ci-gates 208: a tightened pattern is not a second opinion - #2808
Merged
Conversation
Refs #2807 A scanner hunting for the adamw.t27 defect reported 91 files with repeated top-level definitions. Tightening it to module indentation gave 43 and a barely-moved count, which read as corroboration. Both runs shared the bug: specs/numeric/gf16.t27 has 110 Zig test blocks each opening `const a` at four spaces, and the scanner was counting local variables. Its real count is zero. Settled by duplicated section banners -- a signal that does not pass through the name scanner. Exactly one file in specs/ fires both, and it is the one already read by hand. The corpus-wide number is withdrawn. Third scanner artifact this session; none was caught by re-reading the scanner.
Contributor
PR DashboardGenerated at: 2026-08-29 11:48:36 UTC
Summary
Seal Status
|
Contributor
|
📓 NotebookLM Notebook linked to this PR
This notebook contains session context, decisions, and artifacts for this work. |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Refs #2807
A scanner hunting for the
adamw.t27defect elsewhere reported 91 files with repeated top-level definitions, worstspecs/numeric/gf16.t27at 561 extras. Tightening the pattern to module indentation gave 43 files and 539 extras — close enough to feel confirmed.Both runs were the same ruler.
gf16.t27has 110 Zig test blocks, each opening:const aat four spaces, inside a test body. The scanner had found local variables. Its real count is zero, and the corpus-wide number is withdrawn.What settled it was a signal that does not pass through the name scanner at all: duplicated section banners.
adamw.t27carries// 1. Constants/// 2. Types/// 3. Core Functionstwice (11/25/54 and 414/470/509);gf16.t27has none. Exactly one file inspecs/fires both signals, and it is the one already read by hand.The lesson, recorded as ci-gates 208: tightening a pattern is not a second opinion. Both runs shared the assumption that a
constat module indentation is a module-scope declaration, and that assumption was the bug, so the stricter run inherited it intact.Third scanner artifact this session — the
for allcensus matching prose, the type reader dropping everypubfield, and now this. All three produced a confident number; none was caught by re-reading the scanner.