Re: Uncleaned egg-info directory giving lots of bugs about failing to build after successful build
Hi Soren
On 07/09/2023 01:05, Soren Stoutner wrote:
Also, the commit linked to below states:
"if the egg-info directory is included in the upstream source, then
deleting it causes no harm as dpkg-source warns on missing files in
the source package but does not exit with error.”
It might not cause any problems for dpkg-source, but won’t it cause
issues for double builds? For reference:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1044268
<https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1044268>
If so, any package that ships the egg-info directory in the tarball will
get flagged with a FTBFS bug, which might be the desired behavior if
these shouldn’t be in the tarball anyway.
No, the opposite is true.
They _currently_ FTBFS when built twice in a row because:
* the build modifies the egg-info files
* the change to the egg-info is not undone by the clean step that is
run between the two builds
* dpkg-source doesn't want modified upstream files and so exits
with error
With the change you referred to, these packages do not FTBFS when built
twice in a row because:
* the build modifies the egg-info files
* the egg-info is removed in the clean step between that runs between
the two build attempts
* dpkg-source doesn't care about removed upstream files and so
completes its task without error (it warns about them, that's all)
* the build proceeds and regenerates the egg-info
It was pointed out on IRC today that some packages (like poetry) ship
egg-info files that are part of the test suite and so are not
regenerated... and they should not be deleted. That's the next task.
regards
Stuart
--
Stuart Prescott http://www.nanonanonano.net/ stuart@nanonanonano.net
Debian Developer http://www.debian.org/ stuart@debian.org
GPG fingerprint 90E2 D2C1 AD14 6A1B 7EBB 891D BBC1 7EBB 1396 F2F7
Reply to: