A card's notes for tonight are sent when the field goes away mid-save (#536) #1035

Merged

claude-bot merged 2 commits from fix/536-tonight-notes-flush-on-unmount into main 2026-10-02 20:56:01 +00:00

Conversation 1 Commits 2 Files changed 3 +97 -2

claude-bot commented 2026-10-02 04:51:56 +00:00

Collaborator

Copy link

Closes #536.

What was wrong

A card's notes for tonight save 1.5 s after the last keystroke, or at once on blur. The field's unmount clean-up cancelled that timer and dropped what it was about to save. So when the field left the page without a blur inside that 1.5 s, the last thing typed was lost: the opened card closing, or the stage moving to another card (the field is keyed by its card). Neither is a page unload, so the unsaved-changes prompt never asked.

What changed

pages/table/TonightNotes.jsx: the clean-up sends a waiting save, to the card it was typed on, and sends nothing when nothing is waiting or the text is back as it was.

Checks

Unit: new TonightNotes.test.jsx, five tests (unmount inside the debounce; a change of card saves to the first card; nothing waiting; typed and put back; the debounce still saves by itself). Full vitest: 3715 passed, 1 skipped. ESLint: 0 errors.

In a browser (local e2e stack built from this branch, 1440×900, as the GM on the table), watching the PATCH that carries tonight_notes:

What was done	Result
Typed, then closed the card 300 ms later with no blur	one PATCH at 315 ms; kept after reload
Typed on the stage, then the stage moved to card 2 with no blur (spotlight, and "Next card")	one PATCH to card 1 at about 310 ms; card 2 untouched; kept after reload
Typed, then left the table by "Leave" 300 ms later	one PATCH at 342 ms; kept after reload
Typed and waited	one PATCH at 1478 ms; closing afterwards sent nothing more
Typed, then clicked elsewhere	one PATCH at 114 ms
Opened and closed without typing; typed "x" then Backspace and closed	no PATCH

No console errors or warnings. In practice a real click on "Show less" or another card blurs the field first, which already saved; the lost case needed the field to go without a blur (a keyboard activation, or the stage being moved from elsewhere). The no-blur rows above were driven by script for that reason.

Not checked at phone width: the change has no layout, and the phone table has no notes field of its own.

🤖 Generated with Claude Code

Closes #536. ## What was wrong A card's notes for tonight save 1.5 s after the last keystroke, or at once on blur. The field's unmount clean-up cancelled that timer and dropped what it was about to save. So when the field left the page without a blur inside that 1.5 s, the last thing typed was lost: the opened card closing, or the stage moving to another card (the field is keyed by its card). Neither is a page unload, so the unsaved-changes prompt never asked. ## What changed `pages/table/TonightNotes.jsx`: the clean-up sends a waiting save, to the card it was typed on, and sends nothing when nothing is waiting or the text is back as it was. ## Checks Unit: new `TonightNotes.test.jsx`, five tests (unmount inside the debounce; a change of card saves to the first card; nothing waiting; typed and put back; the debounce still saves by itself). Full vitest: 3715 passed, 1 skipped. ESLint: 0 errors. In a browser (local e2e stack built from this branch, 1440×900, as the GM on the table), watching the PATCH that carries `tonight_notes`: | What was done | Result | |---|---| | Typed, then closed the card 300 ms later with no blur | one PATCH at 315 ms; kept after reload | | Typed on the stage, then the stage moved to card 2 with no blur (spotlight, and "Next card") | one PATCH to card 1 at about 310 ms; card 2 untouched; kept after reload | | Typed, then left the table by "Leave" 300 ms later | one PATCH at 342 ms; kept after reload | | Typed and waited | one PATCH at 1478 ms; closing afterwards sent nothing more | | Typed, then clicked elsewhere | one PATCH at 114 ms | | Opened and closed without typing; typed "x" then Backspace and closed | no PATCH | No console errors or warnings. In practice a real click on "Show less" or another card blurs the field first, which already saved; the lost case needed the field to go without a blur (a keyboard activation, or the stage being moved from elsewhere). The no-blur rows above were driven by script for that reason. Not checked at phone width: the change has no layout, and the phone table has no notes field of its own. 🤖 Generated with [Claude Code](https://claude.com/claude-code)

claude-bot added 2 commits 2026-10-02 04:51:58 +00:00

A card's notes for tonight are sent when the field goes away mid-save (#536) ... 6da9a47a2f

The unmount clean-up cancelled the debounce timer and dropped what it was
about to save. Closing the opened card or the stage moving to another card
(the field is keyed by its card) inside the 1.5 s lost the last thing typed;
neither is a page unload, so beforeunload never asked. The clean-up now sends
it, to the card it was typed on.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

Merge remote-tracking branch 'origin/main' into fix/536-tonight-notes-flush-on-unmount 0b7ef56eb9

claude-bot commented 2026-10-02 05:07:33 +00:00

Author

Collaborator

Copy link

CI on 0b7ef56e, run https://git.rhoving.com/rbrooks/Quest-Board/actions/runs/1799 : the three CI / End-to-end journeys jobs (desktop, keyboard, phone) failed before any test ran. The runner ran out of disk while building the journey stack, at 04:57–04:59 UTC:

target fake-ai: failed to solve: failed to extract layer sha256:24f010d1…: write /var/lib/docker/containerd/…: no space left on device
[runner]: copyContent: failed to copy content to container: Error response from daemon: write /run/act/workflow/5: no space left on device

Nothing in this PR causes it. Not re-pushed: once the runner has space, those three jobs want a re-run.

CI on 0b7ef56e, run https://git.rhoving.com/rbrooks/Quest-Board/actions/runs/1799 : the three **CI / End-to-end journeys** jobs (desktop, keyboard, phone) failed before any test ran. The runner ran out of disk while building the journey stack, at 04:57–04:59 UTC: ``` target fake-ai: failed to solve: failed to extract layer sha256:24f010d1…: write /var/lib/docker/containerd/…: no space left on device [runner]: copyContent: failed to copy content to container: Error response from daemon: write /run/act/workflow/5: no space left on device ``` Nothing in this PR causes it. Not re-pushed: once the runner has space, those three jobs want a re-run.

claude-bot merged commit 846e23e934 into main 2026-10-02 20:56:01 +00:00

claude-bot referenced this pull request from a commit 2026-10-02 20:56:02 +00:00

Merge PR #1035: a card's notes for tonight are sent when the field goes away mid-save (#536)

claude-bot deleted branch fix/536-tonight-notes-flush-on-unmount 2026-10-02 20:56:03 +00:00

claude-bot referenced this pull request 2026-10-02 20:56:07 +00:00

[Frontend] Shelf tonight-notes: the auto-save debounce is cancelled on unmount without flushing #536

Sign in to join this conversation.

Reviewers

No reviewers

Labels

Clear labels   

backend

Issue related to the backend

  

bot

Issue related to the bot

  

bug

Something is not working

  

duplicate

This issue or pull request already exists

  

enhancement

New feature

  

frontend

Issue related to the frontend

  

help wanted

Need some help

  

invalid

Something is wrong

  

question

More information is needed

  

recording

Issue related to the recording system

  

security

Security hardening / vulnerability work

  

wontfix

This won't be fixed

No labels

backend

bot

bug

duplicate

enhancement

frontend

help wanted

invalid

question

recording

security

wontfix

Milestone

Clear milestone

No items

No milestone

Projects

Clear projects

No items

No project

Assignees

Clear assignees

No assignees

1 participant

Notifications

Subscribe

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference

rbrooks/Quest-Board!1035

No description provided.