savethenurse/ai-media-hub
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Document rollback after search regression
build-push / docker (push) Successful in 4m9s
Details

Browse Source
This commit is contained in:
AI Assistant
2026-03-17 13:22:10 +09:00
parent 3f824c4bdf
commit 4844394334
1 changed files with 22 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
TODO.md
+22
View File
@@ -33,6 +33,7 @@
- Result modal sizing is now being constrained to the viewport, and modal-only source-summary translation is now part of the active implementation path. - Result modal sizing is now being constrained to the viewport, and modal-only source-summary translation is now part of the active implementation path.
- Card summaries now also translate lazily to Korean, and Gemini negative-assessment handling now drives stronger follow-up search behavior than before. - Card summaries now also translate lazily to Korean, and Gemini negative-assessment handling now drives stronger follow-up search behavior than before.
- Search preview delivery is now moving away from persistent on-disk preview caching toward live proxy / live transcode behavior, with Google Video preview reuse added to result cards and modal playback. - Search preview delivery is now moving away from persistent on-disk preview caching toward live proxy / live transcode behavior, with Google Video preview reuse added to result cards and modal playback.
- The latest search-breadth / modal-fitting experiment from `5ca7aef` has been rolled back after live regression was confirmed.
## Current Architecture ## Current Architecture
- `backend/main.go` - `backend/main.go`
@@ -231,6 +232,7 @@
- Source Summary translation now depends on Google Translate HTTP availability; frontend silently falls back to original summary text if translation fails. - Source Summary translation now depends on Google Translate HTTP availability; frontend silently falls back to original summary text if translation fails.
- The result modal should now stay within viewport height, but this still needs real browser confirmation on multiple short-height displays because CSS-only constraints were the source of the latest user-visible regression. - The result modal should now stay within viewport height, but this still needs real browser confirmation on multiple short-height displays because CSS-only constraints were the source of the latest user-visible regression.
- Artgrid preview playback now has a server-side ffmpeg transcode path for `.m3u8` style preview URLs, but this trades storage savings for runtime CPU cost. - Artgrid preview playback now has a server-side ffmpeg transcode path for `.m3u8` style preview URLs, but this trades storage savings for runtime CPU cost.
- The reverted `5ca7aef` experiment showed that simply widening collector caps and Gemini candidate count can backfire when the added candidates are weak: final visible count fell sharply even though backend raw candidate count increased.
- The local self-test script is better than before, but it is still a smoke test, not full integration coverage. - The local self-test script is better than before, but it is still a smoke test, not full integration coverage.
## Current Risks Around Search Quality ## Current Risks Around Search Quality
@@ -551,6 +553,7 @@
## Highest-Value Next Steps ## Highest-Value Next Steps
- [ ] Reduce `/api/search` latency further without collapsing result count - [ ] Reduce `/api/search` latency further without collapsing result count
- [ ] Rebuild the reverted search-expansion work from the previous stable baseline, but only after measuring where candidate quality collapses between ranked pool and final merge
- [ ] Build a repeatable repo-local bootstrap script or documented setup command set for non-root machines so fresh PC setup does not depend on shell history - [ ] Build a repeatable repo-local bootstrap script or documented setup command set for non-root machines so fresh PC setup does not depend on shell history
- [ ] Improve Envato / Artgrid preview acquisition reliability so Gemini Vision sees real frames more often - [ ] Improve Envato / Artgrid preview acquisition reliability so Gemini Vision sees real frames more often
- [ ] Browser-verify the new result modal at multiple viewport heights and confirm translated Source Summary readability on real long descriptions - [ ] Browser-verify the new result modal at multiple viewport heights and confirm translated Source Summary readability on real long descriptions
@@ -621,6 +624,25 @@
- If behavior in the browser does not match the latest backend/frontend code, the first assumption should be stale frontend assets until proven otherwise - If behavior in the browser does not match the latest backend/frontend code, the first assumption should be stale frontend assets until proven otherwise
## Recent Change Log ## Recent Change Log
- Date: `2026-03-17`
- What changed:
- Reverted commit `5ca7aef` (`Strengthen search breadth and modal fitting`) to restore the previous stable search/modal baseline.
- Revalidated the rollback state locally.
- Why it changed:
- The user reported that search results became too sparse again and that `search returned partial results to avoid gateway timeout` reappeared.
- The provided log `ai-media-hub-2026-03-17T04-19-08-889Z.log` showed:
- backend candidate pool still reaching `30`
- Gemini candidate cap widened to `24`
- `visualRejectCount: 12`
- final visible result count collapsing to `5`
- That indicates the widened experiment increased weak/low-visual candidates and pushed the request back toward deadline pressure without improving visible result count.
- How it was verified:
- `go test ./...`
- `bash scripts/selftest.sh`
- What is still risky or incomplete:
- This rollback restores the earlier baseline but does not solve the underlying request to improve visible result count.
- The next attempt should start from the reverted baseline and target the actual bottleneck between ranked pool and final merge, instead of only widening raw query breadth.
- Date: `2026-03-17` - Date: `2026-03-17`
- What changed: - What changed:
- If the first search pass plus Gemini filtering still leaves too few visible results, the backend now performs an additional coverage-expansion search/evaluation pass before final fallback filling. - If the first search pass plus Gemini filtering still leaves too few visible results, the backend now performs an additional coverage-expansion search/evaluation pass before final fallback filling.