
+        task_ids=request.tasks,
+        event_type=request.event_type,
     )
     call.result.data = {
         "metrics": [{"task": task, "metrics": metrics} for (task, metrics) in res]
@@ -955,20 +1001,22 @@ def clear_task_log(call: APICall, company_id: str, request: ClearTaskLogRequest)
     )
 
 
-def _get_top_iter_unique_events_per_task(events, max_iters, tasks):
-    key = itemgetter("metric", "variant", "task", "iter")
-
+def _get_top_iter_unique_events_per_task(
+    events, max_iters: int, task_names: Mapping[str, str]
+):
+    key_fields = ("metric", "variant", "task")
     unique_events = itertools.chain.from_iterable(
         itertools.islice(group, max_iters)
         for _, group in itertools.groupby(
-            sorted(events, key=key, reverse=True), key=key
+            sorted(events, key=itemgetter(*(key_fields + ("iter",))), reverse=True),
+            key=itemgetter(*key_fields),
         )
     )
 
     def collect(evs, fields):
         if not fields:
             evs = list(evs)
-            return {"name": tasks.get(evs[0].get("task")), "plots": evs}
+            return {"name": task_names.get(evs[0].get("task")), "plots": evs}
         return {
             str(k): collect(group, fields[1:])
             for k, group in itertools.groupby(evs, key=itemgetter(fields[0]))
@@ -1034,14 +1082,16 @@ def scalar_metrics_iter_raw(
         request.batch_size = request.batch_size or scroll.request.batch_size
 
     task_id = request.task
-    task_or_model = _assert_task_or_model_exists(company_id, task_id, model_events=request.model_events)[0]
+    task_or_model = _assert_task_or_model_exists(
+        company_id, task_id, model_events=request.model_events
+    )[0]
     metric_variants = _get_metric_variants_from_request([request.metric])
 
     if request.count_total and total is None:
         total = event_bll.events_iterator.count_task_events(
             event_type=EventType.metrics_scalar,
             company_id=task_or_model.get_index_company(),
-            task_id=task_id,
+            task_ids=[task_id],
             metric_variants=metric_variants,
         )
 

apiserver/services/models.py

@@ -1,6 +1,6 @@
 from datetime import datetime
 from functools import partial
-from typing import Sequence
+from typing import Sequence, Union
 
 from mongoengine import Q, EmbeddedDocument
 
@@ -22,6 +22,7 @@ from apiserver.apimodels.models import (
     ModelsDeleteManyRequest,
     ModelsGetRequest,
 )
+from apiserver.apimodels.tasks import UpdateTagsRequest
 from apiserver.bll.model import ModelBLL, Metadata
 from apiserver.bll.organization import OrgBLL, Tags
 from apiserver.bll.project import ProjectBLL
@@ -59,23 +60,28 @@ org_bll = OrgBLL()
 project_bll = ProjectBLL()
 
 
+def conform_model_data(call: APICall, model_data: Union[Sequence[dict], dict]):
+    conform_output_tags(call, model_data)
+    unescape_metadata(call, model_data)
+
+
 @endpoint("models.get_by_id", required_fields=["model"])
 def get_by_id(call: APICall, company_id, _):
     model_id = call.data["model"]
-
-    Metadata.escape_query_parameters(call)
+    call_data = Metadata.escape_query_parameters(call.data)
     models = Model.get_many(
         company=company_id,
-        query_dict=call.data,
+        query_dict=call_data,
         query=Q(id=model_id),
         allow_public=True,
     )
     if not models:
         raise errors.bad_request.InvalidModelId(
-            "no such public or company model", id=model_id, company=company_id,
+            "no such public or company model",
+            id=model_id,
+            company=company_id,
         )
-    conform_output_tags(call, models[0])
-    unescape_metadata(call, models[0])
+    conform_model_data(call, models[0])
     call.result.data = {"model": models[0]}
 
 
@@ -99,35 +105,38 @@ def get_by_task_id(call: APICall, company_id, _):
     ).first()
     if not model:
         raise errors.bad_request.InvalidModelId(
-            "no such public or company model", id=model_id, company=company_id,
+            "no such public or company model",
+            id=model_id,
+            company=company_id,
         )
     model_dict = model.to_proper_dict()
-    conform_output_tags(call, model_dict)
-    unescape_metadata(call, model_dict)
+    conform_model_data(call, model_dict)
     call.result.data = {"model": model_dict}
 
 
 @endpoint("models.get_all_ex", request_data_model=ModelsGetRequest)
 def get_all_ex(call: APICall, company_id, request: ModelsGetRequest):
     conform_tag_fields(call, call.data)
-    process_include_subprojects(call.data)
-    Metadata.escape_query_parameters(call)
+    call_data = Metadata.escape_query_parameters(call.data)
+    process_include_subprojects(call_data)
     ret_params = {}
     models = Model.get_many_with_join(
         company=company_id,
-        query_dict=call.data,
-        allow_public=True,
+        query_dict=call_data,
+        allow_public=request.allow_public,
         ret_params=ret_params,
     )
-    conform_output_tags(call, models)
-    unescape_metadata(call, models)
+    conform_model_data(call, models)
 
     if not request.include_stats:
         call.result.data = {"models": models, **ret_params}
         return
 
     model_ids = {model["id"] for model in models}
-    stats = ModelBLL.get_model_stats(company=company_id, model_ids=list(model_ids),)
+    stats = ModelBLL.get_model_stats(
+        company=company_id,
+        model_ids=list(model_ids),
+    )
 
     for model in models:
         model["stats"] = stats.get(model["id"])
@@ -138,29 +147,28 @@ def get_all_ex(call: APICall, company_id, request: ModelsGetRequest):
 @endpoint("models.get_by_id_ex", required_fields=["id"])
 def get_by_id_ex(call: APICall, company_id, _):
     conform_tag_fields(call, call.data)
-    Metadata.escape_query_parameters(call)
+    call_data = Metadata.escape_query_parameters(call.data)
     models = Model.get_many_with_join(
-        company=company_id, query_dict=call.data, allow_public=True
+        company=company_id, query_dict=call_data, allow_public=True
     )
-    conform_output_tags(call, models)
-    unescape_metadata(call, models)
+    conform_model_data(call, models)
     call.result.data = {"models": models}
 
 
 @endpoint("models.get_all", required_fields=[])
 def get_all(call: APICall, company_id, _):
     conform_tag_fields(call, call.data)
-    Metadata.escape_query_parameters(call)
+    call_data = Metadata.escape_query_parameters(call.data)
+    process_include_subprojects(call_data)
     ret_params = {}
     models = Model.get_many(
         company=company_id,
-        parameters=call.data,
-        query_dict=call.data,
+        parameters=call_data,
+        query_dict=call_data,
         allow_public=True,
         ret_params=ret_params,
     )
-    conform_output_tags(call, models)
-    unescape_metadata(call, models)
+    conform_model_data(call, models)
     call.result.data = {"models": models, **ret_params}
 
 
@@ -212,7 +220,7 @@ def _update_cached_tags(company: str, project: str, fields: dict):
     org_bll.update_tags(
         company,
         Tags.Model,
-        project=project,
+        projects=[project],
         tags=fields.get("tags"),
         system_tags=fields.get("system_tags"),
     )
@@ -220,7 +228,9 @@ def _update_cached_tags(company: str, project: str, fields: dict):
 
 def _reset_cached_tags(company: str, projects: Sequence[str]):
     org_bll.reset_tags(
-        company, Tags.Model, projects=projects,
+        company,
+        Tags.Model,
+        projects=projects,
     )
 
 
@@ -283,6 +293,8 @@ def update_for_task(call: APICall, company_id, _):
             id=database.utils.id(),
             created=now,
             last_update=now,
+            last_change=now,
+            last_changed_by=call.identity.user,
             user=call.identity.user,
             company=company_id,
             project=task.project,
@@ -301,6 +313,7 @@ def update_for_task(call: APICall, company_id, _):
     TaskBLL.update_statistics(
         task_id=task_id,
         company_id=company_id,
+        user_id=call.identity.user,
         last_iteration_max=iteration,
         models__output=[
             ModelItem(
@@ -320,7 +333,6 @@ def update_for_task(call: APICall, company_id, _):
     response_data_model=CreateModelResponse,
 )
 def create(call: APICall, company_id, req_model: CreateModelRequest):
-
     if req_model.public:
         company_id = ""
 
@@ -345,6 +357,8 @@ def create(call: APICall, company_id, req_model: CreateModelRequest):
         company=company_id,
         created=now,
         last_update=now,
+        last_change=now,
+        last_changed_by=call.identity.user,
         **fields,
     )
     model.save()
@@ -414,12 +428,20 @@ def edit(call: APICall, company_id, _):
     task_id = model.task or fields.get("task")
     if task_id and iteration is not None:
         TaskBLL.update_statistics(
-            task_id=task_id, company_id=company_id, last_iteration_max=iteration,
+            task_id=task_id,
+            company_id=company_id,
+            user_id=call.identity.user,
+            last_iteration_max=iteration,
         )
 
     if fields:
+        now = datetime.utcnow()
+        fields.update(
+            last_change=now,
+            last_changed_by=call.identity.user,
+        )
         if any(uf in fields for uf in last_update_fields):
-            fields.update(last_update=datetime.utcnow())
+            fields.update(last_update=now)
 
         updated = model.update(upsert=False, **fields)
         if updated:
@@ -428,8 +450,7 @@ def edit(call: APICall, company_id, _):
                 _reset_cached_tags(company_id, projects=[new_project, model.project])
             else:
                 _update_cached_tags(company_id, project=model.project, fields=fields)
-        conform_output_tags(call, fields)
-        unescape_metadata(call, fields)
+        conform_model_data(call, fields)
         call.result.data_model = UpdateResponse(updated=updated, fields=fields)
     else:
         call.result.data_model = UpdateResponse(updated=0)
@@ -446,13 +467,25 @@ def _update_model(call: APICall, company_id, model_id=None):
     iteration = data.get("iteration")
     if task_id and iteration is not None:
         TaskBLL.update_statistics(
-            task_id=task_id, company_id=company_id, last_iteration_max=iteration,
+            task_id=task_id,
+            company_id=company_id,
+            user_id=call.identity.user,
+            last_iteration_max=iteration,
         )
 
-    updated_count, updated_fields = Model.safe_update(company_id, model.id, data)
+    now = datetime.utcnow()
+    updated_count, updated_fields = Model.safe_update(
+        company_id,
+        model.id,
+        data,
+        injected_update=dict(
+            last_change=now,
+            last_changed_by=call.identity.user,
+        ),
+    )
     if updated_count:
         if any(uf in updated_fields for uf in last_update_fields):
-            model.update(upsert=False, last_update=datetime.utcnow())
+            model.update(upsert=False, last_update=now)
 
         new_project = updated_fields.get("project", model.project)
         if new_project != model.project:
@@ -461,8 +494,7 @@ def _update_model(call: APICall, company_id, model_id=None):
             _update_cached_tags(
                 company_id, project=model.project, fields=updated_fields
             )
-    conform_output_tags(call, updated_fields)
-    unescape_metadata(call, updated_fields)
+    conform_model_data(call, updated_fields)
     return UpdateResponse(updated=updated_count, fields=updated_fields)
 
 
@@ -482,6 +514,7 @@ def set_ready(call: APICall, company_id: str, request: PublishModelRequest):
     updated, published_task = ModelBLL.publish_model(
         model_id=request.model,
         company_id=company_id,
+        user_id=call.identity.user,
         force_publish_task=request.force_publish_task,
         publish_task_func=publish_task if request.publish_task else None,
     )
@@ -500,6 +533,7 @@ def publish_many(call: APICall, company_id, request: ModelsPublishManyRequest):
         func=partial(
             ModelBLL.publish_model,
             company_id=company_id,
+            user_id=call.identity.user,
             force_publish_task=request.force_publish_task,
             publish_task_func=publish_task if request.publish_task else None,
         ),
@@ -522,7 +556,11 @@ def publish_many(call: APICall, company_id, request: ModelsPublishManyRequest):
 @endpoint("models.delete", request_data_model=DeleteModelRequest)
 def delete(call: APICall, company_id, request: DeleteModelRequest):
     del_count, model = ModelBLL.delete_model(
-        model_id=request.model, company_id=company_id, force=request.force
+        model_id=request.model,
+        company_id=company_id,
+        user_id=call.identity.user,
+        force=request.force,
+        delete_external_artifacts=request.delete_external_artifacts,
     )
     if del_count:
         _reset_cached_tags(
@@ -539,7 +577,13 @@ def delete(call: APICall, company_id, request: DeleteModelRequest):
 )
 def delete(call: APICall, company_id, request: ModelsDeleteManyRequest):
     results, failures = run_batch_operation(
-        func=partial(ModelBLL.delete_model, company_id=company_id, force=request.force),
+        func=partial(
+            ModelBLL.delete_model,
+            company_id=company_id,
+            user_id=call.identity.user,
+            force=request.force,
+            delete_external_artifacts=request.delete_external_artifacts,
+        ),
         ids=request.ids,
     )
 
@@ -563,7 +607,10 @@ def delete(call: APICall, company_id, request: ModelsDeleteManyRequest):
 )
 def archive_many(call: APICall, company_id, request: BatchRequest):
     results, failures = run_batch_operation(
-        func=partial(ModelBLL.archive_model, company_id=company_id), ids=request.ids,
+        func=partial(
+            ModelBLL.archive_model, company_id=company_id, user_id=call.identity.user
+        ),
+        ids=request.ids,
     )
     call.result.data_model = BatchResponse(
         succeeded=[dict(id=_id, archived=bool(archived)) for _id, archived in results],
@@ -578,7 +625,8 @@ def archive_many(call: APICall, company_id, request: BatchRequest):
 )
 def unarchive_many(call: APICall, company_id, request: BatchRequest):
     results, failures = run_batch_operation(
-        func=partial(ModelBLL.unarchive_model, company_id=company_id), ids=request.ids,
+        func=partial(ModelBLL.unarchive_model, company_id=company_id, user_id=call.identity.user),
+        ids=request.ids,
     )
     call.result.data_model = BatchResponse(
         succeeded=[
@@ -591,7 +639,11 @@ def unarchive_many(call: APICall, company_id, request: BatchRequest):
 @endpoint("models.make_public", min_version="2.9", request_data_model=MakePublicRequest)
 def make_public(call: APICall, company_id, request: MakePublicRequest):
     call.result.data = Model.set_public(
-        company_id, ids=request.ids, invalid_cls=InvalidModelId, enabled=True
+        company_id=company_id,
+        user_id=call.identity.user,
+        ids=request.ids,
+        invalid_cls=InvalidModelId,
+        enabled=True,
     )
 
 
@@ -600,13 +652,17 @@ def make_public(call: APICall, company_id, request: MakePublicRequest):
 )
 def make_public(call: APICall, company_id, request: MakePublicRequest):
     call.result.data = Model.set_public(
-        company_id, request.ids, invalid_cls=InvalidModelId, enabled=False
+        company_id=company_id,
+        user_id=call.identity.user,
+        ids=request.ids,
+        invalid_cls=InvalidModelId,
+        enabled=False,
     )
 
 
 @endpoint("models.move", request_data_model=MoveRequest)
 def move(call: APICall, company_id: str, request: MoveRequest):
-    if not (request.project or request.project_name):
+    if not ("project" in call.data or request.project_name):
         raise errors.bad_request.MissingRequiredFields(
             "project or project_name is required"
         )
@@ -623,30 +679,51 @@ def move(call: APICall, company_id: str, request: MoveRequest):
     }
 
 
+@endpoint("models.update_tags")
+def update_tags(_, company_id: str, request: UpdateTagsRequest):
+    return {
+        "updated": org_bll.edit_entity_tags(
+            company_id=company_id,
+            entity_cls=Model,
+            entity_ids=request.ids,
+            add_tags=request.add_tags,
+            remove_tags=request.remove_tags,
+        )
+    }
+
+
 @endpoint("models.add_or_update_metadata", min_version="2.13")
 def add_or_update_metadata(
-    _: APICall, company_id: str, request: AddOrUpdateMetadataRequest
+    call: APICall, company_id: str, request: AddOrUpdateMetadataRequest
 ):
     model_id = request.model
     model = ModelBLL.get_company_model_by_id(company_id=company_id, model_id=model_id)
+    now = datetime.utcnow()
     return {
         "updated": Metadata.edit_metadata(
             model,
             items=request.metadata,
             replace_metadata=request.replace_metadata,
-            last_update=datetime.utcnow(),
+            last_update=now,
+            last_change=now,
+            last_changed_by=call.identity.user,
         )
     }
 
 
 @endpoint("models.delete_metadata", min_version="2.13")
-def delete_metadata(_: APICall, company_id: str, request: DeleteMetadataRequest):
+def delete_metadata(call: APICall, company_id: str, request: DeleteMetadataRequest):
     model_id = request.model
     model = ModelBLL.get_company_model_by_id(
         company_id=company_id, model_id=model_id, only_fields=("id",)
     )
+    now = datetime.utcnow()
     return {
         "updated": Metadata.delete_metadata(
-            model, keys=request.keys, last_update=datetime.utcnow()
+            model,
+            keys=request.keys,
+            last_update=now,
+            last_change=now,
+            last_changed_by=call.identity.user,
         )
     }

apiserver/services/organization.py

@@ -1,21 +1,45 @@
+import csv
 from collections import defaultdict
+from concurrent.futures import ThreadPoolExecutor
+from io import StringIO
 from operator import itemgetter
-from typing import Mapping, Type
+from typing import Mapping, Type, Sequence, Optional, Callable, Hashable
 
+from flask import stream_with_context
 from mongoengine import Q
 
-from apiserver.apimodels.organization import TagsRequest, EntitiesCountRequest
+from apiserver.apierrors import errors
+from apiserver.apimodels.organization import (
+    TagsRequest,
+    EntitiesCountRequest,
+    DownloadForGetAllRequest,
+    EntityType,
+    PrepareDownloadForGetAllRequest,
+)
+from apiserver.bll.model import Metadata
 from apiserver.bll.organization import OrgBLL, Tags
 from apiserver.bll.project import ProjectBLL
+from apiserver.config_repo import config
 from apiserver.database.model import User, AttributedDocument, EntityVisibility
 from apiserver.database.model.model import Model
 from apiserver.database.model.project import Project
-from apiserver.database.model.task.task import Task
+from apiserver.database.model.task.task import Task, TaskType
+from apiserver.redis_manager import redman
 from apiserver.service_repo import endpoint, APICall
+from apiserver.services.models import conform_model_data
+from apiserver.services.tasks import (
+    escape_execution_parameters,
+    _hidden_query,
+    conform_task_data,
+)
 from apiserver.services.utils import get_tags_filter_dictionary, sort_tags_response
+from apiserver.utilities import json
+from apiserver.utilities.dicts import nested_get
 
 org_bll = OrgBLL()
 project_bll = ProjectBLL()
+redis = redman.connection("apiserver")
+conf = config.get("services.organization")
 
 
 @endpoint("organization.get_tags", request_data_model=TagsRequest)
@@ -24,7 +48,10 @@ def get_tags(call: APICall, company, request: TagsRequest):
     ret = defaultdict(set)
     for entity in Tags.Model, Tags.Task:
         tags = org_bll.get_tags(
-            company, entity, include_system=request.include_system, filter_=filter_dict,
+            company,
+            entity,
+            include_system=request.include_system,
+            filter_=filter_dict,
         )
         for field, vals in tags.items():
             ret[field] |= vals
@@ -59,6 +86,7 @@ def get_entities_count(call: APICall, company, request: EntitiesCountRequest):
         "models": Model,
         "pipelines": Project,
         "datasets": Project,
+        "reports": Task,
     }
     ret = {}
     for field, entity_cls in entity_classes.items():
@@ -66,16 +94,20 @@ def get_entities_count(call: APICall, company, request: EntitiesCountRequest):
         if data is None:
             continue
 
+        if field == "reports":
+            data["type"] = TaskType.report
+            data["include_subprojects"] = True
+
         if request.active_users:
             if entity_cls is Project:
                 requested_ids = data.get("id")
                 if isinstance(requested_ids, str):
                     requested_ids = [requested_ids]
-                ids, _ = project_bll.get_projects_with_active_user(
+                ids, _ = project_bll.get_projects_with_selected_children(
                     company=company,
                     users=request.active_users,
                     project_ids=requested_ids,
-                    allow_public=True,
+                    allow_public=request.allow_public,
                 )
                 if not ids:
                     ret[field] = 0
@@ -85,11 +117,218 @@ def get_entities_count(call: APICall, company, request: EntitiesCountRequest):
                 data["user"] = request.active_users
 
         query = Q()
-        if entity_cls in (Project, Task) and not request.search_hidden:
+        if (
+            entity_cls in (Project, Task)
+            and field not in ("reports", "pipelines", "datasets")
+            and not request.search_hidden
+        ):
             query &= Q(system_tags__ne=EntityVisibility.hidden.value)
 
         ret[field] = entity_cls.get_count(
-            company=company, query_dict=data, query=query, allow_public=True,
+            company=company,
+            query_dict=data,
+            query=query,
+            allow_public=request.allow_public,
         )
 
     call.result.data = ret
+
+
+def _get_download_getter_fn(
+    company: str,
+    call: APICall,
+    call_data: dict,
+    allow_public: bool,
+    entity_type: EntityType,
+) -> Optional[Callable[[int, int], Sequence[dict]]]:
+    def get_task_data() -> Sequence[dict]:
+        tasks = Task.get_many_with_join(
+            company=company,
+            query_dict=call_data,
+            query=_hidden_query(call_data),
+            allow_public=allow_public,
+        )
+        conform_task_data(call, tasks)
+        return tasks
+
+    def get_model_data() -> Sequence[dict]:
+        models = Model.get_many_with_join(
+            company=company,
+            query_dict=call_data,
+            allow_public=allow_public,
+        )
+        conform_model_data(call, models)
+        return models
+
+    if entity_type == EntityType.task:
+        call_data = escape_execution_parameters(call_data)
+        get_fn = get_task_data
+    elif entity_type == EntityType.model:
+        call_data = Metadata.escape_query_parameters(call_data)
+        get_fn = get_model_data
+    else:
+        raise errors.bad_request.ValidationError(
+            f"Unsupported entity type: {str(entity_type)}"
+        )
+
+    def getter(page: int, page_size: int) -> Sequence[dict]:
+        call_data.pop("scroll_id", None)
+        call_data.pop("start", None)
+        call_data.pop("size", None)
+        call_data.pop("refresh_scroll", None)
+        call_data["page"] = page
+        call_data["page_size"] = page_size
+        return get_fn()
+
+    return getter
+
+
+@endpoint("organization.prepare_download_for_get_all")
+def prepare_download_for_get_all(
+    call: APICall, company: str, request: PrepareDownloadForGetAllRequest
+):
+    # validate input params
+    field_names = set()
+    for fm in request.field_mappings:
+        name = fm.name or fm.field
+        if name in field_names:
+            raise errors.bad_request.ValidationError(
+                f"Field_name appears more than once in field_mappings: {str(name)}"
+            )
+        field_names.add(name)
+        if fm.values:
+            value_keys = set()
+            for v in fm.values:
+                if v.key in value_keys:
+                    raise errors.bad_request.ValidationError(
+                        f"Value key appears more than once in field_mappings: {str(v.key)}"
+                    )
+                value_keys.add(v.key)
+
+    getter = _get_download_getter_fn(
+        company,
+        call,
+        call_data=call.data.copy(),
+        allow_public=request.allow_public,
+        entity_type=request.entity_type,
+    )
+    # retrieve one element just to make sure that there are no issues with the call parameters
+    if getter:
+        getter(0, 1)
+
+    redis.setex(
+        f"get_all_download_{call.id}",
+        int(conf.get("download.redis_timeout_sec", 300)),
+        json.dumps(call.data),
+    )
+
+    call.result.data = dict(prepare_id=call.id)
+
+
+@endpoint("organization.download_for_get_all")
+def download_for_get_all(call: APICall, company, request: DownloadForGetAllRequest):
+    request_data = redis.get(f"get_all_download_{request.prepare_id}")
+    if not request_data:
+        raise errors.bad_request.InvalidId(
+            f"prepare ID not found", prepare_id=request.prepare_id
+        )
+
+    try:
+        call_data = json.loads(request_data)
+        request = PrepareDownloadForGetAllRequest(**call_data)
+    except Exception as ex:
+        raise errors.server_error.DataError("failed parsing prepared data", ex=ex)
+
+    class SingleLine:
+        @staticmethod
+        def write(line: str) -> str:
+            return line
+
+    def generate():
+        field_mappings = {
+            mapping.get("name", mapping["field"]): {
+                "field_path": mapping["field"].split("."),
+                "values": {
+                    v.get("key"): v.get("value")
+                    for v in (mapping.get("values") or [])
+                },
+            }
+            for mapping in call_data.get("field_mappings", [])
+        }
+        get_fn = _get_download_getter_fn(
+            company,
+            call,
+            call_data=call_data,
+            allow_public=request.allow_public,
+            entity_type=request.entity_type,
+        )
+        if not get_fn:
+            yield csv.writer(SingleLine()).writerow(field_mappings)
+            return
+
+        def get_entity_field_as_str(
+            data: dict, field_path: Sequence[str], values: Mapping
+        ) -> str:
+            val = nested_get(data, field_path, "")
+            if isinstance(val, dict):
+                val = val.get("id", "")
+            if values and isinstance(val, Hashable):
+                val = values.get(val, val)
+
+            return str(val)
+
+        def get_projected_fields(data: dict) -> Sequence[str]:
+            return [
+                get_entity_field_as_str(
+                    data, field_path=m["field_path"], values=m["values"]
+                )
+                for m in field_mappings.values()
+            ]
+
+        with ThreadPoolExecutor(1) as pool:
+            page = 0
+            page_size = int(conf.get("download.batch_size", 500))
+            items_left = int(conf.get("download.max_download_items", 1000))
+            future = pool.submit(get_fn, page, min(page_size, items_left))
+            while items_left > 0:
+                result = future.result()
+                if not result:
+                    break
+
+                items_left -= len(result)
+                page += 1
+                if items_left > 0:
+                    future = pool.submit(get_fn, page, min(page_size, items_left))
+
+                with StringIO() as fp:
+                    writer = csv.writer(fp)
+                    if page == 1:
+                        fp.write("\ufeff")  # utf-8 signature
+                        writer.writerow(field_mappings)
+                    writer.writerows(get_projected_fields(r) for r in result)
+                    yield fp.getvalue()
+
+        if page == 0:
+            yield csv.writer(SingleLine()).writerow(field_mappings)
+
+    def get_project_name() -> Optional[str]:
+        projects = call_data.get("project")
+        if not projects or not isinstance(projects, (list, str)):
+            return
+        if isinstance(projects, list):
+            if len(projects) > 1:
+                return
+            projects = projects[0]
+            if projects is None:
+                return "root"
+        project: Project = Project.objects(id=projects).only("basename").first()
+        if not project:
+            return
+
+        return project.basename[: conf.get("download.max_project_name_length", 60)]
+
+    call.result.filename = "-".join(
+        filter(None, ("clearml", get_project_name(), f"{request.entity_type}s.csv"))
+    )
+    call.result.content_type = "text/csv"
+    call.result.raw_data = stream_with_context(generate())

apiserver/services/pipelines.py

@@ -1,12 +1,21 @@
 import re
+from functools import partial
 
-from apiserver.apimodels.pipelines import StartPipelineResponse, StartPipelineRequest
+import attr
+
+from apiserver.apierrors.errors.bad_request import CannotRemoveAllRuns
+from apiserver.apimodels.pipelines import (
+    StartPipelineResponse,
+    StartPipelineRequest,
+    DeleteRunsRequest,
+)
 from apiserver.bll.organization import OrgBLL
 from apiserver.bll.project import ProjectBLL
 from apiserver.bll.task import TaskBLL
-from apiserver.bll.task.task_operations import enqueue_task
+from apiserver.bll.task.task_operations import enqueue_task, delete_task
+from apiserver.bll.util import run_batch_operation
 from apiserver.database.model.project import Project
-from apiserver.database.model.task.task import Task
+from apiserver.database.model.task.task import Task, TaskType
 from apiserver.service_repo import APICall, endpoint
 
 org_bll = OrgBLL()
@@ -31,6 +40,45 @@ def _update_task_name(task: Task):
     task.update(name=new_name)
 
 
+@endpoint("pipelines.delete_runs")
+def delete_runs(call: APICall, company_id: str, request: DeleteRunsRequest):
+    existing_runs = set(
+        Task.objects(project=request.project, type=TaskType.controller).scalar("id")
+    )
+    if not existing_runs.difference(request.ids):
+        raise CannotRemoveAllRuns(project=request.project)
+
+    # make sure that only controller tasks are deleted
+    ids = existing_runs.intersection(request.ids)
+    if not ids:
+        return dict(succeeded=[], failed=[])
+
+    results, failures = run_batch_operation(
+        func=partial(
+            delete_task,
+            company_id=company_id,
+            user_id=call.identity.user,
+            move_to_trash=False,
+            force=True,
+            return_file_urls=False,
+            delete_output_models=True,
+            status_message="",
+            status_reason="Pipeline run deleted",
+            delete_external_artifacts=True,
+        ),
+        ids=list(ids),
+    )
+
+    succeeded = []
+    if results:
+        for _id, (deleted, task, cleanup_res) in results:
+            succeeded.append(
+                dict(id=_id, deleted=bool(deleted), **attr.asdict(cleanup_res))
+            )
+
+    call.result.data = dict(succeeded=succeeded, failed=failures)
+
+
 @endpoint(
     "pipelines.start_pipeline", response_data_model=StartPipelineResponse,
 )
@@ -60,6 +108,7 @@ def start_pipeline(call: APICall, company_id: str, request: StartPipelineRequest
     queued, res = enqueue_task(
         task_id=task.id,
         company_id=company_id,
+        user_id=call.identity.user,
         queue_id=request.queue,
         status_message="Starting pipeline",
         status_reason="",

apiserver/services/projects.py

@@ -1,4 +1,4 @@
-from typing import Sequence
+from typing import Sequence, Optional, Tuple
 
 import attr
 from mongoengine import Q
@@ -7,27 +7,33 @@ from apiserver.apierrors import errors
 from apiserver.apierrors.errors.bad_request import InvalidProjectId
 from apiserver.apimodels.base import UpdateResponse, MakePublicRequest, IdResponse
 from apiserver.apimodels.projects import (
+    DeleteRequest,
     GetParamsRequest,
     ProjectTagsRequest,
     ProjectTaskParentsRequest,
     ProjectHyperparamValuesRequest,
     ProjectsGetRequest,
-    DeleteRequest,
     MoveRequest,
     MergeRequest,
-    ProjectOrNoneRequest,
     ProjectRequest,
     ProjectModelMetadataValuesRequest,
+    ProjectChildrenType,
+    GetUniqueMetricsRequest,
+    ProjectUserNamesRequest,
+    EntityTypeEnum,
 )
 from apiserver.bll.organization import OrgBLL, Tags
 from apiserver.bll.project import ProjectBLL, ProjectQueries
+from apiserver.bll.project.project_bll import pipeline_tag, reports_tag
 from apiserver.bll.project.project_cleanup import (
     delete_project,
     validate_project_delete,
 )
 from apiserver.database.errors import translate_errors_context
 from apiserver.database.model import EntityVisibility
+from apiserver.database.model.model import Model
 from apiserver.database.model.project import Project
+from apiserver.database.model.task.task import TaskType, Task
 from apiserver.database.utils import (
     parse_from_call,
     get_company_or_none_constraint,
@@ -96,30 +102,78 @@ def _adjust_search_parameters(data: dict, shallow_search: bool):
         data["parent"] = [None]
 
 
-@endpoint("projects.get_all_ex", request_data_model=ProjectsGetRequest)
+def _get_project_stats_filter(
+    request: ProjectsGetRequest,
+) -> Tuple[Optional[dict], bool]:
+    if request.include_stats_filter or not request.children_type:
+        return request.include_stats_filter, request.search_hidden
+
+    if request.children_tags_filter:
+        stats_filter = {"tags": request.children_tags_filter}
+    elif request.children_tags:
+        stats_filter = {"tags": request.children_tags}
+    else:
+        stats_filter = {}
+
+    if request.children_type == ProjectChildrenType.pipeline:
+        return (
+            {
+                **stats_filter,
+                "system_tags": [pipeline_tag],
+                "type": [TaskType.controller],
+            },
+            True,
+        )
+    if request.children_type == ProjectChildrenType.report:
+        return (
+            {**stats_filter, "system_tags": [reports_tag], "type": [TaskType.report]},
+            True,
+        )
+    return stats_filter, request.search_hidden
+
+
+@endpoint("projects.get_all_ex")
 def get_all_ex(call: APICall, company_id: str, request: ProjectsGetRequest):
     data = call.data
     conform_tag_fields(call, data)
-    allow_public = not request.non_public
+    allow_public = (
+        data["allow_public"]
+        if "allow_public" in data
+        else not data["non_public"]
+        if "non_public" in data
+        else request.allow_public
+    )
+
     requested_ids = data.get("id")
     if isinstance(requested_ids, str):
         requested_ids = [requested_ids]
     _adjust_search_parameters(
         data, shallow_search=request.shallow_search,
     )
-    user_active_project_ids = None
-    if request.active_users:
-        ids, user_active_project_ids = project_bll.get_projects_with_active_user(
+    selected_project_ids = None
+    if request.active_users or request.children_type:
+        ids, selected_project_ids = project_bll.get_projects_with_selected_children(
             company=company_id,
             users=request.active_users,
             project_ids=requested_ids,
             allow_public=allow_public,
+            children_type=request.children_type,
+            children_tags=request.children_tags,
+            children_tags_filter=request.children_tags_filter,
         )
         if not ids:
             return {"projects": []}
         data["id"] = ids
 
     ret_params = {}
+
+    remove_system_tags = False
+    if request.search_hidden:
+        only_fields = data.get("only_fields")
+        if isinstance(only_fields, list) and "system_tags" not in only_fields:
+            only_fields.append("system_tags")
+            remove_system_tags = True
+
     projects: Sequence[dict] = Project.get_many_with_join(
         company=company_id,
         query_dict=data,
@@ -130,29 +184,61 @@ def get_all_ex(call: APICall, company_id: str, request: ProjectsGetRequest):
     if not projects:
         return {"projects": projects, **ret_params}
 
+    if request.search_hidden:
+        for p in projects:
+            system_tags = (
+                p.pop("system_tags", [])
+                if remove_system_tags
+                else p.get("system_tags", [])
+            )
+            if EntityVisibility.hidden.value in system_tags:
+                p["hidden"] = True
+
+    conform_output_tags(call, projects)
     project_ids = list({project["id"] for project in projects})
+
+    stats_filter, stats_search_hidden = _get_project_stats_filter(request)
     if request.check_own_contents:
-        contents = project_bll.calc_own_contents(
-            company=company_id,
-            project_ids=project_ids,
-            filter_=request.include_stats_filter,
-            users=request.active_users,
-        )
+        if request.children_type == ProjectChildrenType.dataset:
+            contents = project_bll.calc_own_datasets(
+                company=company_id,
+                project_ids=project_ids,
+                filter_=stats_filter,
+                users=request.active_users,
+            )
+        else:
+            contents = project_bll.calc_own_contents(
+                company=company_id,
+                project_ids=project_ids,
+                filter_=stats_filter,
+                specific_state=request.stats_for_state,
+                users=request.active_users,
+            )
+
         for project in projects:
             project.update(**contents.get(project["id"], {}))
 
-    conform_output_tags(call, projects)
     if request.include_stats:
-        stats, children = project_bll.get_project_stats(
-            company=company_id,
-            project_ids=project_ids,
-            specific_state=request.stats_for_state,
-            include_children=request.stats_with_children,
-            search_hidden=request.search_hidden,
-            filter_=request.include_stats_filter,
-            users=request.active_users,
-            user_active_project_ids=user_active_project_ids,
-        )
+        if request.children_type == ProjectChildrenType.dataset:
+            stats, children = project_bll.get_project_dataset_stats(
+                company=company_id,
+                project_ids=project_ids,
+                include_children=request.stats_with_children,
+                filter_=stats_filter,
+                users=request.active_users,
+                selected_project_ids=selected_project_ids,
+            )
+        else:
+            stats, children = project_bll.get_project_stats(
+                company=company_id,
+                project_ids=project_ids,
+                specific_state=request.stats_for_state,
+                include_children=request.stats_with_children,
+                search_hidden=stats_search_hidden,
+                filter_=stats_filter,
+                users=request.active_users,
+                selected_project_ids=selected_project_ids,
+            )
 
         for project in projects:
             project["stats"] = stats[project["id"]]
@@ -277,28 +363,30 @@ def delete(call: APICall, company_id: str, request: DeleteRequest):
         project_id=request.project,
         force=request.force,
         delete_contents=request.delete_contents,
+        delete_external_artifacts=request.delete_external_artifacts,
     )
     _reset_cached_tags(company_id, projects=list(affected_projects))
     call.result.data = {**attr.asdict(res)}
 
 
 @endpoint(
-    "projects.get_unique_metric_variants", request_data_model=ProjectOrNoneRequest
+    "projects.get_unique_metric_variants", request_data_model=GetUniqueMetricsRequest
 )
 def get_unique_metric_variants(
-    call: APICall, company_id: str, request: ProjectOrNoneRequest
+    call: APICall, company_id: str, request: GetUniqueMetricsRequest
 ):
 
     metrics = project_queries.get_unique_metric_variants(
         company_id,
         [request.project] if request.project else None,
         include_subprojects=request.include_subprojects,
+        model_metrics=request.model_metrics,
     )
 
     call.result.data = {"metrics": metrics}
 
 
-@endpoint("projects.get_model_metadata_keys",)
+@endpoint("projects.get_model_metadata_keys")
 def get_model_metadata_keys(call: APICall, company_id: str, request: GetParamsRequest):
     total, remaining, keys = project_queries.get_model_metadata_keys(
         company_id,
@@ -325,6 +413,8 @@ def get_model_metadata_values(
         key=request.key,
         include_subprojects=request.include_subprojects,
         allow_public=request.allow_public,
+        page=request.page,
+        page_size=request.page_size,
     )
     call.result.data = {
         "total": total,
@@ -369,6 +459,9 @@ def get_hyperparam_values(
         name=request.name,
         include_subprojects=request.include_subprojects,
         allow_public=request.allow_public,
+        pattern=request.pattern,
+        page=request.page,
+        page_size=request.page_size,
     )
     call.result.data = {
         "total": total,
@@ -420,7 +513,11 @@ def get_tags(call: APICall, company, request: ProjectTagsRequest):
 )
 def make_public(call: APICall, company_id, request: MakePublicRequest):
     call.result.data = Project.set_public(
-        company_id, ids=request.ids, invalid_cls=InvalidProjectId, enabled=True
+        company_id=company_id,
+        user_id=call.identity.user,
+        ids=request.ids,
+        invalid_cls=InvalidProjectId,
+        enabled=True,
     )
 
 
@@ -429,7 +526,11 @@ def make_public(call: APICall, company_id, request: MakePublicRequest):
 )
 def make_public(call: APICall, company_id, request: MakePublicRequest):
     call.result.data = Project.set_public(
-        company_id, ids=request.ids, invalid_cls=InvalidProjectId, enabled=False
+        company_id=company_id,
+        user_id=call.identity.user,
+        ids=request.ids,
+        invalid_cls=InvalidProjectId,
+        enabled=False,
     )
 
 
@@ -442,10 +543,23 @@ def get_task_parents(
     call: APICall, company_id: str, request: ProjectTaskParentsRequest
 ):
     call.result.data = {
-        "parents": project_bll.get_task_parents(
+        "parents": ProjectBLL.get_task_parents(
             company_id,
             projects=request.projects,
             include_subprojects=request.include_subprojects,
             state=request.tasks_state,
+            name=request.task_name,
+        )
+    }
+
+
+@endpoint("projects.get_user_names")
+def get_user_names(call: APICall, company_id: str, request: ProjectUserNamesRequest):
+    call.result.data = {
+        "users": ProjectBLL.get_entity_users(
+            company_id,
+            entity_cls=Model if request.entity == EntityTypeEnum.model else Task,
+            projects=request.projects,
+            include_subprojects=request.include_subprojects,
         )
     }

apiserver/services/queues.py

@@ -1,3 +1,5 @@
+from typing import Union, Sequence
+
 from mongoengine import Q
 
 from apiserver.apimodels.base import UpdateResponse
@@ -21,6 +23,7 @@ from apiserver.apimodels.queues import (
 )
 from apiserver.bll.model import Metadata
 from apiserver.bll.queue import QueueBLL
+from apiserver.bll.queue.queue_bll import MOVE_FIRST, MOVE_LAST
 from apiserver.bll.workers import WorkerBLL
 from apiserver.config_repo import config
 from apiserver.database.model.task.task import Task
@@ -38,14 +41,18 @@ worker_bll = WorkerBLL()
 queue_bll = QueueBLL(worker_bll)
 
 
+def conform_queue_data(call: APICall, queue_data: Union[Sequence[dict], dict]):
+    conform_output_tags(call, queue_data)
+    unescape_metadata(call, queue_data)
+
+
 @endpoint("queues.get_by_id", min_version="2.4", request_data_model=GetByIdRequest)
 def get_by_id(call: APICall, company_id, request: GetByIdRequest):
     queue = queue_bll.get_by_id(
         company_id, request.queue, max_task_entries=request.max_task_entries
     )
     queue_dict = queue.to_proper_dict()
-    conform_output_tags(call, queue_dict)
-    unescape_metadata(call, queue_dict)
+    conform_queue_data(call, queue_dict)
     call.result.data = {"queue": queue_dict}
 
 
@@ -76,16 +83,15 @@ def get_all_ex(call: APICall, company: str, request: GetAllRequest):
     conform_tag_fields(call, call.data)
     ret_params = {}
 
-    Metadata.escape_query_parameters(call)
+    call_data = Metadata.escape_query_parameters(call.data)
     queues = queue_bll.get_queue_infos(
         company_id=company,
-        query_dict=call.data,
-        query=_hidden_query(call.data),
+        query_dict=call_data,
+        query=_hidden_query(call_data),
         max_task_entries=request.max_task_entries,
         ret_params=ret_params,
     )
-    conform_output_tags(call, queues)
-    unescape_metadata(call, queues)
+    conform_queue_data(call, queues)
     call.result.data = {"queues": queues, **ret_params}
 
 
@@ -93,16 +99,15 @@ def get_all_ex(call: APICall, company: str, request: GetAllRequest):
 def get_all(call: APICall, company: str, request: GetAllRequest):
     conform_tag_fields(call, call.data)
     ret_params = {}
-    Metadata.escape_query_parameters(call)
+    call_data = Metadata.escape_query_parameters(call.data)
     queues = queue_bll.get_all(
         company_id=company,
-        query_dict=call.data,
-        query=_hidden_query(call.data),
+        query_dict=call_data,
+        query=_hidden_query(call_data),
         max_task_entries=request.max_task_entries,
         ret_params=ret_params,
     )
-    conform_output_tags(call, queues)
-    unescape_metadata(call, queues)
+    conform_queue_data(call, queues)
     call.result.data = {"queues": queues, **ret_params}
 
 
@@ -134,15 +139,17 @@ def update(call: APICall, company_id, req_model: UpdateRequest):
     updated, fields = queue_bll.update(
         company_id=company_id, queue_id=req_model.queue, **data
     )
-    conform_output_tags(call, fields)
-    unescape_metadata(call, fields)
+    conform_queue_data(call, fields)
     call.result.data_model = UpdateResponse(updated=updated, fields=fields)
 
 
 @endpoint("queues.delete", min_version="2.4", request_data_model=DeleteRequest)
 def delete(call: APICall, company_id, req_model: DeleteRequest):
     queue_bll.delete(
-        company_id=company_id, queue_id=req_model.queue, force=req_model.force
+        company_id=company_id,
+        user_id=call.identity.user,
+        queue_id=req_model.queue,
+        force=req_model.force,
     )
     call.result.data = {"deleted": 1}
 
@@ -164,7 +171,7 @@ def get_next_task(call: APICall, company_id, request: GetNextTaskRequest):
     if entry:
         data = {"entry": entry.to_proper_dict()}
         if request.get_task_info:
-            task = Task.objects(id=entry.task).first()
+            task = Task.objects(id=entry.task).only("company", "user").first()
             if task:
                 data["task_info"] = {"company": task.company, "user": task.user}
 
@@ -192,7 +199,7 @@ def move_task_forward(call: APICall, company_id, req_model: MoveTaskRequest):
             company_id=company_id,
             queue_id=req_model.queue,
             task_id=req_model.task,
-            pos_func=lambda p: max(0, p - req_model.count),
+            move_count=-req_model.count,
         )
     )
 
@@ -209,7 +216,7 @@ def move_task_backward(call: APICall, company_id, req_model: MoveTaskRequest):
             company_id=company_id,
             queue_id=req_model.queue,
             task_id=req_model.task,
-            pos_func=lambda p: max(0, p + req_model.count),
+            move_count=req_model.count,
         )
     )
 
@@ -226,7 +233,7 @@ def move_task_to_front(call: APICall, company_id, req_model: TaskRequest):
             company_id=company_id,
             queue_id=req_model.queue,
             task_id=req_model.task,
-            pos_func=lambda p: 0,
+            move_count=MOVE_FIRST,
         )
     )
 
@@ -243,7 +250,7 @@ def move_task_to_back(call: APICall, company_id, req_model: TaskRequest):
             company_id=company_id,
             queue_id=req_model.queue,
             task_id=req_model.task,
-            pos_func=lambda p: -1,
+            move_count=MOVE_LAST,
         )
     )
 

apiserver/services/reports.py

@@ -0,0 +1,420 @@
+import textwrap
+from datetime import datetime
+from itertools import chain
+from typing import Sequence
+
+from mongoengine import Q
+
+from apiserver.apimodels.reports import (
+    CreateReportRequest,
+    UpdateReportRequest,
+    PublishReportRequest,
+    ArchiveReportRequest,
+    DeleteReportRequest,
+    MoveReportRequest,
+    GetTasksDataRequest,
+    EventsRequest,
+    GetAllRequest,
+)
+from apiserver.apierrors import errors
+from apiserver.apimodels.base import UpdateResponse
+from apiserver.bll.project.project_bll import reports_project_name, reports_tag
+from apiserver.database.model.model import Model
+from apiserver.services.models import conform_model_data
+from apiserver.services.utils import process_include_subprojects, sort_tags_response
+from apiserver.bll.organization import OrgBLL
+from apiserver.bll.project import ProjectBLL
+from apiserver.bll.task import TaskBLL, ChangeStatusRequest
+from apiserver.database.model import EntityVisibility
+from apiserver.database.model.project import Project
+from apiserver.database.model.task.task import Task, TaskType, TaskStatus
+from apiserver.service_repo import APICall, endpoint
+from apiserver.services.events import (
+    _get_task_or_model_index_companies,
+    event_bll,
+    _get_metrics_response,
+    _get_metric_variants_from_request,
+    _get_multitask_plots,
+    _get_single_value_metrics_response,
+)
+from apiserver.services.tasks import (
+    escape_execution_parameters,
+    _hidden_query,
+    conform_task_data,
+)
+
+org_bll = OrgBLL()
+project_bll = ProjectBLL()
+task_bll = TaskBLL()
+
+
+update_fields = {
+    "name",
+    "tags",
+    "comment",
+    "report",
+    "report_assets",
+}
+
+
+def _assert_report(company_id, task_id, only_fields=None, requires_write_access=True):
+    if only_fields and "type" not in only_fields:
+        only_fields += ("type",)
+
+    task = TaskBLL.get_task_with_access(
+        task_id=task_id,
+        company_id=company_id,
+        only=only_fields,
+        requires_write_access=requires_write_access,
+    )
+    if task.type != TaskType.report:
+        raise errors.bad_request.OperationSupportedOnReportsOnly(id=task_id)
+
+    return task
+
+
+@endpoint("reports.update", response_data_model=UpdateResponse)
+def update_report(call: APICall, company_id: str, request: UpdateReportRequest):
+    task = _assert_report(
+        task_id=request.task,
+        company_id=company_id,
+        only_fields=("status",),
+    )
+
+    partial_update_dict = {
+        field: value for field, value in call.data.items() if field in update_fields
+    }
+    if not partial_update_dict:
+        return UpdateResponse(updated=0)
+
+    allowed_for_published = set(partial_update_dict.keys()).issubset(
+        {"tags", "name", "comment"}
+    )
+    if task.status != TaskStatus.created and not allowed_for_published:
+        raise errors.bad_request.InvalidTaskStatus(
+            expected=TaskStatus.created, status=task.status
+        )
+
+    now = datetime.utcnow()
+    more_updates = {"last_change": now, "last_changed_by": call.identity.user}
+    if not allowed_for_published:
+        more_updates["last_update"] = now
+
+    updated = task.update(upsert=False, **partial_update_dict, **more_updates)
+    if not updated:
+        return UpdateResponse(updated=0)
+
+    updated_tags = partial_update_dict.get("tags")
+    if updated_tags:
+        partial_update_dict["tags"] = sorted(updated_tags)
+    updated_report = partial_update_dict.get("report")
+    if updated_report:
+        partial_update_dict["report"] = textwrap.shorten(updated_report, width=100)
+
+    return UpdateResponse(updated=updated, fields=partial_update_dict)
+
+
+def _ensure_reports_project(company: str, user: str, name: str):
+    name = name.strip("/")
+    _, _, basename = name.rpartition("/")
+    if basename != reports_project_name:
+        name = f"{name}/{reports_project_name}"
+
+    return project_bll.find_or_create(
+        user=user,
+        company=company,
+        project_name=name,
+        description="Reports project",
+        system_tags=[reports_tag, EntityVisibility.hidden.value],
+    )
+
+
+@endpoint("reports.create")
+def create_report(call: APICall, company_id: str, request: CreateReportRequest):
+    user_id = call.identity.user
+    project_id = request.project
+    if request.project:
+        project = Project.get_for_writing(
+            company=company_id, id=project_id, _only=("name",)
+        )
+        project_name = project.name
+    else:
+        project_name = ""
+
+    project_id = _ensure_reports_project(
+        company=company_id, user=user_id, name=project_name
+    )
+    task = task_bll.create(
+        company=company_id,
+        user=user_id,
+        fields=dict(
+            project=project_id,
+            name=request.name,
+            tags=request.tags,
+            comment=request.comment,
+            type=TaskType.report,
+            system_tags=[reports_tag, EntityVisibility.hidden.value],
+        ),
+    )
+    task.save()
+
+    call.result.data = {"id": task.id, "project_id": project_id}
+
+
+def _delete_reports_project_if_empty(project_id):
+    project = Project.objects(id=project_id).only("basename").first()
+    if (
+        project
+        and project.basename == reports_project_name
+        and Task.objects(project=project_id).count() == 0
+    ):
+        project.delete()
+
+
+@endpoint("reports.get_all_ex")
+def get_all_ex(call: APICall, company_id, request: GetAllRequest):
+    call_data = call.data
+    call_data["type"] = TaskType.report
+    process_include_subprojects(call_data)
+    # bring projects one level down in case not the .reports project was passed
+    if "project" in call_data:
+        project_ids = call_data["project"]
+        if not isinstance(project_ids, list):
+            project_ids = [project_ids]
+
+        query = Q(parent__in=project_ids) | Q(id__in=project_ids)
+        project_ids = Project.objects(query & Q(basename=reports_project_name)).scalar(
+            "id"
+        )
+        if not project_ids:
+            return {"tasks": []}
+        call_data["project"] = list(project_ids)
+
+    ret_params = {}
+    tasks = Task.get_many_with_join(
+        company=company_id,
+        query_dict=call_data,
+        allow_public=request.allow_public,
+        ret_params=ret_params,
+    )
+    conform_task_data(call, tasks)
+
+    call.result.data = {"tasks": tasks, **ret_params}
+
+
+def _get_task_metrics_from_request(
+    task_ids: Sequence[str], request: EventsRequest
+) -> dict:
+    task_metrics = {}
+    for task in task_ids:
+        task_dict = {}
+        for mv in request.metrics:
+            task_dict[mv.metric] = mv.variants
+        task_metrics[task] = task_dict
+
+    return task_metrics
+
+
+@endpoint("reports.get_task_data")
+def get_task_data(call: APICall, company_id, request: GetTasksDataRequest):
+    if request.model_events:
+        entity_cls = Model
+        conform_data = conform_model_data
+    else:
+        entity_cls = Task
+        conform_data = conform_task_data
+
+    call_data = escape_execution_parameters(call.data)
+    process_include_subprojects(call_data)
+
+    ret_params = {}
+    tasks = entity_cls.get_many_with_join(
+        company=company_id,
+        query_dict=call_data,
+        query=_hidden_query(call_data),
+        allow_public=request.allow_public,
+        ret_params=ret_params,
+    )
+    conform_data(call, tasks)
+    res = {"tasks": tasks, **ret_params}
+    if not (
+        request.debug_images
+        or request.plots
+        or request.scalar_metrics_iter_histogram
+        or request.single_value_metrics
+    ):
+        return res
+
+    task_ids = [task["id"] for task in tasks]
+    companies = _get_task_or_model_index_companies(
+        company_id, task_ids=task_ids, model_events=request.model_events
+    )
+    if request.debug_images:
+        result = event_bll.debug_images_iterator.get_task_events(
+            companies={
+                t.id: t.company for t in chain.from_iterable(companies.values())
+            },
+            task_metrics=_get_task_metrics_from_request(task_ids, request.debug_images),
+            iter_count=request.debug_images.iters,
+        )
+        res["debug_images"] = [
+            r.to_struct() for r in _get_metrics_response(result.metric_events)
+        ]
+
+    if request.plots:
+        res["plots"] = _get_multitask_plots(
+            companies=companies,
+            last_iters=request.plots.iters,
+            metrics=_get_metric_variants_from_request(request.plots.metrics),
+            last_iters_per_task_metric=request.plots.last_iters_per_task_metric,
+        )[0]
+
+    if request.scalar_metrics_iter_histogram:
+        res[
+            "scalar_metrics_iter_histogram"
+        ] = event_bll.metrics.compare_scalar_metrics_average_per_iter(
+            companies=companies,
+            samples=request.scalar_metrics_iter_histogram.samples,
+            key=request.scalar_metrics_iter_histogram.key,
+            metric_variants=_get_metric_variants_from_request(
+                request.scalar_metrics_iter_histogram.metrics
+            ),
+        )
+
+    if request.single_value_metrics:
+        res["single_value_metrics"] = _get_single_value_metrics_response(
+            companies=companies,
+            value_metrics=event_bll.metrics.get_task_single_value_metrics(
+                companies=companies
+            ),
+        )
+
+    call.result.data = res
+
+
+@endpoint("reports.move")
+def move(call: APICall, company_id: str, request: MoveReportRequest):
+    if not ("project" in call.data or request.project_name):
+        raise errors.bad_request.MissingRequiredFields(
+            "project or project_name is required"
+        )
+
+    task = _assert_report(
+        company_id=company_id,
+        task_id=request.task,
+        only_fields=("project",),
+    )
+    user_id = call.identity.user
+    project_name = request.project_name
+    if not project_name:
+        if request.project:
+            project = Project.get_for_writing(
+                company=company_id, id=request.project, _only=("name",)
+            )
+            project_name = project.name
+        else:
+            project_name = ""
+
+    project_id = _ensure_reports_project(
+        company=company_id, user=user_id, name=project_name
+    )
+
+    project_bll.move_under_project(
+        entity_cls=Task,
+        user=call.identity.user,
+        company=company_id,
+        ids=[request.task],
+        project=project_id,
+    )
+
+    _delete_reports_project_if_empty(task.project)
+
+    return {"project_id": project_id}
+
+
+@endpoint(
+    "reports.publish",
+    response_data_model=UpdateResponse,
+)
+def publish(call: APICall, company_id, request: PublishReportRequest):
+    task = _assert_report(company_id=company_id, task_id=request.task)
+    updates = ChangeStatusRequest(
+        task=task,
+        new_status=TaskStatus.published,
+        force=True,
+        status_reason="",
+        status_message=request.message,
+        user_id=call.identity.user,
+    ).execute(published=datetime.utcnow())
+
+    call.result.data_model = UpdateResponse(**updates)
+
+
+@endpoint("reports.archive")
+def archive(call: APICall, company_id, request: ArchiveReportRequest):
+    task = _assert_report(company_id=company_id, task_id=request.task)
+    archived = task.update(
+        status_message=request.message,
+        status_reason="",
+        add_to_set__system_tags=EntityVisibility.archived.value,
+        last_change=datetime.utcnow(),
+        last_changed_by=call.identity.user,
+    )
+
+    return {"archived": archived}
+
+
+@endpoint("reports.unarchive")
+def unarchive(call: APICall, company_id, request: ArchiveReportRequest):
+    task = _assert_report(company_id=company_id, task_id=request.task)
+    unarchived = task.update(
+        status_message=request.message,
+        status_reason="",
+        pull__system_tags=EntityVisibility.archived.value,
+        last_change=datetime.utcnow(),
+        last_changed_by=call.identity.user,
+    )
+    return {"unarchived": unarchived}
+
+
+# @endpoint("reports.share")
+# def share(call: APICall, company_id, request: ShareReportRequest):
+#     _assert_report(
+#         company_id=company_id, user_id=call.identity.user, task_id=request.task
+#     )
+#     call.result.data = {
+#         "changed": task_bll.share_task(
+#             company_id=company_id, task_ids=[request.task], share=request.share
+#         )
+#     }
+
+
+@endpoint("reports.delete")
+def delete(call: APICall, company_id, request: DeleteReportRequest):
+    task = _assert_report(
+        company_id=company_id,
+        task_id=request.task,
+        only_fields=("project",),
+    )
+    if (
+        task.status != TaskStatus.created
+        and EntityVisibility.archived.value not in task.system_tags
+        and not request.force
+    ):
+        raise errors.bad_request.TaskCannotBeDeleted(
+            "due to status, use force=True",
+            task=task.id,
+            expected=TaskStatus.created,
+            current=task.status,
+        )
+
+    task.delete()
+    _delete_reports_project_if_empty(task.project)
+
+    call.result.data = {"deleted": 1}
+
+
+@endpoint("reports.get_tags")
+def get_tags(call: APICall, company_id: str, _):
+    tags = Task.objects(company=company_id, type=TaskType.report).distinct(field="tags")
+    call.result.data = sort_tags_response({"tags": tags})

apiserver/services/tasks.py

@@ -64,6 +64,10 @@ from apiserver.apimodels.tasks import (
     ResetBatchItem,
     CompletedRequest,
     CompletedResponse,
+    GetAllReq,
+    DequeueRequest,
+    DequeueManyRequest,
+    UpdateTagsRequest,
 )
 from apiserver.bll.event import EventBLL
 from apiserver.bll.model import ModelBLL
@@ -73,7 +77,6 @@ from apiserver.bll.queue import QueueBLL
 from apiserver.bll.task import (
     TaskBLL,
     ChangeStatusRequest,
-    update_project_time,
 )
 from apiserver.bll.task.artifacts import (
     artifacts_prepare_for_save,
@@ -98,7 +101,7 @@ from apiserver.bll.task.task_operations import (
     move_tasks_to_trash,
 )
 from apiserver.bll.task.utils import update_task, get_task_for_update, deleted_prefix
-from apiserver.bll.util import SetFieldsResolver, run_batch_operation
+from apiserver.bll.util import run_batch_operation, update_project_time
 from apiserver.database.errors import translate_errors_context
 from apiserver.database.model import EntityVisibility
 from apiserver.database.model.task.output import Output
@@ -109,7 +112,11 @@ from apiserver.database.model.task.task import (
     ModelItem,
     TaskModelTypes,
 )
-from apiserver.database.utils import get_fields_attr, parse_from_call, get_options
+from apiserver.database.utils import (
+    get_fields_attr,
+    parse_from_call,
+    get_options,
+)
 from apiserver.service_repo import APICall, endpoint
 from apiserver.services.utils import (
     conform_tag_fields,
@@ -136,32 +143,15 @@ project_bll = ProjectBLL()
 
 
 def set_task_status_from_call(
-    request: UpdateRequest, company_id, new_status=None, **set_fields
+    request: UpdateRequest, company_id: str, user_id: str, new_status=None, **set_fields
 ) -> dict:
-    fields_resolver = SetFieldsResolver(set_fields)
     task = TaskBLL.get_task_with_access(
         request.task,
         company_id=company_id,
-        only=tuple(
-            {"status", "project", "started", "duration"} | fields_resolver.get_names()
-        ),
+        only=("id", "status", "project"),
         requires_write_access=True,
     )
 
-    if "duration" not in fields_resolver.get_names():
-        if new_status == Task.started:
-            fields_resolver.add_fields(min__duration=max(0, task.duration or 0))
-        elif new_status in (
-            TaskStatus.completed,
-            TaskStatus.failed,
-            TaskStatus.stopped,
-        ):
-            fields_resolver.add_fields(
-                duration=int((task.started - datetime.utcnow()).total_seconds())
-                if task.started
-                else 0
-            )
-
     status_reason = request.status_reason
     status_message = request.status_message
     force = request.force
@@ -171,7 +161,8 @@ def set_task_status_from_call(
         status_reason=status_reason,
         status_message=status_message,
         force=force,
-    ).execute(**fields_resolver.get_fields(task))
+        user_id=user_id,
+    ).execute(**set_fields)
 
 
 @endpoint("tasks.get_by_id", request_data_model=TaskRequest)
@@ -180,17 +171,17 @@ def get_by_id(call: APICall, company_id, req_model: TaskRequest):
         req_model.task, company_id=company_id, allow_public=True
     )
     task_dict = task.to_proper_dict()
-    unprepare_from_saved(call, task_dict)
+    conform_task_data(call, task_dict)
     call.result.data = {"task": task_dict}
 
 
-def escape_execution_parameters(call: APICall) -> dict:
-    if not call.data:
-        return call.data
+def escape_execution_parameters(call_data: dict) -> dict:
+    if not call_data:
+        return call_data
 
-    keys = list(call.data)
+    keys = list(call_data)
     call_data = {
-        safe_key: call.data[key] for key, safe_key in zip(keys, escape_paths(keys))
+        safe_key: call_data[key] for key, safe_key in zip(keys, escape_paths(keys))
     }
 
     projection = Task.get_projection(call_data)
@@ -214,44 +205,40 @@ def _hidden_query(data: dict) -> Q:
     return Q(system_tags__ne=EntityVisibility.hidden.value)
 
 
-@endpoint("tasks.get_all_ex", required_fields=[])
-def get_all_ex(call: APICall, company_id, _):
+@endpoint("tasks.get_all_ex")
+def get_all_ex(call: APICall, company_id, request: GetAllReq):
     conform_tag_fields(call, call.data)
-
-    call_data = escape_execution_parameters(call)
-
+    call_data = escape_execution_parameters(call.data)
     process_include_subprojects(call_data)
     ret_params = {}
     tasks = Task.get_many_with_join(
         company=company_id,
         query_dict=call_data,
         query=_hidden_query(call_data),
-        allow_public=True,
+        allow_public=request.allow_public,
         ret_params=ret_params,
     )
-    unprepare_from_saved(call, tasks)
+    conform_task_data(call, tasks)
     call.result.data = {"tasks": tasks, **ret_params}
 
 
 @endpoint("tasks.get_by_id_ex", required_fields=["id"])
 def get_by_id_ex(call: APICall, company_id, _):
     conform_tag_fields(call, call.data)
-
-    call_data = escape_execution_parameters(call)
-
+    call_data = escape_execution_parameters(call.data)
     tasks = Task.get_many_with_join(
         company=company_id, query_dict=call_data, allow_public=True,
     )
 
-    unprepare_from_saved(call, tasks)
+    conform_task_data(call, tasks)
     call.result.data = {"tasks": tasks}
 
 
 @endpoint("tasks.get_all", required_fields=[])
 def get_all(call: APICall, company_id, _):
     conform_tag_fields(call, call.data)
-
-    call_data = escape_execution_parameters(call)
+    call_data = escape_execution_parameters(call.data)
+    process_include_subprojects(call_data)
 
     ret_params = {}
     tasks = Task.get_many(
@@ -262,7 +249,7 @@ def get_all(call: APICall, company_id, _):
         allow_public=True,
         ret_params=ret_params,
     )
-    unprepare_from_saved(call, tasks)
+    conform_task_data(call, tasks)
     call.result.data = {"tasks": tasks, **ret_params}
 
 
@@ -291,6 +278,7 @@ def stop(call: APICall, company_id, req_model: UpdateRequest):
         **stop_task(
             task_id=req_model.task,
             company_id=company_id,
+            user_id=call.identity.user,
             user_name=call.identity.user_name,
             status_reason=req_model.status_reason,
             force=req_model.force,
@@ -308,6 +296,7 @@ def stop_many(call: APICall, company_id, request: StopManyRequest):
         func=partial(
             stop_task,
             company_id=company_id,
+            user_id=call.identity.user,
             user_name=call.identity.user_name,
             status_reason=request.status_reason,
             force=request.force,
@@ -329,7 +318,8 @@ def stopped(call: APICall, company_id, req_model: UpdateRequest):
     call.result.data_model = UpdateResponse(
         **set_task_status_from_call(
             req_model,
-            company_id,
+            company_id=company_id,
+            user_id=call.identity.user,
             new_status=TaskStatus.stopped,
             completed=datetime.utcnow(),
         )
@@ -345,7 +335,8 @@ def started(call: APICall, company_id, req_model: UpdateRequest):
     res = StartedResponse(
         **set_task_status_from_call(
             req_model,
-            company_id,
+            company_id=company_id,
+            user_id=call.identity.user,
             new_status=TaskStatus.in_progress,
             min__started=datetime.utcnow(),  # don't override a previous, smaller "started" field value
         )
@@ -359,7 +350,12 @@ def started(call: APICall, company_id, req_model: UpdateRequest):
 )
 def failed(call: APICall, company_id, req_model: UpdateRequest):
     call.result.data_model = UpdateResponse(
-        **set_task_status_from_call(req_model, company_id, new_status=TaskStatus.failed)
+        **set_task_status_from_call(
+            req_model,
+            company_id=company_id,
+            user_id=call.identity.user,
+            new_status=TaskStatus.failed,
+        )
     )
 
 
@@ -368,7 +364,12 @@ def failed(call: APICall, company_id, req_model: UpdateRequest):
 )
 def close(call: APICall, company_id, req_model: UpdateRequest):
     call.result.data_model = UpdateResponse(
-        **set_task_status_from_call(req_model, company_id, new_status=TaskStatus.closed)
+        **set_task_status_from_call(
+            req_model,
+            company_id=company_id,
+            user_id=call.identity.user,
+            new_status=TaskStatus.closed,
+        )
     )
 
 
@@ -415,7 +416,7 @@ def prepare_for_save(call: APICall, fields: dict, previous_task: Task = None):
     return fields
 
 
-def unprepare_from_saved(call: APICall, tasks_data: Union[Sequence[dict], dict]):
+def conform_task_data(call: APICall, tasks_data: Union[Sequence[dict], dict]):
     if isinstance(tasks_data, dict):
         tasks_data = [tasks_data]
 
@@ -475,7 +476,9 @@ def _validate_and_get_task_from_call(call: APICall, **kwargs) -> Tuple[Task, dic
         field_does_not_exist_cls=errors.bad_request.ValidationError
     ):
         fields = prepare_create_fields(call, **kwargs)
-        task = task_bll.create(call, fields)
+        task = task_bll.create(
+            company=call.identity.company, user=call.identity.user, fields=fields
+        )
 
     task_bll.validate(task)
 
@@ -494,7 +497,7 @@ def _update_cached_tags(company: str, project: str, fields: dict):
     org_bll.update_tags(
         company,
         Tags.Task,
-        project=project,
+        projects=[project],
         tags=fields.get("tags"),
         system_tags=fields.get("system_tags"),
     )
@@ -578,7 +581,9 @@ def update(call: APICall, company_id, req_model: UpdateRequest):
             company_id=company_id,
             id=task_id,
             partial_update_dict=partial_update_dict,
-            injected_update=dict(last_change=datetime.utcnow()),
+            injected_update=dict(
+                last_change=datetime.utcnow(), last_changed_by=call.identity.user,
+            ),
         )
         if updated_count:
             new_project = updated_fields.get("project", task.project)
@@ -589,7 +594,7 @@ def update(call: APICall, company_id, req_model: UpdateRequest):
                     company_id, project=task.project, fields=updated_fields
                 )
             update_project_time(updated_fields.get("project"))
-        unprepare_from_saved(call, updated_fields)
+        conform_task_data(call, updated_fields)
         return UpdateResponse(updated=updated_count, fields=updated_fields)
 
 
@@ -611,7 +616,11 @@ def set_requirements(call: APICall, company_id, req_model: SetRequirementsReques
             raise errors.bad_request.MissingTaskFields(
                 "Task has no script field", task=task.id
             )
-        res = update_task(task, update_cmds=dict(script__requirements=requirements))
+        res = update_task(
+            task,
+            user_id=call.identity.user,
+            update_cmds=dict(script__requirements=requirements),
+        )
         call.result.data_model = UpdateResponse(updated=res)
         if res:
             call.result.data_model.fields = {"script.requirements": requirements}
@@ -646,7 +655,9 @@ def update_batch(call: APICall, company_id, _):
             partial_update_dict = Task.get_safe_update_dict(fields)
             if not partial_update_dict:
                 continue
-            partial_update_dict.update(last_change=now)
+            partial_update_dict.update(
+                last_change=now, last_changed_by=call.identity.user,
+            )
             update_op = UpdateOne(
                 {"_id": id, "company": company_id}, {"$set": partial_update_dict}
             )
@@ -710,7 +721,11 @@ def edit(call: APICall, company_id, req_model: UpdateRequest):
                 d.update(value)
                 fields[key] = d
 
-        task_bll.validate(task_bll.create(call, fields))
+        task_bll.validate(
+            task_bll.create(
+                company=call.identity.company, user=call.identity.user, fields=fields
+            )
+        )
 
         # make sure field names do not end in mongoengine comparison operators
         fixed_fields = {
@@ -719,7 +734,7 @@ def edit(call: APICall, company_id, req_model: UpdateRequest):
         }
         if fixed_fields:
             now = datetime.utcnow()
-            last_change = dict(last_change=now)
+            last_change = dict(last_change=now, last_changed_by=call.identity.user)
             if not set(fields).issubset(Task.user_set_allowed()):
                 last_change.update(last_update=now)
             fields.update(**last_change)
@@ -734,7 +749,7 @@ def edit(call: APICall, company_id, req_model: UpdateRequest):
                         company_id, project=task.project, fields=fixed_fields
                     )
                 update_project_time(fields.get("project"))
-            unprepare_from_saved(call, fields)
+            conform_task_data(call, fields)
             call.result.data_model = UpdateResponse(updated=updated, fields=fields)
         else:
             call.result.data_model = UpdateResponse(updated=0)
@@ -756,6 +771,7 @@ def edit_hyper_params(call: APICall, company_id, request: EditHyperParamsRequest
     call.result.data = {
         "updated": HyperParams.edit_params(
             company_id,
+            user_id=call.identity.user,
             task_id=request.task,
             hyperparams=request.hyperparams,
             replace_hyperparams=request.replace_hyperparams,
@@ -769,6 +785,7 @@ def delete_hyper_params(call: APICall, company_id, request: DeleteHyperParamsReq
     call.result.data = {
         "deleted": HyperParams.delete_params(
             company_id,
+            user_id=call.identity.user,
             task_id=request.task,
             hyperparams=request.hyperparams,
             force=request.force,
@@ -813,6 +830,7 @@ def edit_configuration(call: APICall, company_id, request: EditConfigurationRequ
     call.result.data = {
         "updated": HyperParams.edit_configuration(
             company_id,
+            user_id=call.identity.user,
             task_id=request.task,
             configuration=request.configuration,
             replace_configuration=request.replace_configuration,
@@ -828,6 +846,7 @@ def delete_configuration(
     call.result.data = {
         "deleted": HyperParams.delete_configuration(
             company_id,
+            user_id=call.identity.user,
             task_id=request.task,
             configuration=request.configuration,
             force=request.force,
@@ -844,6 +863,7 @@ def enqueue(call: APICall, company_id, request: EnqueueRequest):
     queued, res = enqueue_task(
         task_id=request.task,
         company_id=company_id,
+        user_id=call.identity.user,
         queue_id=request.queue,
         status_message=request.status_message,
         status_reason=request.status_reason,
@@ -868,6 +888,7 @@ def enqueue_many(call: APICall, company_id, request: EnqueueManyRequest):
         func=partial(
             enqueue_task,
             company_id=company_id,
+            user_id=call.identity.user,
             queue_id=request.queue,
             status_message=request.status_message,
             status_reason=request.status_reason,
@@ -894,32 +915,34 @@ def enqueue_many(call: APICall, company_id, request: EnqueueManyRequest):
 
 
 @endpoint(
-    "tasks.dequeue",
-    request_data_model=UpdateRequest,
-    response_data_model=DequeueResponse,
+    "tasks.dequeue", response_data_model=DequeueResponse,
 )
-def dequeue(call: APICall, company_id, request: UpdateRequest):
+def dequeue(call: APICall, company_id, request: DequeueRequest):
     dequeued, res = dequeue_task(
         task_id=request.task,
         company_id=company_id,
+        user_id=call.identity.user,
         status_message=request.status_message,
         status_reason=request.status_reason,
+        remove_from_all_queues=request.remove_from_all_queues,
+        new_status=request.new_status,
     )
     call.result.data_model = DequeueResponse(dequeued=dequeued, **res)
 
 
 @endpoint(
-    "tasks.dequeue_many",
-    request_data_model=TaskBatchRequest,
-    response_data_model=DequeueManyResponse,
+    "tasks.dequeue_many", response_data_model=DequeueManyResponse,
 )
-def dequeue_many(call: APICall, company_id, request: TaskBatchRequest):
+def dequeue_many(call: APICall, company_id, request: DequeueManyRequest):
     results, failures = run_batch_operation(
         func=partial(
             dequeue_task,
             company_id=company_id,
+            user_id=call.identity.user,
             status_message=request.status_message,
             status_reason=request.status_reason,
+            remove_from_all_queues=request.remove_from_all_queues,
+            new_status=request.new_status,
         ),
         ids=request.ids,
     )
@@ -1007,12 +1030,21 @@ def archive(call: APICall, company_id, request: ArchiveRequest):
     tasks = TaskBLL.assert_exists(
         company_id,
         task_ids=request.tasks,
-        only=("id", "execution", "status", "project", "system_tags", "enqueue_status"),
+        only=(
+            "id",
+            "company",
+            "execution",
+            "status",
+            "project",
+            "system_tags",
+            "enqueue_status",
+        ),
     )
     archived = 0
     for task in tasks:
         archived += archive_task(
             company_id=company_id,
+            user_id=call.identity.user,
             task=task,
             status_message=request.status_message,
             status_reason=request.status_reason,
@@ -1031,6 +1063,7 @@ def archive_many(call: APICall, company_id, request: TaskBatchRequest):
         func=partial(
             archive_task,
             company_id=company_id,
+            user_id=call.identity.user,
             status_message=request.status_message,
             status_reason=request.status_reason,
         ),
@@ -1052,6 +1085,7 @@ def unarchive_many(call: APICall, company_id, request: TaskBatchRequest):
         func=partial(
             unarchive_task,
             company_id=company_id,
+            user_id=call.identity.user,
             status_message=request.status_message,
             status_reason=request.status_reason,
         ),
@@ -1130,6 +1164,7 @@ def publish(call: APICall, company_id, request: PublishRequest):
     updates = publish_task(
         task_id=request.task,
         company_id=company_id,
+        user_id=call.identity.user,
         force=request.force,
         publish_model_func=ModelBLL.publish_model if request.publish_model else None,
         status_reason=request.status_reason,
@@ -1148,6 +1183,7 @@ def publish_many(call: APICall, company_id, request: PublishManyRequest):
         func=partial(
             publish_task,
             company_id=company_id,
+            user_id=call.identity.user,
             force=request.force,
             publish_model_func=ModelBLL.publish_model
             if request.publish_model
@@ -1174,7 +1210,8 @@ def completed(call: APICall, company_id, request: CompletedRequest):
     res = CompletedResponse(
         **set_task_status_from_call(
             request,
-            company_id,
+            company_id=company_id,
+            user_id=call.identity.user,
             new_status=TaskStatus.completed,
             completed=datetime.utcnow(),
         )
@@ -1184,6 +1221,7 @@ def completed(call: APICall, company_id, request: CompletedRequest):
         publish_res = publish_task(
             task_id=request.task,
             company_id=company_id,
+            user_id=call.identity.user,
             force=request.force,
             publish_model_func=ModelBLL.publish_model,
             status_reason=request.status_reason,
@@ -1198,9 +1236,12 @@ def completed(call: APICall, company_id, request: CompletedRequest):
 
 
 @endpoint("tasks.ping", request_data_model=PingRequest)
-def ping(_, company_id, request: PingRequest):
+def ping(call: APICall, company_id, request: PingRequest):
     TaskBLL.set_last_update(
-        task_ids=[request.task], company_id=company_id, last_update=datetime.utcnow()
+        task_ids=[request.task],
+        company_id=company_id,
+        user_id=call.identity.user,
+        last_update=datetime.utcnow(),
     )
 
 
@@ -1215,6 +1256,7 @@ def add_or_update_artifacts(
     call.result.data = {
         "updated": Artifacts.add_or_update_artifacts(
             company_id=company_id,
+            user_id=call.identity.user,
             task_id=request.task,
             artifacts=request.artifacts,
             force=True,
@@ -1231,6 +1273,7 @@ def delete_artifacts(call: APICall, company_id, request: DeleteArtifactsRequest)
     call.result.data = {
         "deleted": Artifacts.delete_artifacts(
             company_id=company_id,
+            user_id=call.identity.user,
             task_id=request.task,
             artifact_ids=request.artifacts,
             force=True,
@@ -1241,20 +1284,28 @@ def delete_artifacts(call: APICall, company_id, request: DeleteArtifactsRequest)
 @endpoint("tasks.make_public", min_version="2.9", request_data_model=MakePublicRequest)
 def make_public(call: APICall, company_id, request: MakePublicRequest):
     call.result.data = Task.set_public(
-        company_id, request.ids, invalid_cls=InvalidTaskId, enabled=True
+        company_id=company_id,
+        user_id=call.identity.user,
+        ids=request.ids,
+        invalid_cls=InvalidTaskId,
+        enabled=True,
     )
 
 
 @endpoint("tasks.make_private", min_version="2.9", request_data_model=MakePublicRequest)
 def make_public(call: APICall, company_id, request: MakePublicRequest):
     call.result.data = Task.set_public(
-        company_id, request.ids, invalid_cls=InvalidTaskId, enabled=False
+        company_id=company_id,
+        user_id=call.identity.user,
+        ids=request.ids,
+        invalid_cls=InvalidTaskId,
+        enabled=False,
     )
 
 
 @endpoint("tasks.move", request_data_model=MoveRequest)
 def move(call: APICall, company_id: str, request: MoveRequest):
-    if not (request.project or request.project_name):
+    if not ("project" in call.data or request.project_name):
         raise errors.bad_request.MissingRequiredFields(
             "project or project_name is required"
         )
@@ -1278,8 +1329,21 @@ def move(call: APICall, company_id: str, request: MoveRequest):
     return {"project_id": project_id}
 
 
+@endpoint("tasks.update_tags")
+def update_tags(_, company_id: str, request: UpdateTagsRequest):
+    return {
+        "updated": org_bll.edit_entity_tags(
+            company_id=company_id,
+            entity_cls=Task,
+            entity_ids=request.ids,
+            add_tags=request.add_tags,
+            remove_tags=request.remove_tags,
+        )
+    }
+
+
 @endpoint("tasks.add_or_update_model", min_version="2.13")
-def add_or_update_model(_: APICall, company_id: str, request: AddUpdateModelRequest):
+def add_or_update_model(call: APICall, company_id: str, request: AddUpdateModelRequest):
     get_task_for_update(company_id=company_id, task_id=request.task, force=True)
 
     models_field = f"models__{request.type}"
@@ -1290,6 +1354,7 @@ def add_or_update_model(_: APICall, company_id: str, request: AddUpdateModelRequ
     updated = TaskBLL.update_statistics(
         task_id=request.task,
         company_id=company_id,
+        user_id=call.identity.user,
         last_iteration_max=request.iteration,
         **({f"push__{models_field}": model} if not updated else {}),
     )
@@ -1298,7 +1363,7 @@ def add_or_update_model(_: APICall, company_id: str, request: AddUpdateModelRequ
 
 
 @endpoint("tasks.delete_models", min_version="2.13")
-def delete_models(_: APICall, company_id: str, request: DeleteModelsRequest):
+def delete_models(call: APICall, company_id: str, request: DeleteModelsRequest):
     task = get_task_for_update(company_id=company_id, task_id=request.task, force=True)
 
     delete_names = {
@@ -1311,5 +1376,7 @@ def delete_models(_: APICall, company_id: str, request: DeleteModelsRequest):
         if names
     }
 
-    updated = task.update(last_change=datetime.utcnow(), **commands,)
+    updated = task.update(
+        last_change=datetime.utcnow(), last_changed_by=call.identity.user, **commands,
+    )
     return {"updated": updated}

apiserver/services/users.py

@@ -98,9 +98,7 @@ def get_current_user(call: APICall, company_id, _):
     user_id = call.identity.user
 
     projection = (
-        {"company.name"}
-        .union(User.get_fields())
-        .difference(User.get_exclude_fields())
+        {"company.name"}.union(User.get_fields()).difference(User.get_exclude_fields())
     )
     res = User.get_many_with_join(
         query=Q(id=user_id),
@@ -114,9 +112,13 @@ def get_current_user(call: APICall, company_id, _):
     user = res[0]
     user["role"] = call.identity.role
 
-    resp = {
-        "user": user,
-        "getting_started": config.get("apiserver.getting_started_info", None),
+    resp = dict(
+        user=user, getting_started=config.get("apiserver.getting_started_info", None)
+    )
+    resp["settings"] = {
+        "max_download_items": int(
+            config.get("services.organization.download.max_download_items", 1000)
+        )
     }
     call.result.data = resp
 

apiserver/services/utils.py

@@ -101,7 +101,7 @@ def _upgrade_tags(call: APICall, tags: Sequence, system_tags: Sequence):
 def validate_tags(tags: Sequence[str], system_tags: Sequence[str]):
     for values in filter(None, (tags, system_tags)):
         unsupported = [
-            t for t in values if t.startswith(GetMixin.ListFieldBucketHelper.op_prefix)
+            t for t in values if t.startswith(GetMixin.NewListFieldBucketHelper.op_prefix)
         ]
         if unsupported:
             raise errors.bad_request.FieldsValueError(

apiserver/services/workers.py

@@ -22,6 +22,7 @@ from apiserver.apimodels.workers import (
     GetActivityReportRequest,
     GetActivityReportResponse,
     ActivityReportSeries,
+    GetCountRequest,
 )
 from apiserver.bll.workers import WorkerBLL
 from apiserver.config_repo import config
@@ -50,12 +51,29 @@ def get_all(call: APICall, company_id: str, request: GetAllRequest):
     )
 
 
+@endpoint(
+    "workers.get_count", request_data_model=GetCountRequest,
+)
+def get_all(call: APICall, company_id: str, request: GetCountRequest):
+    call.result.data = {
+        "count": worker_bll.get_count(
+            company_id,
+            request.last_seen,
+            tags=request.tags,
+            system_tags=request.system_tags,
+        )
+    }
+
+
 @endpoint("workers.register", min_version="2.4", request_data_model=RegisterRequest)
 def register(call: APICall, company_id, request: RegisterRequest):
     worker = request.worker
     timeout = request.timeout
     queues = request.queues
 
+    if not timeout:
+        timeout = config.get("apiserver.workers.default_timeout", 10 * 60)
+
     if not timeout or timeout <= 0:
         raise bad_request.WorkerRegistrationFailed(
             "invalid timeout", timeout=timeout, worker=worker