Lane 8 journeys: J5 through the new Suggested wiki updates, J6 reads the wiki as a player (#734) #928

Merged

claude-bot merged 5 commits from build/lane8-journeys into main 2026-09-29 10:09:44 +00:00

Conversation 0 Commits 5 Files changed 9 +439 -68

claude-bot commented 2026-09-29 08:43:56 +00:00

Collaborator

Copy link

Lane 8's journeys PR (#734; survey §5). The journeys now walk lane 8's pages, and doing so found two bugs, fixed here.

J5, "after the session"

J5 used to stop at the old suggestions page because it failed the contrast scan. Now it:

• follows the checklist's row to Suggested wiki updates and scans it (IS-002's list of where each session stands);
• Adds the new page the session found;
• opens the update's merge draft (WIKI-9), waits for the worker, reads the merged text, approves it, and reads it on the page.

The fake AI extracts a second entry, an update to Sister Quill (a page the seed has), matched to her page by title, and answers the merge draft with the page followed by what the session added.

Three projects, one database. All three evenings are the same recording, so after the first project adds and merges, a later evening's run correctly finds nothing new: the server skips what a page already says, and the row reads "None left to review". That project asserts this and scans Suggested wiki updates on its own layout, and its add and merge steps skip with the reason given.

J6, "a player finds out what happened", act 5

As Priya, on lane 8's pages, with axe on each:

• Search: Sister Quill is found; a GM-only page that mentions her never is.
• Her page, rendered for her: the table's text and her :::for Ossian Vale block are there, the GM's block is not; only public timeline events; Mentioned in lists the linking page.
• People, from the Contents inside an article (#734's first comment).
• The quest log and a quest's page: the objective the table knows, not the GM's.
• The party: what Ossian carries.
• Handouts: the one shown to her alone ("Just you"), not the one the GM hasn't shown.

The seed

• Ossian Vale, Priya's character, active. identities.json always said she plays him; the seed never made him.
• Sister Quill's page gains a GM block and a block for Ossian.
• canon.wiki_scenery: a linking page, a GM-only page, a quest with a hidden objective, a line Ossian carries, and two handouts.
• With a character, Priya's recorded lines are Ossian Vale's (the speaker legend's rule), so J6 counts speakers that way.

Bugs the journeys found (fixed)

• A player's Handouts page threw when a handout was shown with no file (Tile read attachments[0].kind), so none of her handouts appeared. Fixed, with a unit test.
• The party page failed axe (link-in-text-block: character links 2.23:1 against the text around them, no underline). They're underlined now.

Checked

Full journey suite locally, all three projects: 118 passed, 11 skipped (the designed skips, above). Frontend wiki tests pass (172), lint clean.

Part of #734.

🤖 Generated with Claude Code

Lane 8's journeys PR (#734; survey §5). The journeys now walk lane 8's pages, and doing so found two bugs, fixed here. ## J5, "after the session" J5 used to stop at the old suggestions page because it failed the contrast scan. Now it: - follows the checklist's row to **Suggested wiki updates** and scans it (IS-002's list of where each session stands); - **Adds** the new page the session found; - opens the **update**'s merge draft (WIKI-9), waits for the worker, reads the merged text, **approves** it, and reads it on the page. The fake AI extracts a second entry, an update to Sister Quill (a page the seed has), matched to her page by title, and answers the merge draft with the page followed by what the session added. **Three projects, one database.** All three evenings are the same recording, so after the first project adds and merges, a later evening's run correctly finds nothing new: the server skips what a page already says, and the row reads "None left to review". That project asserts this and scans Suggested wiki updates on its own layout, and its add and merge steps skip with the reason given. ## J6, "a player finds out what happened", act 5 As Priya, on lane 8's pages, with axe on each: - **Search:** Sister Quill is found; a GM-only page that mentions her never is. - **Her page, rendered for her:** the table's text and her `:::for Ossian Vale` block are there, **the GM's block is not**; only public timeline events; Mentioned in lists the linking page. - **People**, from the Contents inside an article (#734's first comment). - **The quest log and a quest's page:** the objective the table knows, not the GM's. - **The party:** what Ossian carries. - **Handouts:** the one shown to her alone ("Just you"), not the one the GM hasn't shown. ## The seed - **Ossian Vale**, Priya's character, active. `identities.json` always said she plays him; the seed never made him. - Sister Quill's page gains a GM block and a block for Ossian. - `canon.wiki_scenery`: a linking page, a GM-only page, a quest with a hidden objective, a line Ossian carries, and two handouts. - With a character, Priya's recorded lines are Ossian Vale's (the speaker legend's rule), so J6 counts speakers that way. ## Bugs the journeys found (fixed) - **A player's Handouts page threw** when a handout was shown with no file (`Tile` read `attachments[0].kind`), so none of her handouts appeared. Fixed, with a unit test. - **The party page failed axe** (`link-in-text-block`: character links 2.23:1 against the text around them, no underline). They're underlined now. ## Checked Full journey suite locally, all three projects: **118 passed, 11 skipped** (the designed skips, above). Frontend wiki tests pass (172), lint clean. Part of #734. 🤖 Generated with [Claude Code](https://claude.com/claude-code)

claude-bot added 3 commits 2026-09-29 08:43:57 +00:00

test(e2e): J5 goes through the door: Suggested wiki updates, a new page, a merge (#734) ... 9945ad5882

J5 used to stop at the suggestions page because the old one failed the
contrast scan. Lane 8's page is scanned now, and the journey:
- follows the checklist's row to Suggested wiki updates (IS-002's list),
- adds the new page the session found,
- opens the update's merge draft (WIKI-9), waits for the worker, reads
  the merged text, approves it, and reads it on the page.

The fake AI finds a second entry, an update to Sister Quill (a page the
seed has), matched to her page by title, and answers the merge draft
with the page followed by what the session added.

The three projects share one database and the same recording: after the
first adds and merges, a later evening's run finds nothing new (the
server skips what a page already says), which its row says ("None left
to review"). That evening scans the page on its own layout, and its add
and merge steps skip with the reason.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

test(e2e): J6 reads lane 8's wiki as a player; the seed gives it scenery (#734) ... aa7a8919c2

J6's act 5 walks lane 8's pages as Priya: search (a GM-only page that
mentions Sister Quill is never among the results), Sister Quill's page
rendered for her (the table's text and her :::for Ossian block, never the
GM's block; the page's public events only; Mentioned in), People from the
Contents inside an article, the quest log and a quest's page (the
objective the table knows, not the GM's), the party (what Ossian carries)
and Handouts (the one shown to her alone, "Just you"; not the one the GM
hasn't shown). axe on each.

The seed makes Ossian Vale, Priya's character (identities.json always said
she plays him), writes the blocks into Sister Quill's page, and adds the
scenery in canon.wiki_scenery. With a character, Priya's recorded lines
are Ossian Vale's (the speaker legend's rule), so J6 counts speakers that
way.

Two bugs J6 found, fixed:
- A player's Handouts page threw when a handout was shown with no file
  (Tile read attachments[0].kind), so none of her handouts showed.
- The party page's character links failed axe's link-in-text-block
  (2.23:1, no underline).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

Merge remote-tracking branch 'origin/main' into build/lane8-journeys f14f5e04fc

rbrooks added 1 commit 2026-09-29 08:55:29 +00:00

style(e2e): ruff format the fake AI and the seed (#734) ... 1cd9e132b1

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

rbrooks added 1 commit 2026-09-29 09:29:24 +00:00

ci: re-run (the runner's Docker daemon refused to start the migrate container) ... 5135a705cf

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

claude-bot merged commit af7fe0147e into main 2026-09-29 10:09:44 +00:00

claude-bot referenced this pull request from a commit 2026-09-29 10:09:46 +00:00

Merge PR #928: lane 8 journeys (#734)

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

2 participants

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!928

No description provided.