https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=41719
--- Comment #5 from Chris Nighswonger <[email protected]> --- (In reply to Owen Leonard from comment #2) Owen: the failure you hit was PDF::Reuse 0.39 — its prEnd() crashes on the TTFont lifecycle (fixed in 0.43, Bug 41717), and the old test swallowed that crash, leaving an empty file and a confusing downstream failure. The revised patch fixes the experience directly: both test files now skip_all with a clear message when PDF::Reuse < 0.43 is installed, so you get "PDF::Reuse >= 0.43 required (see Bug 41717)" instead of a mystery. Revision 2 notes (replaces obsoleted attachment 192076): - The end-to-end and barcode assertions in revision 1 could pass even when the Koha rendering code did nothing (file-size / %PDF-header checks are produced by PDF::Reuse unconditionally; Label.pm converts barcode-generation failures to warns). Revision 2 asserts on Koha-originated output: per-type absence of the barcode-generation warning, text blocks at layout-computed coordinates, and the guide-box rectangle at its exact geometry. - Extending patroncard barcode coverage to all five dispatched types exposed a real crash: CODE39MOD / CODE39MOD10 die with "Undefined subroutine CheckDigits" because C4::Patroncards::Patroncard never imports Algorithm::CheckDigits (C4::Labels::Label does). Filed as Bug 43095 (one-line fix, patch attached there); this bug now depends on it. The two test rows fail by design until 43095 is applied — see the test plan. - Warning filters are scoped to the exact known Bug 41718 / PDF::Reuse warning text and forward anything unexpected to Test::NoWarnings, instead of the blanket suppression in revision 1. Test plan is in the commit message. Verified in koha-testing-docker: Labels 11/11; Patroncards all pass with 43095 applied, and fails only the two designed rows without it. AI Disclosure: revision 2 was developed with AI assistance (Claude Fable 5 reviewed revision 1 and drafted the revision; OpenAI Codex reviewed the result). Human author directed the approach, chose to surface rather than skip the Bug 43095 failure, reviewed all changes, and verified test behavior. Full disclosure in the commit message per the draft AI guideline. -- You are receiving this mail because: You are watching all bug changes. _______________________________________________ Koha-bugs mailing list -- [email protected] To unsubscribe send an email to [email protected] website : http://www.koha-community.org/ git : http://git.koha-community.org/ bugs : http://bugs.koha-community.org/
