Bug#1143976: python3-reportlab: regression in reproducibility? should the tests even be shipped?
Paul Gevers <[email protected]>
| Newsgroups | gmane.linux.debian.devel.bugs.rc |
|---|---|
| Message-ID | <[email protected]> |
Package: python3-reportlab Version: 5.0.0-3 Severity: serious X-Debbugs-CC: Jochen Sprickerhof <[email protected]> Hi, You version of src:python-reportlab in unstable fails to migrate to testing because the QA infrastructure couldn't reproduce the python3-reportlab binary version in unstable while it could reproduce the version in testing. We no longer automatically allow regressions [1, 2]. I quickly looked at the diff that can be found via the excuses page [3] and from a cursory look this appears like a regression in the package (not some of the unsolved flaky toolchain we still have). Thinking about the files which show up differently, I was wondering if the package ships these files under *tests* on purpose; not knowing the package at all it appears to me that stop shipping those files could be an option to regain reproducibility. If you need help debugging this, the reproducibility project has lots of experience, so I suggest reaching out to them. Paul [1] https://lists.debian.org/debian-devel-announce/2026/05/msg00001.html [2] https://lists.debian.org/debian-devel/2026/05/msg00383.html [3] https://qa.debian.org/excuses.php?package=python-reportlab
OpenPGP_signature.asc
(application/pgp-signature, 585 B)
-----BEGIN PGP SIGNATURE----- wsC7BAABCABvBYJqeCeBCRCcXJnrBb11CkcUAAAAAAAeACBzYWx0QG5vdGF0aW9u cy5zZXF1b2lhLXBncC5vcmfShzn4SAImyUMxeZ6p6PqHZj4/MapT/KmSOQKT+KmP lhYhBFi2bUhza+k7BS3mcpxcmesFvXUKAADDNwgAtny+U9S/XKuaBdlkbkB37T0A BL5RTYUthVJihnRGPZ34pqypI7LfabfKy4J/cKvB/yToWUCSeAaZsII4xhTbzf1C 5H+C9aX5WbFX4vZbdgeD+lOU0rAzW9L0mOQXcmzD+Mku+AsKM6SW/O0M6Ta2Oak7 VsZfgKtK1knVxapjr36a3SKbc8yIx0tS4VSknDlcxbYyST/WgEh3zkY0/4xbYHBr +ogo0bYukfFtxtq2fN+vxOy8LqxhgAthxDYzL63+z1wvGVg/xDzz6E1mN0IeCkg2 4bwhzxDp0xq7ChmwwdnBjG8JpLYl3tLE2qEtOmXJFMMbjLFcR97wualNBrUJXA== =WnuI -----END PGP SIGNATURE-----