* On 2026-05-15 at 10:23 BST, Jonathan Perkin wrote:
New Failures Since 20260513T220202Z Breaks Previously ---------------------------------------------------------------------------- R-curl-6.2.0nb6 35 absent mupdf-1.27.2nb1 10 absent gnupg-1.4.23nb30 7 absent ...
I'm tightening up the match for these so that PKGREVISION bumps and similar are no longer marked as new failures. I was incorrectly matching PKGNAME rather than PKGBASE.
ruby34-bcrypt-3.1.22 16 indirect-failed ruby33-bcrypt-3.1.22 16 indirect-failed ...
These are caused by rust breaking and then being fixed. I'm still not sure on the best way to handle these. I wonder if they should just be excluded - yes it's useful when something is genuinely fixed and it exposes new failures that were blocked, but I think this is very much the exception rather than the rule.
-- Jonathan Perkin pkgsrc.smartos.org Open Source Complete Cloud www.tritondatacenter.com