Lane 11: a sheet's rank takes a name as well as an ordinal, as the evaluator reads it (#737) #966

Merged

claude-bot merged 3 commits from fix/lane11-rank-names into main 2026-09-30 10:43:57 +00:00

Conversation 1 Commits 3 Files changed 6 +850 -24

claude-bot commented 2026-09-30 08:48:23 +00:00

Collaborator

Copy link

A follow-up to #964 (lane 11 PR 3), found by PR 6.

The server's write validation refused rank names that the evaluator accepts. For example, "skills.athletics.prof": "proficient" on a 5e sheet got a 422 "is a rank from 0 to 2", but PR 2's rank() reads either an ordinal or a name on the field's scale, and the built-in 5e fixtures store names.

Fix (sheet_values.check, the rank branch):

• Accepted: a name that exactly matches the field's scale (the evaluator's rule) is accepted and stored as given; so is a whole ordinal from 0 to the top of the scale.
• Refused: anything else gets the same 422 shape, naming both forms: "is a rank: 0–2 or one of none, proficient, expertise".

Tests (test_sheets_server.py): run for a PF2e-style scale and a 5e-style scale.

• A name is accepted, stored and evaluated (expert → 7, proficient → 5), and so is an ordinal.
• A name in the wrong case, an unknown name, an out-of-range ordinal, −1 and 1.5 are each refused with the exact body.
• tests/sheets passes (478), and ruff (including alembic/) is clean.

No user-visible change yet (the sheet page arrives with PR 9), so there's no changelog fragment.

🤖 Generated with Claude Code

A follow-up to #964 (lane 11 PR 3), found by PR 6. The server's write validation refused rank **names** that the evaluator accepts. For example, `"skills.athletics.prof": "proficient"` on a 5e sheet got a 422 "is a rank from 0 to 2", but PR 2's `rank()` reads either an ordinal or a name on the field's scale, and the built-in 5e fixtures store names. **Fix** (`sheet_values.check`, the rank branch): - **Accepted:** a name that exactly matches the field's scale (the evaluator's rule) is accepted and stored as given; so is a whole ordinal from 0 to the top of the scale. - **Refused:** anything else gets the same 422 shape, naming both forms: "is a rank: 0–2 or one of none, proficient, expertise". **Tests** (`test_sheets_server.py`): run for a PF2e-style scale and a 5e-style scale. - A name is accepted, stored and evaluated (`expert` → 7, `proficient` → 5), and so is an ordinal. - A name in the wrong case, an unknown name, an out-of-range ordinal, −1 and 1.5 are each refused with the exact body. - `tests/sheets` passes (478), and ruff (including `alembic/`) is clean. No user-visible change yet (the sheet page arrives with PR 9), so there's no changelog fragment. 🤖 Generated with [Claude Code](https://claude.com/claude-code)

claude-bot added 1 commit 2026-09-30 08:48:24 +00:00

Lane 11 PR 3: a rank is an ordinal or a name on its scale, as the evaluator reads it (#737) ... 0efb756e99

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

claude-bot commented 2026-09-30 09:17:38 +00:00

Author

Collaborator

Copy link

This PR now carries more than the rank-name fix. The sheet page's render check (PR 9) wrote the built-in fixtures through these routes and found two more places where the server refused data the evaluator accepts:

• <resource>.max as data, for a resource whose maximum is typed by hand (a creature's hp.max). This is PR 7's rule, taken exactly (legacy.spec_for), so the two branches merge cleanly. .current and .temp stay out of data.
• Entry-row overrides accept any JSON value, objects included (PF2e runes), capped at 20,000 bytes each. EntryAdd also takes overrides.

The new test: tests/sheets/test_fixtures_through_the_server.py writes every built-in fixture (all 21: PF2e, 5e and generic) through the routes: POST, then PATCH overrides, then resources, with gone rows made genuinely gone. It asserts the server's evaluated values equal each fixture's expectations. It would have caught all three problems, and it keeps the server's validation and the fixtures from drifting apart.

Checks: tests/sheets plus the lore-restore and wiki-page tests pass (578), and ruff is clean, including alembic/.

This PR now carries more than the rank-name fix. The sheet page's render check (PR 9) wrote the built-in fixtures through these routes and found two more places where the server refused data the evaluator accepts: - **`<resource>.max` as data**, for a resource whose maximum is typed by hand (a creature's `hp.max`). This is PR 7's rule, taken exactly (`legacy.spec_for`), so the two branches merge cleanly. `.current` and `.temp` stay out of `data`. - **Entry-row overrides accept any JSON value**, objects included (PF2e runes), capped at 20,000 bytes each. `EntryAdd` also takes `overrides`. **The new test:** `tests/sheets/test_fixtures_through_the_server.py` writes **every built-in fixture (all 21: PF2e, 5e and generic)** through the routes: POST, then PATCH overrides, then resources, with gone rows made genuinely gone. It asserts the server's evaluated values equal each fixture's expectations. It would have caught all three problems, and it keeps the server's validation and the fixtures from drifting apart. **Checks:** `tests/sheets` plus the lore-restore and wiki-page tests pass (578), and ruff is clean, including `alembic/`.

rbrooks added 1 commit 2026-09-30 09:17:39 +00:00

Lane 11: the server keeps what the built-in fixtures hold — a hand-typed resource max, object overrides on entry rows; every fixture through the routes (#737) ... 570080217c

<resource>.max is accepted exactly as PR 7 does it (legacy.spec_for, taken
from build/lane11-7-convert as-is so the branches merge cleanly).

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

rbrooks added 1 commit 2026-09-30 09:46:01 +00:00

Merge remote-tracking branch 'origin/main' into fix/lane11-rank-names 828caf72b7

claude-bot merged commit 5ef457214d into main 2026-09-30 10:43:57 +00:00

claude-bot referenced this pull request from a commit 2026-09-30 10:44:00 +00:00

Merge PR #966: lane 11 — the server keeps rank names, a hand-typed max and object overrides; every fixture through the routes (#737)

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!966

No description provided.