OpenWebUI/open-webui
1
0
Fork 0
mirror of https://github.com/open-webui/open-webui synced 2025-06-26 18:26:48 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Merge pull request #10486 from Micca/feature/document_intelligence_support

Browse Source 
Feat: Adding Support for Azure AI Document Intelligence for Content Extraction (Revised)
This commit is contained in:
Timothy Jaeryang Baek
2025-02-21 10:56:18 -08:00
committed by GitHub
parent a3c2ff61ed 35f3824932
commit ab1b910d80
8 changed files with 107 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
backend/open_webui/config.py
View File

@@ -1579,6 +1579,18 @@ TIKA_SERVER_URL = PersistentConfig(
os.getenv("TIKA_SERVER_URL", "http://tika:9998"), # Default for sidecar deployment
)
DOCUMENT_INTELLIGENCE_ENDPOINT = PersistentConfig(
"DOCUMENT_INTELLIGENCE_ENDPOINT",
"rag.document_intelligence_endpoint",
os.getenv("DOCUMENT_INTELLIGENCE_ENDPOINT", ""),
)
DOCUMENT_INTELLIGENCE_KEY = PersistentConfig(
"DOCUMENT_INTELLIGENCE_KEY",
"rag.document_intelligence_key",
os.getenv("DOCUMENT_INTELLIGENCE_KEY", ""),
)
RAG_TOP_K = PersistentConfig(
"RAG_TOP_K", "rag.top_k", int(os.environ.get("RAG_TOP_K", "3"))
)

4
backend/open_webui/main.py
View File

@@ -180,6 +180,8 @@ from open_webui.config import (
CHUNK_SIZE,
CONTENT_EXTRACTION_ENGINE,
TIKA_SERVER_URL,
DOCUMENT_INTELLIGENCE_ENDPOINT,
DOCUMENT_INTELLIGENCE_KEY,
RAG_TOP_K,
RAG_TEXT_SPLITTER,
TIKTOKEN_ENCODING_NAME,
@@ -533,6 +535,8 @@ app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION = (
app.state.config.CONTENT_EXTRACTION_ENGINE = CONTENT_EXTRACTION_ENGINE
app.state.config.TIKA_SERVER_URL = TIKA_SERVER_URL
app.state.config.DOCUMENT_INTELLIGENCE_ENDPOINT = DOCUMENT_INTELLIGENCE_ENDPOINT
app.state.config.DOCUMENT_INTELLIGENCE_KEY = DOCUMENT_INTELLIGENCE_KEY
app.state.config.TEXT_SPLITTER = RAG_TEXT_SPLITTER
app.state.config.TIKTOKEN_ENCODING_NAME = TIKTOKEN_ENCODING_NAME

22
backend/open_webui/retrieval/loaders/main.py
View File

@@ -4,6 +4,7 @@ import ftfy
import sys
from langchain_community.document_loaders import (
AzureAIDocumentIntelligenceLoader,
BSHTMLLoader,
CSVLoader,
Docx2txtLoader,
@@ -147,6 +148,27 @@ class Loader:
file_path=file_path,
mime_type=file_content_type,
)
elif (
self.engine == "document_intelligence"
and self.kwargs.get("DOCUMENT_INTELLIGENCE_ENDPOINT") != ""
and self.kwargs.get("DOCUMENT_INTELLIGENCE_KEY") != ""
and (
file_ext in ["pdf", "xls", "xlsx", "docx", "ppt", "pptx"]
or file_content_type
in [
"application/vnd.ms-excel",
"application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
"application/vnd.openxmlformats-officedocument.wordprocessingml.document",
"application/vnd.ms-powerpoint",
"application/vnd.openxmlformats-officedocument.presentationml.presentation",
]
)
):
loader = AzureAIDocumentIntelligenceLoader(
file_path=file_path,
api_endpoint=self.kwargs.get("DOCUMENT_INTELLIGENCE_ENDPOINT"),
api_key=self.kwargs.get("DOCUMENT_INTELLIGENCE_KEY"),
)
else:
if file_ext == "pdf":
loader = PyPDFLoader(

27
backend/open_webui/routers/retrieval.py
View File

@@ -356,6 +356,10 @@ async def get_rag_config(request: Request, user=Depends(get_admin_user)):
"content_extraction": {
"engine": request.app.state.config.CONTENT_EXTRACTION_ENGINE,
"tika_server_url": request.app.state.config.TIKA_SERVER_URL,
"document_intelligence_config": {
"endpoint": request.app.state.config.DOCUMENT_INTELLIGENCE_ENDPOINT,
"key": request.app.state.config.DOCUMENT_INTELLIGENCE_KEY,
},
},
"chunk": {
"text_splitter": request.app.state.config.TEXT_SPLITTER,
@@ -411,9 +415,15 @@ class FileConfig(BaseModel):
max_count: Optional[int] = None
class DocumentIntelligenceConfigForm(BaseModel):
endpoint: str
key: str
class ContentExtractionConfig(BaseModel):
engine: str = ""
tika_server_url: Optional[str] = None
document_intelligence_config: Optional[DocumentIntelligenceConfigForm] = None
class ChunkParamUpdateForm(BaseModel):
@@ -501,13 +511,22 @@ async def update_rag_config(
request.app.state.config.FILE_MAX_COUNT = form_data.file.max_count
if form_data.content_extraction is not None:
log.info(f"Updating text settings: {form_data.content_extraction}")
log.info(
f"Updating content extraction: {request.app.state.config.CONTENT_EXTRACTION_ENGINE} to {form_data.content_extraction.engine}"
)
request.app.state.config.CONTENT_EXTRACTION_ENGINE = (
form_data.content_extraction.engine
)
request.app.state.config.TIKA_SERVER_URL = (
form_data.content_extraction.tika_server_url
)
if form_data.content_extraction.document_intelligence_config is not None:
request.app.state.config.DOCUMENT_INTELLIGENCE_ENDPOINT = (
form_data.content_extraction.document_intelligence_config.endpoint
)
request.app.state.config.DOCUMENT_INTELLIGENCE_KEY = (
form_data.content_extraction.document_intelligence_config.key
)
if form_data.chunk is not None:
request.app.state.config.TEXT_SPLITTER = form_data.chunk.text_splitter
@@ -604,6 +623,10 @@ async def update_rag_config(
"content_extraction": {
"engine": request.app.state.config.CONTENT_EXTRACTION_ENGINE,
"tika_server_url": request.app.state.config.TIKA_SERVER_URL,
"document_intelligence_config": {
"endpoint": request.app.state.config.DOCUMENT_INTELLIGENCE_ENDPOINT,
"key": request.app.state.config.DOCUMENT_INTELLIGENCE_KEY,
},
},
"chunk": {
"text_splitter": request.app.state.config.TEXT_SPLITTER,
@@ -937,6 +960,8 @@ def process_file(
engine=request.app.state.config.CONTENT_EXTRACTION_ENGINE,
TIKA_SERVER_URL=request.app.state.config.TIKA_SERVER_URL,
PDF_EXTRACT_IMAGES=request.app.state.config.PDF_EXTRACT_IMAGES,
DOCUMENT_INTELLIGENCE_ENDPOINT=request.app.state.config.DOCUMENT_INTELLIGENCE_ENDPOINT,
DOCUMENT_INTELLIGENCE_KEY=request.app.state.config.DOCUMENT_INTELLIGENCE_KEY,
)
docs = loader.load(
file.filename, file.meta.get("content_type"), file_path