mirror of
https://github.com/deepseek-ai/deepseek-harness
synced 2026-08-15 21:04:50 +00:00
fix(user-interaction): preserve multi-select custom answers
This commit is contained in:
Yichen Jiang
@@ -0,0 +1,6 @@
|
||||
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write .agents/notes/implemented/bug-fix/2026-07-30-multi-select-custom-answer-composition.md
|
||||
2026-07-30-multi-select-custom-answer-composition.md: 7194f4a79f1dd49eba4a9b626d75203fced06544
|
||||
2026-07-30-multi-select-custom-answer-composition.zh.md: fac09c8db0ebf2dd4a84ade7aa7868128656025d
|
||||
@@ -0,0 +1,25 @@
|
||||
# Agent Note: Multi-select custom answer composition
|
||||
|
||||
Status: implemented
|
||||
|
||||
English | [中文](2026-07-30-multi-select-custom-answer-composition.zh.md)
|
||||
|
||||
## Problem
|
||||
|
||||
The user-interaction result vocabulary carries selected option labels and optional custom text in separate fields, but its original semantics made them mutually exclusive for every question. On a multi-select question, opening or typing the custom answer discarded labels the user had already selected. The TUI returned only the custom text, and the Web host rejected a client response that preserved both fields.
|
||||
|
||||
## Decision
|
||||
|
||||
For a question with `multiSelect: true`, one answer item may contain both a non-empty `selected` array and non-empty `custom` text. Web drafts preserve both values regardless of whether the user selects an option or types custom text first; the TUI projects its checked option set when custom text is submitted; and the Web host accepts the combined response after applying its existing id, label, uniqueness, batch, and non-empty-text validation.
|
||||
|
||||
Single-select and optionless questions keep exclusive semantics: custom text overrides any selected option. The result shape remains `{ id, selected, custom? }`, so no wire or tool-output schema changes.
|
||||
|
||||
## Alternatives considered
|
||||
|
||||
**Encode custom text as another `selected` label.** Rejected because it would erase the distinction between caller-provided option labels and human-authored text, weakening validation and forcing consumers to infer which value was custom.
|
||||
|
||||
**Allow `selected` and `custom` together for every question.** Rejected because a single-select question represents one answer; permitting a selected option plus custom text would make its cardinality ambiguous. The combined form is limited to questions that explicitly opt into multiple answers.
|
||||
|
||||
## Consequences
|
||||
|
||||
Multi-select UIs can represent the user's complete answer without discarding either source. Providers and consumers retain the existing DTO, while request-aware validators interpret the allowed combination from `multiSelect`. Web, TUI, host-response, tool-projection, and assembled keyless TUI coverage pin the combined result; single-select host coverage pins the remaining exclusivity rule.
|
||||
@@ -0,0 +1,25 @@
|
||||
# Agent Note: 多选题自定义答案组合
|
||||
|
||||
Status: implemented
|
||||
|
||||
[English](2026-07-30-multi-select-custom-answer-composition.md) | 中文
|
||||
|
||||
## 问题
|
||||
|
||||
用户交互结果的词汇分别通过不同字段携带选中的选项标签和可选的自定义文本，但最初的语义要求每个问题的这两个字段互斥。对于多选题，打开自定义答案或输入文本会丢弃用户已选中的标签。TUI 只返回自定义文本，而 Web 宿主会拒绝同时保留两个字段的客户端响应。
|
||||
|
||||
## 决策
|
||||
|
||||
对于 `multiSelect: true` 的问题，一个回答项可以同时包含非空 `selected` 数组与非空 `custom` 文本。无论用户先选择选项还是先输入自定义文本，Web 草稿都会保留两个值；提交自定义文本时，TUI 会投影其已勾选的选项集合；Web 宿主则在应用现有的 id、标签、唯一性、批次和非空文本校验后接受组合响应。
|
||||
|
||||
单选题和无选项问题仍保持互斥语义：自定义文本会覆盖任何已选中的选项。结果形状仍为 `{ id, selected, custom? }`，因此协议或工具输出 schema 均无需变更。
|
||||
|
||||
## 考虑过的替代方案
|
||||
|
||||
**把自定义文本编码为另一个 `selected` 标签。** 不予采纳，因为这样会抹去调用方提供的选项标签与用户填写文本之间的区别，削弱校验，并迫使消费方推断哪个值属于自定义内容。
|
||||
|
||||
**允许所有问题同时使用 `selected` 与 `custom`。** 不予采纳，因为单选题只表示一个回答；允许选中选项与自定义文本并存会使其基数含义模糊。组合形式仅适用于显式选择多项回答的问题。
|
||||
|
||||
## 后果
|
||||
|
||||
多选 UI 可以完整表达用户的回答，不会丢弃任一来源。提供方和消费方继续使用现有 DTO，而请求感知的校验器会根据 `multiSelect` 判断是否允许组合。Web、TUI、宿主响应、工具投影和组装后的无密钥 TUI 覆盖会固定组合结果；单选题的宿主覆盖则固定其余的互斥规则。
|
||||
@@ -1,6 +1,6 @@
|
||||
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write
|
||||
user-interaction.md: 798a9790f424683775284a98421be08e6e1399e3
|
||||
user-interaction.zh.md: 12bfcffe4fe4caaacb54e90126eac55e333d64a5
|
||||
# pnpm run verify-translation-pairing --write docs/core-data-structures/user-interaction.md
|
||||
user-interaction.md: db6ac5010ada9d02319bf148566792659711d2e4
|
||||
user-interaction.zh.md: a8306b421a03563ba9ae2ee48d04898d00eb668e
|
||||
|
||||
@@ -60,14 +60,14 @@ interface AskUserQuestionRequest {
|
||||
|
||||
## Answer
|
||||
|
||||
Providers return one answer item per question id. `selected` contains selected option labels, and `custom` carries a free-form "Other" answer when the user typed one. When `custom` is present, `selected` is empty; custom text is an answer override, not a supplement to selected choices. A UI may also use an item with empty `selected` and no `custom` to preserve a skipped question in an otherwise completed batch.
|
||||
Providers return one answer item per question id. `selected` contains selected option labels, and `custom` carries a free-form "Other" answer when the user typed one. For a single-select question, `custom` overrides the selected choice and `selected` is empty. For a multi-select question, `custom` may supplement the labels in `selected`. A UI may also use an item with empty `selected` and no `custom` to preserve a skipped question in an otherwise completed batch.
|
||||
|
||||
```ts type-equiv
|
||||
/** Answer to one question. */
|
||||
interface AskUserQuestionAnswerItem {
|
||||
/** The answered question id. */
|
||||
id: string
|
||||
/** Selected option labels. Empty for custom or unanswered choices. */
|
||||
/** Selected option labels. May accompany custom text for a multi-select question. */
|
||||
selected: string[]
|
||||
/** Optional free-text "Other" answer. */
|
||||
custom?: string
|
||||
|
||||
@@ -60,14 +60,14 @@ interface AskUserQuestionRequest {
|
||||
|
||||
## 回答
|
||||
|
||||
提供方为每个问题 id 返回一个回答项。`selected` 包含选中的选项标签，`custom` 在用户输入自由文本时携带「其他」回答。当 `custom` 存在时，`selected` 为空；自定义文本是对选中项的覆盖，而非补充。UI 也可以使用 `selected` 为空且不含 `custom` 的回答项，在其余问题均已完成的批次中保留被跳过的问题。
|
||||
提供方为每个问题 id 返回一个回答项。`selected` 包含选中的选项标签，`custom` 在用户输入自由文本时携带「其他」回答。对于单选题，`custom` 会覆盖选中的选项，且 `selected` 为空。对于多选题，`custom` 可以补充 `selected` 中的标签。UI 也可以使用 `selected` 为空且不含 `custom` 的回答项，在其余问题均已完成的批次中保留被跳过的问题。
|
||||
|
||||
```ts type-equiv
|
||||
/** Answer to one question. */
|
||||
interface AskUserQuestionAnswerItem {
|
||||
/** The answered question id. */
|
||||
id: string
|
||||
/** Selected option labels. Empty for custom or unanswered choices. */
|
||||
/** Selected option labels. May accompany custom text for a multi-select question. */
|
||||
selected: string[]
|
||||
/** Optional free-text "Other" answer. */
|
||||
custom?: string
|
||||
|
||||
@@ -110,6 +110,12 @@ class ScriptedTuiAdapter extends LlmAdapter {
|
||||
|
||||
const hasToolResult = lastMessage?.content.some(block => block.type === 'tool-result') ?? false
|
||||
if (hasToolResult) {
|
||||
const toolResultText = lastMessage?.content.flatMap(block => block.type === 'tool-result'
|
||||
? block.content.flatMap(content => content.type === 'text' ? [content.text] : [])
|
||||
: []).join('\n') ?? ''
|
||||
if (toolResultText !== '{"answers":[{"id":"mode","selected":["Safe"],"custom":"Release notes"}]}') {
|
||||
throw new Error(`the scripted TUI request received an unexpected question answer: ${toolResultText}`)
|
||||
}
|
||||
for (const chunk of textChunks(FINAL_TEXT)) yield chunk
|
||||
return
|
||||
}
|
||||
@@ -119,6 +125,7 @@ class ScriptedTuiAdapter extends LlmAdapter {
|
||||
id: 'mode',
|
||||
header: 'Execution mode',
|
||||
question: 'How should the scripted run proceed?',
|
||||
multi_select: true,
|
||||
options: [
|
||||
{ label: 'Safe', description: 'Use the guarded path.' },
|
||||
{ label: 'Fast', description: 'Use the shorter path.' },
|
||||
|
||||
@@ -138,6 +138,7 @@ const SELECT_PRO_MODEL = [
|
||||
{ waitFor: 'scripted TUI ready.', send: '/model\r' },
|
||||
{ waitFor: 'Select model', send: '\x1b[B\x1b[Z\r' },
|
||||
] as const
|
||||
const ANSWER_MULTI_WITH_CUSTOM = ' \tRelease notes\r'
|
||||
|
||||
describe('tui-agent keyless smoke (real Loader tree in a PTY)', () => {
|
||||
it('boots pi-tui, sweeps the borderless banner in, enters plan mode, and restores the terminal', async () => {
|
||||
@@ -174,7 +175,10 @@ describe('tui-agent keyless smoke (real Loader tree in a PTY)', () => {
|
||||
// The question text first appears in the streamed tool-call card. Wait
|
||||
// for the dialog's input legend so Enter cannot arrive before it owns
|
||||
// terminal input when pre-dispatch policy yields.
|
||||
{ waitFor: 'Tab custom answer • ↑/↓ navigate • Enter submit • Esc interrupt', send: '\r' },
|
||||
{
|
||||
waitFor: 'Tab custom answer • ↑/↓ navigate • Space toggle • Enter submit • Esc interrupt',
|
||||
send: ANSWER_MULTI_WITH_CUSTOM,
|
||||
},
|
||||
{ waitFor: 'Decision received. Scripted TUI run complete.', send: '' },
|
||||
// Session title: the first user message drives the first-message-llm
|
||||
// provider's tool-less title call; the scripted adapter answers it, the
|
||||
@@ -200,6 +204,7 @@ describe('tui-agent keyless smoke (real Loader tree in a PTY)', () => {
|
||||
expect(output).not.toContain('\u001B[999CMODEL_CURSOR')
|
||||
expect(output).not.toContain('\u009B31mMODEL_C1')
|
||||
expect(output).toContain('Safe')
|
||||
expect(output).toContain('Release notes')
|
||||
expect(output).toContain('\u001B]0;scripted session title — DeepSeek Harness\u0007')
|
||||
expect(output).toContain('Session status')
|
||||
expect(output).toContain('Title')
|
||||
@@ -395,7 +400,7 @@ describe('dsh CLI keyless smoke (apps/cli through the same PTY)', () => {
|
||||
actions: [
|
||||
...SELECT_PRO_MODEL,
|
||||
{ waitFor: 'Model selected: tui-scripted/tui-scripted-model-pro.', send: 'exercise the TUI\r' },
|
||||
{ waitFor: 'How should the scripted run proceed?', send: '\r' },
|
||||
{ waitFor: 'How should the scripted run proceed?', send: ANSWER_MULTI_WITH_CUSTOM },
|
||||
{ waitFor: 'Decision received. Scripted TUI run complete.', send: '/exit\r' },
|
||||
],
|
||||
inspect: async (cwd) => { context = await readLoggedRequestContext(cwd) },
|
||||
|
||||
@@ -2,5 +2,5 @@
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write packages/client/ui-question/README.md
|
||||
README.md: 3a3cd639fc2834685230aca7c8087583e0a48c71
|
||||
README.zh.md: 1330578577da7ed7d0890595f675fd272fd5ebc7
|
||||
README.md: c36f1474e175b52c7d35af6b479ab5bfeabcd9ff
|
||||
README.zh.md: 8986dee718a98920a20757aafb1bd4b54ac8f782
|
||||
|
||||
@@ -4,7 +4,7 @@ English | [中文](README.zh.md)
|
||||
|
||||
Web `ask_user_question` feature plugin. Its host half mounts `dsh-tool-ask-user` only when the Web feature is selected; its browser half registers the `question` entry in the conversation-owned `conversation.composer` keyed slot.
|
||||
|
||||
The component renders one question at a time with progress navigation, single- and multi-select choices, recommendation badges derived from label suffixes, and custom answers. Question detail reuses the assistant-output `MarkdownText` primitive, including its GFM rendering and untrusted-content policy. The capped card keeps its title, navigation, and submission actions fixed while long detail and choices share an internal scroll region. Single-select choices advance immediately, and Enter submits once every question is answered or skipped; Enter during IME composition confirms the input candidate without advancing. It submits one structured answer batch for the whole request: “Skip this question” retains other drafts and emits the existing blank `{ selected: [] }` shape for that item, while close rejects the whole wait as `ASK_CANCELLED`.
|
||||
The component renders one question at a time with progress navigation, single- and multi-select choices, recommendation badges derived from label suffixes, and custom answers. A multi-select draft keeps its selected labels while the user opens or edits the custom answer, so its submitted item may carry both `selected` and `custom`; a single-select custom answer remains exclusive. Question detail reuses the assistant-output `MarkdownText` primitive, including its GFM rendering and untrusted-content policy. The capped card keeps its title, navigation, and submission actions fixed while long detail and choices share an internal scroll region. Single-select choices advance immediately, and Enter submits once every question is answered or skipped; Enter during IME composition confirms the input candidate without advancing. It submits one structured answer batch for the whole request: “Skip this question” retains other drafts and emits the existing blank `{ selected: [] }` shape for that item, while close rejects the whole wait as `ASK_CANCELLED`.
|
||||
|
||||
Selection state is local to a component keyed by the request rpcId. A replay with the same id preserves a still-mounted draft, while `question/resolved` from the host removes the composer. The host remains authoritative: successful HTTP delivery does not remove pending state locally.
|
||||
|
||||
|
||||
@@ -4,7 +4,7 @@
|
||||
|
||||
Web `ask_user_question` 功能插件。只有选择 Web 功能时，其主机侧才会挂载 `dsh-tool-ask-user`；浏览器侧会把 `question` 配置项注册到会话拥有的 `conversation.composer` 键控 slot 中。
|
||||
|
||||
组件每次渲染一个问题，提供进度导航、单选和多选选项、由标签后缀派生的推荐徽标，以及自定义答案。问题详情复用助手输出的 `MarkdownText` 原语，包括其 GFM 渲染与不受信内容策略。封顶卡片保持标题、导航与提交动作固定，超长的详情与选项共享内部滚动区。单选选项会立即前进；所有问题均已回答或跳过后，Enter 会提交；IME 输入法组合期间按 Enter 只会确认输入候选，不会前进。组件为整个请求提交一批结构化答案：「跳过此问题」会保留其他草稿，并为该项发出既有的空 `{ selected: [] }` 形状；关闭则以 `ASK_CANCELLED` 拒绝整个等待。
|
||||
组件每次渲染一个问题，提供进度导航、单选和多选选项、由标签后缀派生的推荐徽标，以及自定义答案。用户打开或编辑自定义答案时，多选题草稿会保留已选中的标签，因此提交项可以同时携带 `selected` 与 `custom`；单选题的自定义答案仍保持互斥。问题详情复用助手输出的 `MarkdownText` 原语，包括其 GFM 渲染与不受信内容策略。封顶卡片保持标题、导航与提交动作固定，超长的详情与选项共享内部滚动区。单选选项会立即前进；所有问题均已回答或跳过后，Enter 会提交；IME 输入法组合期间按 Enter 只会确认输入候选，不会前进。组件为整个请求提交一批结构化答案：「跳过此问题」会保留其他草稿，并为该项发出既有的空 `{ selected: [] }` 形状；关闭则以 `ASK_CANCELLED` 拒绝整个等待。
|
||||
|
||||
选择状态只存在于以请求 rpcId 为 key 的组件本地。使用相同 id 回放时，只要组件仍挂载，就会保留草稿；主机发出的 `question/resolved` 则会移除编辑器。主机仍具有最终决定权：HTTP 交付成功不会在本地移除待处理状态。
|
||||
|
||||
|
||||
@@ -86,12 +86,13 @@ function QuestionFlow({ pending }: { pending: PendingQuestion }) {
|
||||
|
||||
const choose = (label: string): void => {
|
||||
updateDraft((current) => {
|
||||
const selected = question.multiSelect === true
|
||||
? current.selected.includes(label)
|
||||
if (question.multiSelect === true) {
|
||||
const selected = current.selected.includes(label)
|
||||
? current.selected.filter(item => item !== label)
|
||||
: [...current.selected, label]
|
||||
: [label]
|
||||
return { selected, custom: '', customOpen: false, skipped: false }
|
||||
return { ...current, selected, skipped: false }
|
||||
}
|
||||
return { selected: [label], custom: '', customOpen: false, skipped: false }
|
||||
})
|
||||
if (question.multiSelect !== true && index < questions.length - 1) {
|
||||
setIndex(current => current + 1)
|
||||
@@ -99,7 +100,12 @@ function QuestionFlow({ pending }: { pending: PendingQuestion }) {
|
||||
}
|
||||
|
||||
const openCustom = (): void => {
|
||||
updateDraft(current => ({ ...current, selected: [], customOpen: true, skipped: false }))
|
||||
updateDraft(current => ({
|
||||
...current,
|
||||
selected: question.multiSelect === true ? current.selected : [],
|
||||
customOpen: true,
|
||||
skipped: false,
|
||||
}))
|
||||
}
|
||||
|
||||
const answered = (item: DraftAnswer): boolean =>
|
||||
@@ -121,7 +127,7 @@ function QuestionFlow({ pending }: { pending: PendingQuestion }) {
|
||||
const custom = value.custom.trim()
|
||||
return {
|
||||
id: item.id,
|
||||
selected: custom === '' ? value.selected : [],
|
||||
selected: custom === '' || item.multiSelect === true ? value.selected : [],
|
||||
...(custom === '' ? {} : { custom }),
|
||||
}
|
||||
}),
|
||||
@@ -269,7 +275,11 @@ function QuestionFlow({ pending }: { pending: PendingQuestion }) {
|
||||
onChange={(event) => {
|
||||
const value = event.target.value
|
||||
updateDraft(current => ({
|
||||
...current, selected: [], custom: value, customOpen: true, skipped: false,
|
||||
...current,
|
||||
selected: question.multiSelect === true ? current.selected : [],
|
||||
custom: value,
|
||||
customOpen: true,
|
||||
skipped: false,
|
||||
}))
|
||||
}}
|
||||
onKeyDown={(event) => {
|
||||
|
||||
@@ -96,13 +96,20 @@ describe('QuestionComposer', () => {
|
||||
fireEvent.click(screen.getByRole('checkbox', { name: '系统设计' }))
|
||||
fireEvent.click(screen.getByRole('checkbox', { name: '系统设计' }))
|
||||
fireEvent.click(screen.getByRole('checkbox', { name: '代码质量' }))
|
||||
fireEvent.keyDown(screen.getByRole('checkbox', { name: '代码质量' }), { key: 'Enter' })
|
||||
fireEvent.click(screen.getByRole('button', { name: '其他，请填写自定义答案' }))
|
||||
const multiCustom = screen.getByPlaceholderText('输入你的答案')
|
||||
fireEvent.change(multiCustom, { target: { value: '沟通能力' } })
|
||||
fireEvent.click(screen.getByRole('checkbox', { name: '产品判断' }))
|
||||
expect(screen.getByRole('checkbox', { name: '系统设计' }).getAttribute('aria-checked')).toBe('true')
|
||||
expect(screen.getByRole('checkbox', { name: '代码质量' }).getAttribute('aria-checked')).toBe('true')
|
||||
expect((multiCustom as HTMLTextAreaElement).value).toBe('沟通能力')
|
||||
fireEvent.keyDown(multiCustom, { key: 'Enter' })
|
||||
|
||||
// The domain face encoded the whole batch into one carrier envelope.
|
||||
expect(respond).toHaveBeenCalledWith(answeredEnvelope('question-1', [
|
||||
{ id: 'profile', selected: ['工程落地型 (Recommended)'] },
|
||||
{ id: 'detail', selected: [], custom: '要能独立排查线上问题' },
|
||||
{ id: 'signals', selected: ['系统设计', '代码质量'] },
|
||||
{ id: 'signals', selected: ['系统设计', '代码质量', '产品判断'], custom: '沟通能力' },
|
||||
]))
|
||||
expect(screen.getByRole<HTMLButtonElement>('button', { name: '正在提交…' }).disabled).toBe(true)
|
||||
})
|
||||
|
||||
@@ -2,5 +2,5 @@
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write packages/host/apiproxy/README.md
|
||||
README.md: ca4471454f5be5d3fcba38ce665d4fb3fbd85e74
|
||||
README.zh.md: 953539e1198a52b2bf7cdd9ca1b0d263cc2ae6f9
|
||||
README.md: d517608404239809df03b089e150dbbecbf6d7cc
|
||||
README.zh.md: f37427205fc72ef60f923d9d938adee0d4aa241c
|
||||
|
||||
@@ -10,6 +10,8 @@ Wire messages form a four-quadrant discriminated union — who initiates × requ
|
||||
|
||||
The layering/protocol decisions are recorded in the [GUI layering and RPC protocol RFC](../../../.agents/notes/implemented/architecture/2026-07-19-gui-layering-and-rpc-protocol.md); the browser-side consumption architecture in the [web client architecture RFC](../../../.agents/notes/implemented/architecture/2026-07-19-gui-web-client-architecture.md).
|
||||
|
||||
Question responses are validated against their pending request before the first answer claims it. A multi-select item may carry both requested option labels in `selected` and non-empty `custom` text; a single-select item must use one or the other. Duplicate labels, unknown labels, mismatched ids, incomplete batches, and empty custom text are rejected as `bad-response`.
|
||||
|
||||
`session.history`'s tail page (`beforeSeq` absent) additionally carries an optional `projections` block — the watermark snapshot of every unit registered on `ctx.sessionProjections` (`@deepseek-ai/dsh-session-projection`), with `asOfSeq` = the last event seq the values reflect (`-1` on an empty log). The gateway also subscribes to the registry's change feed and mints a `session/projection` mux frame per changed unit (`{sessionId, key, value, seq}` — live push state, never logged; clients hold one generic per-session value store under higher-seq-wins). The carrier holds zero domain knowledge (each value passed its unit's own schema inside the registry; the wire schemas keep `values`/`value` wide); loadOlder pages never carry the block, and a composition without the registry serves histories without either surface.
|
||||
|
||||
Session titles ride the generic projection pair like every other domain — the history-tail `projections` block plus `session/projection` frames under the `title` key (the bespoke `session/title` frame is retired). Titles do not join `session.list`; cold sessions remain metadata-only there until opening or resuming attaches their logs.
|
||||
|
||||
@@ -10,6 +10,8 @@
|
||||
|
||||
分层与协议决策记录在 [GUI 分层与 RPC 协议 RFC](../../../.agents/notes/implemented/architecture/2026-07-19-gui-layering-and-rpc-protocol.md)中；浏览器侧消费架构记录在 [Web 客户端架构 RFC](../../../.agents/notes/implemented/architecture/2026-07-19-gui-web-client-architecture.md)中。
|
||||
|
||||
首个回答认领待处理请求之前，系统会对照该请求校验问题响应。多选题的回答项可以同时携带 `selected` 中的请求选项标签与非空 `custom` 文本；单选题的回答项必须二选一。标签重复、标签未知、id 不匹配、批次不完整以及自定义文本为空都会以 `bad-response` 拒绝。
|
||||
|
||||
`session.history` 的尾页（不带 `beforeSeq`）额外携带一个可选的 `projections` 块——`ctx.sessionProjections`（`@deepseek-ai/dsh-session-projection`）上每个已注册单元的水位线快照，`asOfSeq` = 这些值共同反映到的最后一个事件 seq（空日志为 `-1`）。网关还订阅注册表的变更流，为每个状态发生变化的单元铸造一个 `session/projection` mux 帧（`{sessionId, key, value, seq}`——实时推送状态，绝不入日志；客户端按 seq 高者胜维护一个按会话的通用值仓）。载体不持有任何领域知识（每个值在注册表内部已过其单元自己的 schema；协议 schema 对 `values`/`value` 保持宽松）；loadOlder 页永不携带该块，未装注册表的组合则两个面都不提供。
|
||||
|
||||
会话标题与其他所有领域一样搭乘这对通用投影机制——历史尾页的 `projections` 块外加 `title` 键下的 `session/projection` 帧（专设的 `session/title` 帧已下线）。标题不会加入 `session.list`；冷会话在其中仍只有元数据，直到打开或恢复操作附加其日志。
|
||||
|
||||
@@ -275,7 +275,7 @@ function matchesQuestions(payload: QuestionResponsePayload, pending: PendingQues
|
||||
if (new Set(answer.selected).size !== answer.selected.length) return false
|
||||
const custom = answer.custom?.trim()
|
||||
if (custom !== undefined && custom === '') return false
|
||||
if (custom !== undefined && answer.selected.length > 0) return false
|
||||
if (custom !== undefined && answer.selected.length > 0 && question.multiSelect !== true) return false
|
||||
if (question.multiSelect !== true && answer.selected.length > 1) return false
|
||||
const labels = new Set(question.options?.map(option => option.label) ?? [])
|
||||
return answer.selected.every(label => labels.has(label))
|
||||
|
||||
116
packages/host/apiproxy/tests/api-proxy-question.spec.ts
Normal file
116
packages/host/apiproxy/tests/api-proxy-question.spec.ts
Normal file
@@ -0,0 +1,116 @@
|
||||
import { describe, expect, it } from 'vitest'
|
||||
import { Context } from 'cordis'
|
||||
import type { Agent } from '@deepseek-ai/dsh-agent'
|
||||
import SessionStore from '@deepseek-ai/dsh-session'
|
||||
import UserInteractionService from '@deepseek-ai/dsh-user-interaction'
|
||||
import type { ApiProxy, MuxFrame, RpcRequest } from '@deepseek-ai/dsh-host-apiproxy/api'
|
||||
import { RpcId } from '@deepseek-ai/dsh-host-apiproxy/api/rpc'
|
||||
import { createApiProxy } from '../src/api-proxy.ts'
|
||||
|
||||
async function harness(): Promise<{ ctx: Context; api: ApiProxy }> {
|
||||
const ctx = new Context()
|
||||
await ctx.plugin(SessionStore)
|
||||
await ctx.plugin(UserInteractionService)
|
||||
return {
|
||||
ctx,
|
||||
api: createApiProxy(ctx, { provider: 'p', model: 'm', cwd: '/tmp', workspaceRoot: '/tmp' }),
|
||||
}
|
||||
}
|
||||
|
||||
function agent(id: string): Agent {
|
||||
return { id } as unknown as Agent
|
||||
}
|
||||
|
||||
function openMux(api: ApiProxy, abort: AbortController): {
|
||||
envelopes: RpcRequest<MuxFrame>[]
|
||||
waitForQuestion(): Promise<RpcRequest<Extract<MuxFrame, { type: 'question/requested' }>>>
|
||||
} {
|
||||
const envelopes: RpcRequest<MuxFrame>[] = []
|
||||
let resolveQuestion!: (value: RpcRequest<Extract<MuxFrame, { type: 'question/requested' }>>) => void
|
||||
const question = new Promise<RpcRequest<Extract<MuxFrame, { type: 'question/requested' }>>>((resolve) => {
|
||||
resolveQuestion = resolve
|
||||
})
|
||||
void (async () => {
|
||||
for await (const envelope of api.events.mux({ rpcId: RpcId('question-mux'), payload: {} }, abort.signal)) {
|
||||
envelopes.push(envelope)
|
||||
if (envelope.payload.type === 'question/requested') {
|
||||
resolveQuestion(envelope as RpcRequest<Extract<MuxFrame, { type: 'question/requested' }>>)
|
||||
}
|
||||
}
|
||||
})()
|
||||
return { envelopes, waitForQuestion: () => question }
|
||||
}
|
||||
|
||||
function answer(
|
||||
envelope: RpcRequest<Extract<MuxFrame, { type: 'question/requested' }>>,
|
||||
selected: string[],
|
||||
custom?: string,
|
||||
): Parameters<ApiProxy['respond']>[0] {
|
||||
return {
|
||||
type: 'client-response',
|
||||
rpcId: envelope.rpcId,
|
||||
result: {
|
||||
ok: true,
|
||||
value: {
|
||||
sessionId: envelope.payload.sessionId,
|
||||
answer: {
|
||||
answers: [{
|
||||
id: envelope.payload.questions[0]?.id,
|
||||
selected,
|
||||
...custom === undefined ? {} : { custom },
|
||||
}],
|
||||
},
|
||||
},
|
||||
},
|
||||
}
|
||||
}
|
||||
|
||||
describe('question response validation', () => {
|
||||
it('accepts selected options with custom text for multi-select questions', async () => {
|
||||
const { ctx, api } = await harness()
|
||||
const abort = new AbortController()
|
||||
const mux = openMux(api, abort)
|
||||
const asked = ctx.userInteraction.ask({
|
||||
agent: agent('session-multi'),
|
||||
questions: [{
|
||||
id: 'targets',
|
||||
question: 'Choose targets and add another',
|
||||
multiSelect: true,
|
||||
options: [{ label: 'Code' }, { label: 'Docs' }],
|
||||
}],
|
||||
})
|
||||
const envelope = await mux.waitForQuestion()
|
||||
|
||||
expect(await api.respond(answer(envelope, ['Code', 'Docs'], 'Release notes')))
|
||||
.toEqual({ accepted: true })
|
||||
await expect(asked).resolves.toEqual({
|
||||
answers: [{ id: 'targets', selected: ['Code', 'Docs'], custom: 'Release notes' }],
|
||||
})
|
||||
expect(mux.envelopes.some(item => item.payload.type === 'question/resolved')).toBe(true)
|
||||
abort.abort()
|
||||
})
|
||||
|
||||
it('keeps selected options and custom text mutually exclusive for single-select questions', async () => {
|
||||
const { ctx, api } = await harness()
|
||||
const abort = new AbortController()
|
||||
const mux = openMux(api, abort)
|
||||
const asked = ctx.userInteraction.ask({
|
||||
agent: agent('session-single'),
|
||||
questions: [{
|
||||
id: 'target',
|
||||
question: 'Choose one target',
|
||||
options: [{ label: 'Code' }, { label: 'Docs' }],
|
||||
}],
|
||||
})
|
||||
const envelope = await mux.waitForQuestion()
|
||||
|
||||
expect(await api.respond(answer(envelope, ['Code'], 'Release notes')))
|
||||
.toEqual({ accepted: false, reason: 'bad-response' })
|
||||
expect(await api.respond(answer(envelope, [], 'Release notes')))
|
||||
.toEqual({ accepted: true })
|
||||
await expect(asked).resolves.toEqual({
|
||||
answers: [{ id: 'target', selected: [], custom: 'Release notes' }],
|
||||
})
|
||||
abort.abort()
|
||||
})
|
||||
})
|
||||
@@ -1,6 +1,6 @@
|
||||
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write
|
||||
README.md: 8e779f4025c20cd200344efb7cb8cd6bc09ba64d
|
||||
README.zh.md: fe1dc5559882532c4f44e705cc6daa2c7f4f8905
|
||||
# pnpm run verify-translation-pairing --write packages/ui/tool-ask-user/README.md
|
||||
README.md: 64da4d75d01a0df0ae51b1557ed1c796317b906f
|
||||
README.zh.md: 8a1eb3ee4f9e9ccc2ea2fe433bf85158c76d3549
|
||||
|
||||
@@ -15,7 +15,7 @@ Model-facing `ask_user_question` tool over `ctx.userInteraction`. It lets the mo
|
||||
- `options` — optional choices with `label` and `description`. If recommending a choice, put it first and append `(Recommended)` to that label.
|
||||
- `multi_select` — whether that question may return more than one selected option.
|
||||
|
||||
The tool calls `ctx.userInteraction.ask()` and returns canonical `{ answers: [{ id, selected, custom? }] }`. `selected` contains option labels; `custom` is present only for a free-form answer and overrides selected choices. The Native renderer preserves the compact JSON text shape `{ "answers": [{ "id": "...", "selected": ["..."], "custom": "..." }] }`.
|
||||
The tool calls `ctx.userInteraction.ask()` and returns canonical `{ answers: [{ id, selected, custom? }] }`. `selected` contains option labels; `custom` carries a free-form answer, supplementing `selected` for a multi-select question and overriding it for a single-select question. The Native renderer preserves the compact JSON text shape `{ "answers": [{ "id": "...", "selected": ["..."], "custom": "..." }] }`.
|
||||
|
||||
## Role
|
||||
|
||||
|
||||
@@ -15,7 +15,7 @@
|
||||
- `options`：可选选项，包含 `label` 和 `description`。如需推荐某个选项，请将其置于首位，并在该标签末尾追加 `(Recommended)`。
|
||||
- `multi_select`：该问题是否可以返回多个选中的选项。
|
||||
|
||||
工具调用 `ctx.userInteraction.ask()`，并返回规范的 `{ answers: [{ id, selected, custom? }] }`。`selected` 包含选项标签；仅当用户自由填写回答时才会出现 `custom`，并覆盖选中的选项。Native renderer 会保留紧凑的 JSON 文本形式 `{ "answers": [{ "id": "...", "selected": ["..."], "custom": "..." }] }`。
|
||||
工具调用 `ctx.userInteraction.ask()`，并返回规范的 `{ answers: [{ id, selected, custom? }] }`。`selected` 包含选项标签；`custom` 携带自由填写的回答，对于多选题会补充 `selected`，对于单选题则会覆盖它。Native renderer 会保留紧凑的 JSON 文本形式 `{ "answers": [{ "id": "...", "selected": ["..."], "custom": "..." }] }`。
|
||||
|
||||
## 职责
|
||||
|
||||
|
||||
@@ -140,7 +140,7 @@ describe('ask_user_question tool', () => {
|
||||
async ask() {
|
||||
return {
|
||||
answers: [
|
||||
{ id: 'targets', selected: ['tests', 'docs'] },
|
||||
{ id: 'targets', selected: ['tests', 'docs'], custom: 'release notes' },
|
||||
{ id: 'notes', selected: [], custom: 'ship today' },
|
||||
],
|
||||
}
|
||||
@@ -168,13 +168,13 @@ describe('ask_user_question tool', () => {
|
||||
if (result.isError) throw new Error('expected ask_user_question success')
|
||||
expect(result.value).toEqual({
|
||||
answers: [
|
||||
{ id: 'targets', selected: ['tests', 'docs'] },
|
||||
{ id: 'targets', selected: ['tests', 'docs'], custom: 'release notes' },
|
||||
{ id: 'notes', selected: [], custom: 'ship today' },
|
||||
],
|
||||
})
|
||||
expect(result.content).toEqual([{
|
||||
type: 'text',
|
||||
text: '{"answers":[{"id":"targets","selected":["tests","docs"]},{"id":"notes","selected":[],"custom":"ship today"}]}',
|
||||
text: '{"answers":[{"id":"targets","selected":["tests","docs"],"custom":"release notes"},{"id":"notes","selected":[],"custom":"ship today"}]}',
|
||||
}])
|
||||
})
|
||||
|
||||
|
||||
@@ -2,5 +2,5 @@
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write packages/ui/tui/README.md
|
||||
README.md: 0b358520b863f0b9ee7a128cf4807f582fc46d8d
|
||||
README.zh.md: 7e89197bd82d16dfbabeb715e953275e2f6dd68b
|
||||
README.md: 3c847828a3d560b85e74809f984bc9ea581e417f
|
||||
README.zh.md: 8872484a5de376e41564756332200198e587d2b3
|
||||
|
||||
@@ -153,7 +153,7 @@ Append-only; newly visible content follows the reusable request prefix and does
|
||||
|
||||
#### What the model sees
|
||||
|
||||
When a consumer calls `ctx.userInteraction.ask()`, this provider presents each question in order and returns selected option labels or `custom` text. Abort, cancellation, or UI disposal becomes `Error: ask_user_question was interrupted before the user answered` through `dsh-tool-ask-user`.
|
||||
When a consumer calls `ctx.userInteraction.ask()`, this provider presents each question in order and returns selected option labels, `custom` text, or both for a multi-select question. Abort, cancellation, or UI disposal becomes `Error: ask_user_question was interrupted before the user answered` through `dsh-tool-ask-user`.
|
||||
|
||||
#### Token effect
|
||||
|
||||
|
||||
@@ -153,7 +153,7 @@ Paths prefixed with @ are files explicitly referenced by the user. Use the read
|
||||
|
||||
#### 模型看到的内容
|
||||
|
||||
消费方调用 `ctx.userInteraction.ask()` 时，此提供方会按顺序显示各个问题，并返回选中选项标签或 `custom` 文本。中止、取消或 UI dispose 会变为 `Error: ask_user_question was interrupted before the user answered`；该转换由 `dsh-tool-ask-user` 完成。
|
||||
消费方调用 `ctx.userInteraction.ask()` 时，此提供方会按顺序显示各个问题，并返回选中选项标签、`custom` 文本，或为多选题同时返回两者。中止、取消或 UI dispose 会变为 `Error: ask_user_question was interrupted before the user answered`；该转换由 `dsh-tool-ask-user` 完成。
|
||||
|
||||
#### Token 影响
|
||||
|
||||
|
||||
@@ -799,12 +799,14 @@ export class QuestionDialog implements Component, Focusable {
|
||||
if (this.selected.has(this.selectedIndex)) this.selected.delete(this.selectedIndex)
|
||||
else this.selected.add(this.selectedIndex)
|
||||
} else if (matchesKey(data, Key.enter)) {
|
||||
const indices = this.question.multiSelect ? [...this.selected].sort((a, b) => a - b) : [this.selectedIndex]
|
||||
if (indices.length === 0) {
|
||||
const selected = this.question.multiSelect
|
||||
? this.selectedOptionLabels()
|
||||
: [options[this.selectedIndex]?.label].filter((label): label is string => label !== undefined)
|
||||
if (selected.length === 0) {
|
||||
this.error = 'Select at least one option, or press Tab for a custom answer.'
|
||||
return
|
||||
}
|
||||
this.done({ selected: indices.map(index => options[index]?.label).filter((label): label is string => label !== undefined) })
|
||||
this.done({ selected })
|
||||
} else if (matchesKey(data, Key.tab) || data.toLowerCase() === 'c') {
|
||||
this.mode = 'custom'
|
||||
this.error = ''
|
||||
@@ -819,7 +821,17 @@ export class QuestionDialog implements Component, Focusable {
|
||||
this.error = 'Enter an answer before submitting.'
|
||||
return
|
||||
}
|
||||
this.done({ selected: [], custom })
|
||||
this.done({
|
||||
selected: this.question.multiSelect ? this.selectedOptionLabels() : [],
|
||||
custom,
|
||||
})
|
||||
}
|
||||
|
||||
private selectedOptionLabels(): string[] {
|
||||
return [...this.selected]
|
||||
.sort((a, b) => a - b)
|
||||
.map(index => this.options[index]?.label)
|
||||
.filter((label): label is string => label !== undefined)
|
||||
}
|
||||
|
||||
render(width: number): string[] {
|
||||
|
||||
@@ -4436,8 +4436,12 @@ describe('TUI user-interaction dialogs', () => {
|
||||
result.terminal.send(' ')
|
||||
result.terminal.send('\x1b[B')
|
||||
result.terminal.send(' ')
|
||||
result.terminal.send('\t')
|
||||
result.terminal.send('Tests')
|
||||
result.terminal.send('\r')
|
||||
await expect(multi).resolves.toEqual({ answers: [{ id: 'targets', selected: ['Code', 'Docs'] }] })
|
||||
await expect(multi).resolves.toEqual({
|
||||
answers: [{ id: 'targets', selected: ['Code', 'Docs'], custom: 'Tests' }],
|
||||
})
|
||||
|
||||
const custom = result.ctx.userInteraction.ask({
|
||||
questions: [{ id: 'other', question: 'Choose or type', options: [{ label: 'Default' }] }],
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Bilingual-pair consistency record (docs/i18n/README.md): the git blob hash of each
|
||||
# side as of the last confirmed-consistent state. Both languages carry equal authority;
|
||||
# after editing either side, bring the other along and re-record with:
|
||||
# pnpm run verify-translation-pairing --write
|
||||
README.md: d234d6677bdd772f1bbd2c979c0d41f90aef5c32
|
||||
README.zh.md: b70a61d6491e0bb0e52215cdeaeea3d728f7f153
|
||||
# pnpm run verify-translation-pairing --write packages/ui/user-interaction/README.md
|
||||
README.md: 2ff29f5fd6244ebcf7e29b86f5de1cde30944532
|
||||
README.zh.md: 7d0d1b06db5be4353e42d1905c71d5dff963b97d
|
||||
|
||||
@@ -19,7 +19,7 @@ Abstract user-interaction seam. It owns `ctx.userInteraction`, the service a mod
|
||||
- `UserInteractionProvider` — UI implementation with `ask(request)`.
|
||||
- `UserInteractionError` — `HarnessError` subclass with codes such as `EMPTY_QUESTIONS`, `NO_PROVIDER`, `DUPLICATE_PROVIDER`, and `ASK_ABORTED`.
|
||||
|
||||
When an answer includes `custom`, `selected` is empty; custom text is an override rather than a supplement to selected choices. A UI may preserve a skipped item as `{ id, selected: [] }`, keeping the existing answer shape while retaining other answers in the batch.
|
||||
For a single-select question, `custom` overrides the selected choice and `selected` is empty. For a multi-select question, `custom` may supplement the labels in `selected`. A UI may preserve a skipped item as `{ id, selected: [] }`, keeping the existing answer shape while retaining other answers in the batch.
|
||||
|
||||
## Role
|
||||
|
||||
|
||||
@@ -19,7 +19,7 @@
|
||||
- `UserInteractionProvider`：包含 `ask(request)` 的 UI 实现。
|
||||
- `UserInteractionError`：`HarnessError` 的子类，包含 `EMPTY_QUESTIONS`、`NO_PROVIDER`、`DUPLICATE_PROVIDER` 和 `ASK_ABORTED` 等代码。
|
||||
|
||||
当回答包含 `custom` 时，`selected` 为空；自定义文本会覆盖所选选项，而不是补充它们。UI 可以把跳过的条目保留为 `{ id, selected: [] }`，既维持现有回答形态，也保留该批次中的其他回答。
|
||||
对于单选题，`custom` 会覆盖选中的选项，且 `selected` 为空。对于多选题，`custom` 可以补充 `selected` 中的标签。UI 可以把跳过的条目保留为 `{ id, selected: [] }`，既维持现有回答形态，也保留该批次中的其他回答。
|
||||
|
||||
## 职责
|
||||
|
||||
|
||||
@@ -33,7 +33,7 @@ export interface AskUserQuestionItem {
|
||||
export interface AskUserQuestionAnswerItem {
|
||||
/** The answered question id. */
|
||||
id: string
|
||||
/** Selected option labels. Empty for custom or unanswered choices. */
|
||||
/** Selected option labels. May accompany custom text for a multi-select question. */
|
||||
selected: string[]
|
||||
/** Optional free-text "Other" answer. */
|
||||
custom?: string
|
||||
|
||||
Reference in New Issue
Block a user