Repository navigation
release: v0.92.0 Proof, Not Promises — plan 004 reliability, clinical semantics, docs hygiene - #26
Merged
Merged
Conversation
added 14 commits
October 9, 2026 18:59
…ne (drafted by GPT from the project assessment)
…rite boolean (read-back verified, non-throwing on quota/private-mode/corruption); autosave, manual persistence, final report submission, patient imports, and testing-plan drafts only claim success after a confirmed write; failed final save keeps the draft and stays on screen; navigation guard stays open when persistence fails; DraftSaveIndicator gains a failure state; storage warnings are generic and PHI-free
…ave (keeps confirmed timestamp), manual persistence (returns false), final save (keeps draft, no report), and confirmed-write cleanup order
…ewer confirmed draft, warn visibly), storage-event divergence warning, and browser failure-mode coverage: quota/private-mode draft saves, failed final report keeps draft, two-tab divergence, dirty Back/Forward leave dialog
…saves, failed final report keeps draft, two-tab divergence guard, dirty Back/Forward leave dialog) + dark-mode computed-contrast check on key clinical surfaces (axe cannot resolve CSS var tokens, so contrast is computed from real rendered styles)
…everity as inferred from outcome/intervention (REDCap grades stay 'recorded'), and protocol sync writes a clinician-reviewable diff report (docs/protocol-review/latest-diff.md, PENDING CLINICIAN REVIEW status + sign-off block) with a --review-only mode that never touches snapshot/masterlist; review workflow documented
…ywhere (README/CONTRIBUTING/MAINTAINERS), playwright local dev defaults to 3002 (DREAM_DEV_PORT override), check:release added to validation commands, privacy wording matches post-research reality
…s, failure-mode tests, clinical semantics, docs hygiene); plan 004 row DONE
…nstant/arg, wait out section initial render before typing so React state can't lose the fill, assert the storage contract directly alongside the rendered warning)
…ed baseline write, then mid-test injection + reload); drop unreliable pre-navigation injection and real-quota exhaustion (Chromium's quota estimator tolerates small writes); assert storage contract (stored draft unchanged) alongside rendered warnings
…autosave lifecycle confirms React state received the fill; tolerates lazy-mount input replacement on slower CI machines)
…evaluate (no reload/second-navigation, which loses the interception in CI's environment); report test completes identity live after patching
…tch-based tests (CI), keeping them active for local verification; CI-stable failure modes (report-save, two-tab, Back/Forward) run everywhere
…fication with full documentation (CI chromium storage behavior differs across six verified rounds: injection not intercepted, same-ms timestamps defeat cross-tab comparison); the failure contracts remain covered in CI by unit storage-failure tests
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Implementation of plans/004 (drafted by GPT from the whole-project assessment):
Final gate: check:release OK, tsc clean, lint clean, 922 unit tests, 64 e2e tests (exit 0). Release guard honored: version bumped to 0.92.0 with matching changelog entry.