Commit Graph

43 Commits

Author SHA1 Message Date
Woody 1e8773469e Merge branch 'Phase4-dev' 2026-05-14 23:29:42 +08:00
Woody 7bff4308b7 feat: Phase 4 — System Audio & Listen Mic capture into ASR → RAG
Adds two new live audio sources alongside file Upload:

- System Audio: getDisplayMedia() captures system/tab audio output,
  pipes through WebSocket → DashScope realtime ASR → RAG.
- Listen Mic: getUserMedia() captures microphone input via the same
  audio pipeline (shared useMediaStreamASR hook).

Backend: feature toggles (system_audio_enabled, mic_enabled) in
config.py, source query param gating in ws_asr.py, 10 config tests.

Bug fix: getDisplayMedia() rejected video:false per W3C spec —
changed to video:true then stop video tracks to allow audio-only
capture on Windows/macOS Chrome.
2026-05-14 22:55:06 +08:00
Woody a8a2cc0940 fix: enable Half Question/Final Submit during interim ASR text
isDisabled, handleSubmit, and Half Question onClick all checked
question.trim() instead of displayValue.trim(). Since question state
is only updated on onFinalTranscript (complete sentences), interim
ASR delta text shown in the textarea via partialText was invisible
to the disabled check — buttons stayed disabled until sentence end.

Fix: use displayValue which includes partialText when user hasn't typed.
2026-05-14 21:55:07 +08:00
Woody 17db487dbb feat: Phase 3 — Half Question button, Final Submit rename, ASR text always black
- Backend: add stop_after_decompose flag to QueryRequest, early-return
  after decomposition in SSE stream with half_question:true event
- Frontend: add decomposeOnly method to useQueryDocumentStream hook
- QueryInput: remove grey italic from ASR partial text, rename Submit
  to Final Submit, add gray Half Question button that decomposes
  without clearing querybox text
- LTTPage: wire handleHalfQuestion to decomposeOnly
2026-05-14 21:27:21 +08:00
Woody b05c361fbd revert: remove Phase 3 YouTube proxy — all 7 sub-phases
Reverts commits 284028b through b4096d6. Phase 4 (System Audio Capture)
will replace the YouTube use case with a more versatile getDisplayMedia approach.

Removed: YouTube router, HLS proxy, YouTubeService, YouTubeInput,
YouTubeVideoPlayer, useYouTubeASR hook, all Phase 3 tests, hls.js dep,
YouTube config fields, YouTube README/plan sections.

Modified files restored to pre-Phase-3 state: LTTPage (no source toggle),
api.ts (no YouTube extract), types (no YouTube types), config.py (no
youtube fields), main.py (no YouTube router), requirements.txt (no yt-dlp),
.env.example (no YouTube vars), package.json (no hls.js).

Relevant Phase 2 code preserved: ws_asr.py (unchanged), useVideoASR,
VideoPlayer, VideoUpload, QueryInput, Full Transcript.
2026-05-09 21:07:21 +08:00
Woody a8eea54c0f feat: Phase 3.4 — YouTube Input + Video Player frontend components
- YouTubeInput.tsx: URL input with validation (youtube.com/watch, youtu.be, /live/, /shorts/),
  loading/error states, Load Stream button, uses useYouTubeExtract mutation
- YouTubeVideoPlayer.tsx: dual hls.js (video + hidden audio), forwardRef,
  thumbnail placeholder until play, LIVE badge, quality capped ≤480p,
  onAudioReady callback for ASR hook exposure, dynamic import('hls.js')
- Types: YouTubeFormat, YouTubeStreamResponse interfaces
- API: extractYouTubeStream() — POST /youtube/extract
- Query: useYouTubeExtract() TanStack Query mutation hook
- Tests: 16 new (7 YouTubeInput, 9 YouTubeVideoPlayer)
- 171/171 total pass (zero regressions)
- Updated plan: 3.4 marked Complete, 4/7 sub-phases done
2026-05-09 16:43:42 +08:00
Woody cb0ac07786 fix: text accumulation — stashes are sliding windows, merge via overlap detection
DashScope stashes are ~7-char rolling windows, not cumulative. Each partial
event replaces the previous. Completed events rarely sent. This caused text to
jump/replace during streaming and disappear on pause.

Backend:
- Add _merge_stash() — finds overlapping suffix between successive stashes
  and appends only new characters, reconstructing full utterance from partials
- format_transcription_event returns raw stash for read_events to merge
- read_events maintains partial_buffer via _merge_stash, clears on completed
- Guard against empty/whitespace-only stashes

Frontend:
- transcriptRef + onFinalTranscriptRef avoid stale closures in pause handler
- stopStreaming fires onFinalTranscript(currentText) before clearing partial
- Removed blind setPartialTranscript('') that erased text on pause

Tests: 16/16 ws_protocol tests pass, frontend tests unchanged
Plan: Updated phase2_implementation_plan.md to Complete with 11-bug log
2026-05-06 20:06:39 +08:00
Woody fcb9ec1f6c fix: Phase 2 ASR pipeline — 9 bugs resolved, Full Transcript works end-to-end
- Vite proxy: forward /api and /ws to backend port 8000
- WebSocket URL: use backend host, not Vite HMR port
- LTTPage: callback ref replaces useRef (video element always null before)
- ws_asr: pass DashScope API key to OmniRealtimeConversation
- asr_client: fix data_url MIME type (audio/wav), omit extra_body when auto
- useFullTranscript: use absolute URL prefix for fetch
- QueryInput: add value prop for external Full Transcript injection
- QueryInput: fix displayValue || logic (partialText '' overrode question)
- ffmpeg: install static binary for audio extraction
- Integration tests: 7 tests (upload→transcribe flow)
- Acceptance tests: real DashScope tests (skippable)
- Structured logging: ws_asr.py + video.py
2026-05-06 18:26:17 +08:00
Woody f3b94381ae feat: Phase 2.5 video player, upload UI, and LTTPage layout refactor
- VideoUpload: native drag-and-drop with axios progress bar, file validation
- VideoPlayer: forwardRef wrapper for <video> element (used by useVideoASR)
- LTTPage: replaced VideoPlaceholder, wired useVideoASR/useFullTranscript,
  Full Transcript button, resizable left/right panels (min 30%)
- Tests: 25 new (VideoUpload 8, VideoPlayer 7, LTTPage integration 10)
2026-05-06 14:31:27 +08:00
Woody a4e067822b feat: Phase 2.3 ASR proxy + full transcript and 2.4 frontend hooks
- Backend: DashScope WebSocket proxy (/ws/asr/{video_id}), DashScopeCallback
  sync-to-async bridge, ffmpeg audio extraction, POST /video/{id}/transcribe
- Frontend: useVideoASR hook (auto on play), useFullTranscript hook,
  QueryInput partialText prop, VideoUploadResponse types, uploadVideo API
- Tests: 41 backend + 26 frontend = 67 new tests, all passing
2026-05-06 13:41:24 +08:00
Woody 76c3bec2ab feat: configurable SubQuestions via Step 1.2 system prompt page
- Split 'Step 1: Query Decomposition' into Step 1.1 (prompt template) and Step 1.2 (format config with description + max_length)

- Add create_subquestions_model() and parse_decompose_format() to decompose.py

- QueryDecomposer reads decompose_format from DB, creates dynamic Pydantic model at runtime

- PromptEditor renders Step 1.2 as textarea (description) + number input (max_length 1-5)

- Graceful fallback to static SubQuestions when decompose_format unavailable
2026-05-04 17:22:14 +08:00
Woody 90269608bc fix: display highlight tracking data in history page UI
- Add highlight_prompt, highlight_response, highlight_time_ms to QueryHistoryDetail type
- Add 'Highlights' bar segment with pink color in TimingBar component
- Pass highlightTimeMs to TimingBar in HistoryCard expanded view
- Add collapsible sections for highlight prompt and response in HistoryCard detail
2026-04-29 13:42:08 +08:00
Woody 41f59b396f feat: track highlight generation prompt, response, and timing in history (Phase 5.5)
- Add 3 columns to query_history: highlight_prompt, highlight_response, highlight_time_ms
- HistoryService.update_highlights() updates existing row after batch LLM call
- ChunkHighlightService measures timing, captures prompt and structured JSON response
- SSE completed event includes history_id for frontend to pass back
- Frontend captures historyId, passes as ?history_id= query param in batch POST
- Highlight time tracked separately (excluded from total_time_ms)
- All 153 tests pass (108 backend + 45 frontend)
2026-04-29 11:18:21 +08:00
Woody b47e37f39b fix: use absolute backend URL for highlight API calls
- Vite dev server doesn't proxy /api/v1/v2/ paths to backend
- Changed fetch URL and getHighlightUrl to use http://localhost:8000
- Fixed inline citation highlight URLs in buildCitationUrl
- Cleaned up debug code
2026-04-29 10:39:01 +08:00
Woody bcf4a853bf feat: add highlight status toast notification (Phase 5.4)
- Shows 'Preparing highlights...' (amber spinner) while LLM batch runs
- Shows 'Highlights ready' (green) for 4 seconds when batch completes
- Fixed position top-left corner, auto-dismisses
2026-04-29 10:00:54 +08:00
Woody 1c490ce2fa fix: inline citations now upgrade to highlighted view (Phase 5.4)
- Added sub_question_text to frontend SourceMetadata type
- SubQuestionSection enriches sources with parent sub-question text
- buildCitationUrl routes to highlight page when sub_question_text present
- processCitations threads highlightReadyKeys through inline citations
2026-04-29 09:54:40 +08:00
Woody c632b9ea3b feat: cited source extraction, background batch trigger, and View PDF link upgrade (Phase 5.4.6-5.4.8)
- citationParser.ts: extractCitedSources() parses answer text for [citations],
  resolves against SourceMetadata, returns deduplicated cited sources
- ResponsePanel.tsx: useEffect fires POST /api/v1/v2/highlights/batch after
  answer renders; View PDF link upgrades in-place to highlighted HTML when
  batch completes; stays as raw PDF on failure
- Updated plan: LLM-based relevance detection, eager background computation,
  single batched LLM call, sqlite cache, regex sentence splitter
- 45 frontend tests: 28 citationParser + 17 ResponsePanel (including 4 new
  sub-question highlight tests)
2026-04-29 09:27:04 +08:00
Woody 4058c7dffe fix(citations): use all sub-question sources for citation lookup
LLMs may cite chunks from one sub-question's context inside another sub-question's answer section. Previously, processCitationsForSubq only looked up the current sub-question's sources, leaving cross-referenced citations unlinked. Now SubQuestionSection passes all sub-question sources and uses processCitations with a combined flat lookup.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-28 17:10:02 +08:00
Woody c43cb372e9 feat: integrate bullet points in ResponsePanel with CSS list-style
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-28 16:43:30 +08:00
Woody aa5f716578 feat(upload): support multiple file upload on RAG Database page 2026-04-28 13:22:25 +08:00
Woody 23796d6a0c feat(prompts): add JSON export/import for profile prompt configurations 2026-04-27 19:44:35 +08:00
Woody a7a22f1494 fix(relevance): tolerate LLM score count mismatches via padding instead of discarding
The per-sub-question filter was all-or-nothing: if the LLM returned
9 scores for 10 chunks (common with qwen3.5-35b), every chunk was
discarded and the user got 'no relevant information found'.

Now: fewer scores → pad with 0.0; more scores → truncate. Changed
from error→warning since this is recoverable.

Also improve LTT page UI: sources collapsed by default in per-sub-q
sections, and the 'Your question' text now shows the full question
instead of being truncated.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-27 14:31:18 +08:00
Woody 3b868a0133 feat(prompts): integrate filter_per_subq with PromptService, fix seed bugs, restructure UI
Break the hardcoded per-sub-q filter prompt into 3 editable PromptService templates (filter_intro, filter_section, filter_outro) with placeholders for the for-loop iteration pattern. Refactor RelevanceFilter._build_per_subq_prompt() to compose them at runtime, falling back to built-in defaults when PromptService is unavailable.

Fix two latent bugs from Package 4:
- generate_per_subq was called by rag.py but never added to _VALID_STEPS or DB seed (would ValueError at runtime)
- _SEED_GENERATE placeholder mismatch: flat generate_response() expects {question}/{context} but Package 4 changed it to {context_sections}. Restored flat template; generate_per_subq now holds {context_sections}.

Add database backfill migration in seed_default_profiles() to INSERT OR IGNORE missing steps into existing profile rows, ensuring all 7 steps exist on restart.

Restructure System Prompts UI: remove unused flat filter/generate steps, replace with Step 2.1-2.3 (filter_intro/section/outro) and Step 3 (generate_per_subq). Update PlaceholderDocs with {context_sections}, {subq_idx}, {subq_question}.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-27 11:14:27 +08:00
Woody 7d072e5ea1 feat(frontend): redesign ResponsePanel for per-sub-question sections with grouped sources
Redesign ResponsePanel with SubQuestionSections component that parses answer markdown on ## Sub-question N: boundaries and renders per-sub-question cards with headers, ReactMarkdown body, and collapsible sources scoped to each section. Extract FlatResponse for backward compatibility when subQuestionSources is null. Add buildCitationLookupForSubq and processCitationsForSubq for per-sub-question citation lookup scope isolation. Add anchor links in ExtractedQuestionsDisplay that smooth-scroll to matching ResponsePanel sections. Pass subQuestionSources through LTTPage.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-26 23:29:51 +08:00
Woody 0d3e8ce0ce fix(frontend): save button always disabled on System Prompts page
Root cause: PromptEditor useEffect synced localPrompts back to match prompts after every keystroke, making isDirty() always false.

- Delegate disabled control to parent via hasChanges prop (no local sync)

- Derive currentPrompts synchronously to avoid empty-textarea flash

- Add key={selectedProfile} for clean remount on profile switch

- Update PromptEditor tests for new hasChanges prop

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-26 18:48:52 +08:00
Woody 9f41a328e3 fix(frontend): remove duplicate History link from NavBar
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-26 18:48:31 +08:00
Woody d7cf785452 feat(frontend): Phase 3.6 — History page with timing bars, expandable cards, and pagination 2026-04-26 13:19:52 +08:00
Woody 8e6597a86e feat(frontend): Phase 3.3 — System Prompt Configuration page
- SystemPromptsPage: profile selector, activation, edit with TanStack Query
- ProfileList: 3 profile cards (A/B/C) with active indicator + edit button
- PromptEditor: 3 monospace textareas, placeholder badges, char count,
  unknown placeholder warnings, per-step reset (↺), action bar
- PlaceholderDocs: info box showing {question}/{chunks}/{context}
- Data layer: +7 types, +6 API functions, +6 TanStack Query hooks
- Routing: /system-prompts route + NavBar link
- Tests: 27 tests (PlaceholderDocs 6, ProfileList 7, PromptEditor 14)
- 0TS errors, 27/27 tests pass, 1 pre-existing e2e failure (unrelated)
2026-04-25 21:26:42 +08:00
Woody 3b741c1844 feat(query): stream extracted questions immediately via SSE
Convert /query endpoint from synchronous JSON to Server-Sent Events (SSE)
streaming. The frontend now receives extracted_questions as soon as the
first LLM call completes, without waiting for retrieval, filtering, and
answer generation.

Backend:
- Add StreamingQueryEvent union type (Decomposed, Retrieving, Filtering,
  Generating, Completed, Error)
- Convert /query to return StreamingResponse with SSE format
- Yield events after each pipeline phase

Frontend:
- Add queryDocumentStream() using fetch + ReadableStream
- Add useQueryDocumentStream() hook with phase-aware state
- Update LTTPage to use streaming instead of mutation
- Update ResponsePanel to show phase messages (Searching documents...,
  Filtering passages..., Generating answer...)
- Update ExtractedQuestionsDisplay to accept null

Tests:
- Update query_flow e2e test to mock queryDocumentStream
- 84/85 tests pass (1 pre-existing failure from removed file-input)
2026-04-25 18:29:22 +08:00
Woody f07e14aafd feat(frontend): render inline citations as clickable PDF links (sub-phase 2.6)
ResponsePanel now calls processCitations() on answer text before rendering.
Custom ReactMarkdown 'a' component adds target="_blank" to all citation links.
Adds tests for citation link rendering and unmatched citation fallback.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-24 17:53:25 +08:00
Woody a3028df4e1 feat(frontend): update View PDF links to open in-browser viewer (sub-phase 2.5)
ResponsePanel and ChunkList View PDF links now open /pdf-viewer page instead of raw PDF download.
Update ChunkList test mock from getChunkPdfUrl to getPdfViewerUrl.
Add DOMMatrix polyfill to test setup for react-pdf/jsdom compatibility.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-24 17:09:50 +08:00
Woody 467d88489c test(frontend): update tests for extracted questions and inline question display (sub-phase 2.2/2.3)
Replace KeywordsDisplay test with ExtractedQuestionsDisplay test. Update e2e mock data for extracted_questions. Fix QueryInput to show submitted question inline with submit button.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-24 16:24:49 +08:00
Woody 4c51758348 feat(frontend): replace KeywordsDisplay with ExtractedQuestionsDisplay (sub-phase 2.3)
Delete KeywordsDisplay (blue pills) and create ExtractedQuestionsDisplay (numbered list). Rename keywords to extracted_questions in types and LTTPage.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-24 16:24:30 +08:00
Woody 87acb8816a feat(frontend): display submitted question below input (sub-phase 2.2)
Show submitted question as italic text below the input area after clicking submit. Clears when user starts typing a new question.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-24 15:56:35 +08:00
Woody d49756f374 feat: add chunk PDF serving endpoint and frontend clickable source links (1.5.6)
- Add page_number and chunk_file_path to SourceMetadata model and query router
- Add GET /chunks/{file_path}/pdf endpoint with path traversal protection
- Add View PDF links in ResponsePanel source cards and ChunkList component
- Update TypeScript types and API helper for chunk PDF URLs
- Add backend tests (5) and frontend ChunkList tests (7)
- Update enhancement plan: all 3 features complete
2026-04-24 11:49:39 +08:00
Woody c10318b7f7 feat(frontend): add RAG Database management page with document CRUD UI
Sub-phase 1.5.3: Full RAG Database page with document listing, expandable chunk viewer, delete with confirmation, and document upload. Adds TypeScript types, API functions, TanStack Query hooks (useQuery + useMutation with cache invalidation), and three new components (DocumentList, ChunkList, DocumentUpload).

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-24 09:41:56 +08:00
Woody 52c09b86cb feat(frontend): add nav bar with routing, markdown rendering, and enhancement plan
- Add react-router-dom with NavBar component (LTT + RAG Database tabs)
- Extract AppContent into LTTPage, add RAGDatabasePage placeholder
- Refactor App.tsx to BrowserRouter + Routes layout
- Switch ResponsePanel to react-markdown for rich formatting
- Fix ResponsePanel test for markdown rendering
- Update RAG prompt to cite source name instead of number
- Save Phase 1 enhancement plan (.plans/phase1_enhancement_plan.md)
2026-04-23 18:37:30 +08:00
Woody e927e5fc60 feat(frontend): polish styling, spacing, and add e2e integration tests for Phase 1.3
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-23 11:46:53 +08:00
Woody f6618fd57e feat(frontend): polish ResponsePanel with collapsible sources, copy button, and enhanced skeletons
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-23 11:46:40 +08:00
Woody 864b684d32 feat(frontend): add PipelineProgress component with 4-stage stepper and tests
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-23 11:46:23 +08:00
Woody a7d5dc610a feat(frontend): add IngestPanel and ErrorBoundary components with tests
IngestPanel: file upload for PDF/DOCX/TXT with progress and success/error feedback.

ErrorBoundary: React error boundary with fallback UI and reload button.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-23 11:24:29 +08:00
Woody 3d76b894cb feat(frontend): add ResponsePanel component with bullet-point rendering and tests
Displays bullet-point answer with source metadata cards. Handles empty, loading, error, and success states.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-23 11:23:54 +08:00
Woody fa94b7c9a3 feat(frontend): add QueryInput and KeywordsDisplay components with tests
QueryInput: textarea with submit button, loading state, Enter-to-submit, clears on submit.

KeywordsDisplay: keyword chips with loading skeletons, animated entrance.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-04-23 11:23:08 +08:00