From cd64cbae8237719a7475c8b8088c46294c0aafbd Mon Sep 17 00:00:00 2001 From: argus Date: Thu, 19 Nov 2026 19:52:29 +0100 Subject: [PATCH] review-classifier: UI state wiring with no computation needs no test of its own An ARIA attribute or pressed flag, a focus call after navigation and a badge shown by an existing boolean are glue like the other UI wiring already listed. Logic that decides from data keeps its test. --- .hyperhive/review-classifier.yml | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/.hyperhive/review-classifier.yml b/.hyperhive/review-classifier.yml index 3a56f969c..786f1431b 100644 --- a/.hyperhive/review-classifier.yml +++ b/.hyperhive/review-classifier.yml @@ -52,8 +52,12 @@ classifications: calling a function that is already tested. So does presentational layout math: a small helper in UI code that only maps a value to a position, size or colour that CSS then draws, such as the place of a - slider's zone edge. A diff whose changed files - are all test code (a fixture under `nix/module-eval/`, + slider's zone edge. So does UI state wiring with no computation + behind it: an ARIA attribute or a pressed flag on an element, a focus + call after navigation, a badge shown or hidden by a boolean the page + already has. Logic that decides something from data, such as a + threshold, a filter or a parse, still needs its test. A diff whose + changed files are all test code (a fixture under `nix/module-eval/`, `nix/checks.nix`, a crate's `tests/` directory, or Rust inside `#[cfg(test)]` or a `#[test]` function) changes no behaviour: the question is not applicable. So is a diff that changes only