4 Commits
Author SHA1 Message Date
ClaudeandClaude Fable 5.2 c461305b36 fix: chat status animation and streamed answer stop stuttering and jumping
Measured in headless Chromium against the real transcript, scroll hook and CSS.

- Status highlight: transform-only sweep (compositor) instead of
  background-position under background-clip:text, which repainted the label
  on the main thread every frame; labels share one sweep phase so a reworded
  step no longer restarts it.
- Step timeline above the answer: folds to its summary line once the answer
  has text, and folds away over 180ms when the run ends instead of being
  dropped between two frames (the answer jumped up 164-260px); a half-written
  heading is no longer a step.
- Streamed markdown: unclosed ** is bold as written, a table is held until
  its delimiter row, bare markers and the half-typed 判断依据 heading wait.
- Answers render as memoised blocks, streaming and settled alike: a frame
  parses only the tail, settling does not rebuild the answer's DOM.
- The report layer (from the first H2) is written out while streaming; it
  used to stay empty until the run ended.
- Follow mode scrolls in the resize callback, before paint; each new line
  used to be painted once below its place.

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DbBXp2Vp62Ur3C7yRWQdSq
2026-11-16 10:31:52 +08:00
jesse-ux 58ccafb67c fix(rectification): stop re-parsing settled messages on every stream frame
Independent Staging Quality Gate / validate (push) Failing after 6m18s
Independent Staging Quality Gate / publish (push) Skipped
2026-09-16 07:13:12 +08:00
jesse-ux 8144fca27d fix(chat): fold natal reports and drop homepage topic cards
Independent Staging Quality Gate / validate (push) Failing after 9m47s
Independent Staging Quality Gate / publish (push) Skipped
Homepage no longer waits on /api/onboarding for six starter questions.
Natal answers show the spoken layer first; the Level 2 skeleton sits in a
collapsed 完整分析 block. Wide tables scroll sideways on a phone. Form
inputs are 16px so iOS does not zoom on focus.
2026-09-15 12:27:16 +08:00
Jesse_ChenandClaude Fable 5.1 ad9dba5c79 perf(chat): coalesce stream events per frame and pace text release
Every NDJSON event used to commit its own React update and re-parse the
whole partial answer through react-markdown, so long replies grew
quadratically slower. Stream events now land in a frame buffer that
flushes at most once per animation frame, releases answer and thinking
text at a steady pace with a twelve-frame catch-up, and settles
synchronously on completion, failure and abort. Streaming markdown is
split at the last completed block so only the tail is re-parsed each
frame. Applied to both the consultation hook and the rectification chat.

BUG-473

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01JUei7K13cYxLHE3Axe4A45
2026-09-02 04:29:01 +00:00