On 2025-01-30 10:08:58 +0100, Johannes Schauer Marin Rodrigues wrote:
> your sources.debian.org link is for mailgraph 1.14-22 but the version in
> bookworm is 1.14-20 and it does not include this regex. I provided the link
> to the bookworm version in my last message.

When I reported

  https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1051496

and provided a patch, this was entirely for bookworm.

Then I don't know why the maintainer
  * downgraded the severity,
  * apparently did not provide a fixed package for bookworm,
  * modified my patch.

What I can tell you is that I tested my patch with both the old
and new formats (after the upgrade, there were still lines with
the old format in the mail logs, i.e. those that where put before
the upgrade), and there were no problems with it.

> And even if I'm wrong and if I'm reading the code wrongly, how do
> you explain the error message that I see with mailgraph in *both* a
> clean Bookworm chroot as well as inside a clean Trixie chroot. What
> kind of reproducer would you like to have if that is not sufficient
> to show the issue?

Perhaps because the maintainer modified my patch?

[...]
> What makes it work is if I add the following to the mailgraph invocation:
> 
>     --logtype rsyslog
> 
> This is not necessary with the patch above which bumps the version of the
> embedded Parse::Syslog.

If you look at

  https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1051496#10

this was not necessary either (this was important, because both
formats had to be used at the same time).

BTW, in the long term, I think that mailgraph should just depend
on libparse-syslog-perl, and Parse::Syslog should no longer be
embedded. I think that this embedding is quite dirty as it starts
with

package Parse::Syslog;

and all the mailgraph code is thus seen as being part of
Parse::Syslog.

-- 
Vincent Lefèvre <vinc...@vinc17.net> - Web: <https://www.vinc17.net/>
100% accessible validated (X)HTML - Blog: <https://www.vinc17.net/blog/>
Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)

Reply via email to