[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[bug#46866] [PATCH] gnu: b4: Fix setup.py version incompatibility.
From: |
Kyle Meyer |
Subject: |
[bug#46866] [PATCH] gnu: b4: Fix setup.py version incompatibility. |
Date: |
Tue, 09 Mar 2021 17:54:04 -0500 |
Efraim Flashner writes:
> Indeed, it does fail at startup. Two things about your patch, you've
> added a patch file but didn't register it in gnu/local.mk. This means
> it'll get dropped during a 'make dist' which would be a problem.
Doh, thanks for catching that.
> The second is that it's a lot of patch for a small amount of code
> change.
I like that a patch fails loudly once it's no longer valid, but I'm
happy to defer to your opinion that it's an overkill here (and will try
to recalibrate for future patches).
> With your permission I'd like to commit the attached patch instead.
[...]
> From: Kyle Meyer <kyle@kyleam.com>
> Date: Tue, 9 Mar 2021 10:06:13 +0200
> Subject: [PATCH] gnu: b4: Fix setup.py version incompatibility.
>
> * gnu/packages/version-control.scm (b4)[source]: Add snippet to not
> require too specific of a version.
>
> Co-authored with Efraim Flashner <efraim@flashner.co.il>
>
> Signed-off-by: Efraim Flashner <efraim@flashner.co.il>
Sounds fine to me. I have a slight preference that you put yourself as
the author and put a
Noticed-by: Kyle Meyer <kyle@kyleam.com>
or
Reported-by: Kyle Meyer <kyle@kyleam.com>
trailer for me, but either way is okay.
> @@ -2347,7 +2347,15 @@ based on a manifest file published by servers.")
> (method url-fetch)
> (uri (pypi-uri "b4" version))
> (sha256
> - (base32 "1j904dy9cwxl85k2ngc498q5cdnqwsmw3jibjr1m55w8aqdck68z"))))
> + (base32 "1j904dy9cwxl85k2ngc498q5cdnqwsmw3jibjr1m55w8aqdck68z"))
> + (modules '((guix build utils)))
> + (snippet
> + '(begin
> + ;; Fixes issue with dependency requirements being too strict. See
> upstream commit:
> + ;;
> https://git.kernel.org/pub/scm/utils/b4/b4.git/commit/?id=31348a14afdb1d39e7faf9576eaddea1ced76e19
> + (substitute* "setup.py"
> + (("\\.0'") "'"))
Just a note: This won't catch a similar problem if dkimpy (1.0.5) is
upgraded before b4, though perhaps that's unlikely and not worth
worrying about.
Thank you for the review.