test(doc-control): address RENDERING.md table cells by name, and check the ladder column - #747
Open
tractorjuice wants to merge 2 commits into
Open
test(doc-control): address RENDERING.md table cells by name, and check the ladder column#747tractorjuice wants to merge 2 commits into
tractorjuice wants to merge 2 commits into
Conversation
…k the ladder column Two gaps flagged in #744's final review. The table assertions read cells positionally (c[1], c[3]), so reordering a column in RENDERING.md would silently change what was asserted — the table still parses and the guard still passes while checking the wrong thing. Cells are now addressed by column name, and a missing column fails loudly with the header it actually found rather than comparing against undefined. The `Classification ladder` column was never checked at all, so a row could name the correct partial while describing a completely different scheme. It is now compared against the partial's own Classification row. The comparison is first-rung-to-last rather than verbatim, because the table legitimately abbreviates — CA writes "Protected A-C" where the partial spells out all three. Both probed: swapping two column headers, and giving CA the UK ladder text while leaving its partial alone, each now fail with a specific message. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VKbHCujejpxkgvh47BndDE
lint-markdown.yml is where the node guards run — regime registration, doc-type dual registration, handoff validation. Its path filters listed the inputs those guards check (doc-types.mjs, plugin.json, hooks, tests/plugin) but never scripts/tests/** itself. So a PR that changes only a guard's own test file does not run that guard. This PR was exactly that shape: it rewrote test-regime-registration.mjs and CI executed only the Python suite, which never invokes it. Adds scripts/tests/** to both the push and pull_request filters. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VKbHCujejpxkgvh47BndDE
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.
Two Minor findings from #744's final review, now that the guard they concern is on
main.Positional cell reads
scripts/tests/test-regime-registration.mjsaddressed table cells by index —c[1]for the partial,c[3]for the routing. Reorder a column inRENDERING.mdand the table still parses, the guard still passes, and it is silently asserting the wrong thing.Cells are now addressed by column name. A missing column fails loudly, reporting the header it actually found rather than comparing against
undefined.The ladder column was never checked
The
Classification laddercolumn existed in the table but nothing verified it. A row could name the correct partial while describing a completely different scheme — precisely the drift the guard exists to prevent, one column over from where it was looking.It is now compared against the partial's own Classification row. The comparison is first rung to last rung, not verbatim, because the table legitimately abbreviates: CA writes
Protected A–Cwheredocument-control-ca.mdspells out all three. Demanding exact equality would force the table to become unreadable, which is the wrong trade for a document a human reads.Probed, not assumed
Both assertions were verified by making the change they are meant to catch:
Swapping two column headers (leaving the data rows alone):
Giving CA the UK ladder text in the table while leaving its partial untouched:
Neither was detectable before.
Verification
Test-only change —
RENDERING.mdand the partials are untouched, so nothing needed propagating.🤖 Generated with Claude Code
https://claude.ai/code/session_01VKbHCujejpxkgvh47BndDE