• testing of piuparts.d.o with adequate from testing/unstable

    From Serafeim (Serafi) Zanikolas@21:1/5 to All on Sat Jan 11 00:40:01 2025
    --f1e772351b3b83ef55fb469e35ab45f32cf6ef569b242d30ad58371d7213 Content-Transfer-Encoding: quoted-printable
    Content-Type: text/plain; charset=UTF-8

    hello piuparts maintainers,

    as you might know, adequate(1) has been rewritten from scratch last July. the rewrite is meant to be a [drop-in] replacement, with only very trivial backwards-incompatible changes (see /usr/share/doc/adequate/NEWS.Debian.gz for full details)

    I'm told that piuparts.d.o runs on Debian stable, and therefore still relies on the old implementation of adequate. Holger has suggested that it'd be good to try out the new implementation in piuparts.d.o (perhaps in a test instance, if one exists) prior to the Trixie release.

    I'm not particularly worried (adequate has not received any bug reports in the last few months, and all of the reports that were filed shortly after the rewrite have been long addressed) but I see no reason why we shouldn't give Holger's suggestion a try. (I'd go even further and suggest that piuparts.d.o should always use adequate from testing but I'd rather leave that for another thread, if you see that as a contentious change). as a one off, we could temporarily pin adequate to the testing release, or manually install the adequate executable from testing/unstable (adequate's Depends in testing/unstable are satisfied by bookworm)

    I'm happy to carry out the aforementioned testing myself, but I'd need somebody to answer questions regarding the piuparts.d.o setup (given that, trivial changes aside, piuparts/docs/README_pejacevic.txt has been written 5y ago) and would need access to at least one piuparts runner instance.

    once we establish that all's good, I'd also like to discuss the possibility of overriding --fail-if-inadequate to true in piuparts.d.o, or even changing the default in piuparts.py

    thanks,
    serafi

    [drop-in] the rewritten adequate is not available in three ports; my
    understanding is that that should be immaterial given that all
    piuparts runners run on amd64

    --f1e772351b3b83ef55fb469e35ab45f32cf6ef569b242d30ad58371d7213
    Content-Type: application/pgp-signature; name="signature.asc"

    -----BEGIN PGP SIGNATURE-----

    iQIzBAABCAAdFiEEA2RWqo7IwLCLSFYbT59tVQ7WEioFAmeBr4IACgkQT59tVQ7W EipJNhAA3AlM/4zECLFmLO5EGd+MQRoMllvJffMi7pD9ZW5XSH8bGPvj/d98Eyav iLoeyzubFwpbRJ4uTctGs4CDSK8PjS2W/Ui9ZAnZvRPVN3mSbyqhW0To7NNBU82V ifPrpX+Y5Y1omUWL2+yeztdHS8MAAChYdsISOtkHzEDGluEFXq18ofcnnOw/nY3z M5cm7Qa++CcHnybtbCp8TfYlt8X5TjghZ0qjxVFoUEHtiAhJr2/K6NI+vuGr5q/P rdCjA/M8MJGFR9vY7xerswjYFBRewWlgafyhj3nFsIj0yvBd66NO4mlw26Ry5879 3934acZ3IwHEYV+/UGnrpuOayJ0ehYW7p4paRuMYBlZ81lCKuVEimzU+MXhURD5k DIpBDlHydrkZyrSiatpbtZm6TuFN85eKqJWHRV3JN1iuq1KdLj00XoC2+AI6suKx p7K4O7yfi+qWHlTOqjr5T40euiitBGONAb3kzOI9DyDdAYRRSB8xcPqmvLt2g+ie bwGHX1aRQOII2cAHLlYeSp2CN7cisrQOHYY/9FnGQL/Wr/WK/E0+X7Suxuaw8etd NmZyeTz2Mn9MNTMUMTRN1Dg3s7TOsxfDBwu/7OGnLD0o6v1IIdmlsuWrqqkCadyL ijpg/0Ky3jgTQIwnapHcHGezZQGG11xg4Jq0imJwHMqScMjaACU=uv8m
    -----END PGP SIGNATURE-----

    --f1e772351b3b83ef55fb469e35ab45f32cf6ef569b242d30ad58371d7213--

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)