rbrooks/Quest-Board
Watch
1
0
Fork
You've already forked Quest-Board
 0
Code Issues 58 Pull requests 7 Projects Releases 76 Packages Wiki Activity Actions

feat(feedback-site): lane 3 backfill — every session state drawn (#659, #696) #700

Closed
claude-bot wants to merge 0 commits from feat/lane3-backfill into main
pull from: feat/lane3-backfill
merge into: rbrooks:main
Conversation 1 Commits - Files changed -
claude-bot commented 2026-09-19 00:10:55 +00:00
Collaborator
Copy link

Step 2 of #696's plan, for the session pages. The coverage audit found lane 3 had left out 83 of its 232 placements and only half-drawn 55 more. This draws all of them.

Settled before drawing (the owner chose the recommendation each time)

  • D50: a pencilled-in session asks softly. Players can say whether they could make it. No reminder is sent and nothing goes in a calendar. The answers carry over when the GM confirms. IS-378.
  • D51: every version of the summary is listed, and any version can be restored. The GM's own edits count as versions. Needs a server change. IS-379.
  • D52: Manage. Move it and Add to your calendar stay as buttons. The rarer actions go in a menu, with the destructive ones last and set apart.

Drawn

  • SESS-1
    • The rest of the vote:
      • the GM's Remove button and its confirmation, adding a time, and the five-time cap;
      • a slot-list error, and per-slot saving and failure;
      • no star while no slot is ahead, and the empty vote;
      • the vote as a traveller sees it (D17);
      • a key for the grid, and names written out instead of a tooltip.
    • Pencilled in: the player's and the GM's views.
    • The page's loading, not-found and action-error states.
  • SESS-2
    • Manage, and every action behind it:
      • edit, including in-world dates (IS-092);
      • mark as played;
      • take out of the series;
      • back to voting, and the times to offer;
      • cancel.
    • Moving the session, the RSVP states, and Previously on hidden or failed.
    • The calendar's two options, the series line, and Prep this session.
    • The GM's recording section and notes.
    • The vote's result, which stays on the page after the vote.
    • A session confirmed from pencilled in, and a session being played.
  • SESS-3
    • Nine transcript states, a limited-setup notice, and "No summary yet".
    • What was shown at the table (IS-017) and an itemised ledger.
    • A moment highlight, an "Unclear" event, and the highlights' loading, error and empty states.
    • The GM's view of an event they struck.
  • SESS-4
    • The recording and transcript:
      • approval, with two different confirmations;
      • the feedback tally and retry;
      • who the recording heard;
      • naming a voice, with pronouns;
      • editing the transcript.
    • Versions, and the rewrite's running, stale, failed and done states.
    • The rest of the event corrections, with transcript citations and failed validations.
    • Attendance proposals, errors, toggles and character pins.
    • Highlight add, edit and delete.
    • The checklist's full vocabulary, the shelf sweep link, and D31's states.
  • SESS-5 and SESS-7: the reinstate confirmation, and moving notes (which replaces the target session's table note).

Decided while drawing

Each is recorded as an "Also settled" line in the catalogue.

  • The status chip names the mode, so there is no separate mode badge.
  • Add to your calendar appears only for sessions still to come.
  • Events on the Review tab stay open.
  • A session being played asks nobody whether they're in.
  • The checklist loses its journal row (D47).

Coverage

  • docs/design/redesign-surfaces/coverage/ now holds the 2026-09-18 audit of lanes 1–4, as #696 promised, with the machine paths removed.
  • coverage/lane3.py reruns the check against the current catalogue: 237 placements, 185 drawn, 29 replaced, 5 behaviour, 14 elsewhere, 4 deferred, and none left unclassified.
  • The catalogue is now 1,648 rows on 103 surfaces.

Rendering caught eleven faults

  1. The first draw broke D17: the travelling GM was shown the table's time.
  2. The vote's slot headings sat at different heights.
  3. The remove confirmation gave the wrong number of answers.
  4. A pencilled-in session claimed to be made by the series.
  5. The GM was asked whether they could make their own pencilled-in session.
  6. A session being played still asked "Are you in?".
  7. The new "Unclear" event was out of time order.
  8. The version list contradicted the corrections still waiting.
  9. One event was relabelled in opposite directions in two stills.
  10. The checklist notes ran into their labels.
  11. Lane 3 had no spacing between grids.

The feedback site's tests pass (101). All eight lanes render. The changed sections were checked in both themes.

🤖 Generated with Claude Code

Step 2 of #696's plan, for the session pages. The coverage audit found lane 3 had left out 83 of its 232 placements and only half-drawn 55 more. This draws all of them. ## Settled before drawing (the owner chose the recommendation each time) - **D50: a pencilled-in session asks softly.** Players can say whether they could make it. No reminder is sent and nothing goes in a calendar. The answers carry over when the GM confirms. `IS-378`. - **D51: every version of the summary is listed, and any version can be restored.** The GM's own edits count as versions. Needs a server change. `IS-379`. - **D52: Manage.** Move it and Add to your calendar stay as buttons. The rarer actions go in a menu, with the destructive ones last and set apart. ## Drawn - **SESS-1** - The rest of the vote: - the GM's Remove button and its confirmation, adding a time, and the five-time cap; - a slot-list error, and per-slot saving and failure; - no star while no slot is ahead, and the empty vote; - the vote as a traveller sees it (D17); - a key for the grid, and names written out instead of a tooltip. - Pencilled in: the player's and the GM's views. - The page's loading, not-found and action-error states. - **SESS-2** - Manage, and every action behind it: - edit, including in-world dates (`IS-092`); - mark as played; - take out of the series; - back to voting, and the times to offer; - cancel. - Moving the session, the RSVP states, and Previously on hidden or failed. - The calendar's two options, the series line, and Prep this session. - The GM's recording section and notes. - The vote's result, which stays on the page after the vote. - A session confirmed from pencilled in, and a session being played. - **SESS-3** - Nine transcript states, a limited-setup notice, and "No summary yet". - What was shown at the table (`IS-017`) and an itemised ledger. - A moment highlight, an "Unclear" event, and the highlights' loading, error and empty states. - The GM's view of an event they struck. - **SESS-4** - The recording and transcript: - approval, with two different confirmations; - the feedback tally and retry; - who the recording heard; - naming a voice, with pronouns; - editing the transcript. - Versions, and the rewrite's running, stale, failed and done states. - The rest of the event corrections, with transcript citations and failed validations. - Attendance proposals, errors, toggles and character pins. - Highlight add, edit and delete. - The checklist's full vocabulary, the shelf sweep link, and D31's states. - **SESS-5 and SESS-7:** the reinstate confirmation, and moving notes (which replaces the target session's table note). ## Decided while drawing Each is recorded as an "Also settled" line in the catalogue. - **The status chip names the mode**, so there is no separate mode badge. - **Add to your calendar appears only for sessions still to come.** - **Events on the Review tab stay open.** - **A session being played asks nobody whether they're in.** - **The checklist loses its journal row** (D47). ## Coverage - `docs/design/redesign-surfaces/coverage/` now holds the 2026-09-18 audit of lanes 1–4, as #696 promised, with the machine paths removed. - `coverage/lane3.py` reruns the check against the current catalogue: 237 placements, 185 drawn, 29 replaced, 5 behaviour, 14 elsewhere, 4 deferred, and none left unclassified. - The catalogue is now 1,648 rows on 103 surfaces. ## Rendering caught eleven faults 1. The first draw broke D17: the travelling GM was shown the table's time. 2. The vote's slot headings sat at different heights. 3. The remove confirmation gave the wrong number of answers. 4. A pencilled-in session claimed to be made by the series. 5. The GM was asked whether they could make their own pencilled-in session. 6. A session being played still asked "Are you in?". 7. The new "Unclear" event was out of time order. 8. The version list contradicted the corrections still waiting. 9. One event was relabelled in opposite directions in two stills. 10. The checklist notes ran into their labels. 11. Lane 3 had no spacing between grids. The feedback site's tests pass (101). All eight lanes render. The changed sections were checked in both themes. 🤖 Generated with [Claude Code](https://claude.com/claude-code)
feat(feedback-site): lane 3 backfill — every session state drawn (#659, #696)
All checks were successful
CI / Summarisation accuracy eval harness (stub provider) (pull_request) Successful in 56s
Details
CI / Frontend tests, audit, and build (pull_request) Successful in 2m47s
Details
CI / Bot tests and audit (pull_request) Successful in 2m1s
Details
CI / Bot/backend version sync (pull_request) Successful in 24s
Details
CI / Backend lint (ruff) (pull_request) Successful in 1m12s
Details
CI / Feedback site tests (pull_request) Successful in 1m57s
Details
CI / Docker image build (pull_request) Successful in 5m7s
Details
CI / Backend migration, tests, and audit (pull_request) Successful in 9m33s
Details
CI / Synthetic session harness (no GPU, no LLM) (pull_request) Successful in 14m19s
Details
9b150a937c 
The first draw of the session pages left 83 of 232 placements out and
half-drew 55. This draws them:

- SESS-1: the rest of the vote. The GM's Remove, adding a time and the
  five-time cap, a slate error, per-slot saving and failure, no star while
  nothing is ahead, the empty vote, a traveller's times (D17), a key for
  the grid, and names instead of a tooltip. Also pencilled in (D50), and
  the page's loading, not-found and action-error states.
- SESS-2: Manage (D52) with every action behind it: edit (with in-world
  dates, IS-092), mark as played, take out of the series, back to voting
  and the times to offer, cancel. Also Move it, the RSVP states,
  Previously on hidden or failed, the calendar's two ways in, the series
  line, Prep this session, the recording section, notes, the vote kept
  after it is won, and a session being played.
- SESS-3: nine transcript states, a limited-setup notice, no summary yet,
  what was shown at the table (IS-017), an itemised ledger, a moment
  highlight, an unclear event, the highlights' own states, and the GM's
  view of a struck event.
- SESS-4: the recording and transcript (approval, feedback tally, retry,
  who was heard, naming a voice, editing), every version of the summary
  (D51, IS-379) and the rewrite's states, the rest of the event
  corrections, attendance proposals and character pins, highlight
  add/edit/delete, the checklist's full vocabulary, and D31's states.
- SESS-5/7: the reinstate question and moving notes, which replaces.

It also fixes the first draw's D17 fault: the travelling GM was shown the
table's time with their own beneath it.

Decisions: D50 (pencilled in asks softly), D51 (versions, any can be put
back), D52 (Manage). The coverage check is committed under
docs/design/redesign-surfaces/coverage/: the 2026-09-18 audit of lanes 1
to 4 as promised on #696, and lane3.py, which reports 237 placements
(185 drawn, 29 replaced, 5 behaviour, 14 elsewhere, 4 deferred) with
none unclassified. The catalogue is 1,648 rows on 103 surfaces.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
claude-bot commented 2026-09-19 00:30:05 +00:00
Author
Collaborator
Copy link

CI passed on run 9991. Merged to main as cefca32.

Deployed to the review site. Checks:

  • Before deploying, the feedback volume was backed up to /home/claude/backups/qb-feedback-data-20260919-002857.tgz.
  • The container reports healthy.
  • The pages for lanes 1, 1b, 2, 2c, 3, 4 and 5 all return 200.
  • The site now serves the new lane 3 page.
  • All 6 stored responses are intact.
CI passed on run 9991. Merged to main as cefca32. Deployed to the review site. Checks: - Before deploying, the feedback volume was backed up to `/home/claude/backups/qb-feedback-data-20260919-002857.tgz`. - The container reports healthy. - The pages for lanes 1, 1b, 2, 2c, 3, 4 and 5 all return 200. - The site now serves the new lane 3 page. - All 6 stored responses are intact.
claude-bot closed this pull request 2026-09-19 00:30:07 +00:00
All checks were successful
CI / Summarisation accuracy eval harness (stub provider) (pull_request) Successful in 56s
Required
Details
CI / Frontend tests, audit, and build (pull_request) Successful in 2m47s
Required
Details
CI / Bot tests and audit (pull_request) Successful in 2m1s
Required
Details
CI / Bot/backend version sync (pull_request) Successful in 24s
Required
Details
CI / Backend lint (ruff) (pull_request) Successful in 1m12s
Required
Details
CI / Feedback site tests (pull_request) Successful in 1m57s
Required
Details
CI / Docker image build (pull_request) Successful in 5m7s
Required
Details
CI / Backend migration, tests, and audit (pull_request) Successful in 9m33s
Required
Details
CI / Synthetic session harness (no GPU, no LLM) (pull_request) Successful in 14m19s
Required
Details

Pull request closed

This pull request cannot be reopened because the branch was deleted.
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
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!700
No description provided.