[Dbconfig-common-devel] dbconfig-common upload?

[sean finney]

hiya,

On Tuesday 22 April 2008 10:08:36 pm Daniel Hahler wrote:
> Yes, the timeline is very tight. The "Universe" repository gets frozen
> tomorrow, so I'll prepare an upload of 1.8.38 with this additional fix.

heh, a bit last minute to be contacting the package maintainer then.  

please make sure to make it clear then that this is a patched version by 
adding something to the changelog stating this, as well as adding something 
distinct to the version string (since this is a native package, we don't have 
the luxury of dropping in a new quilt patch or something).  please also keep 
the version << 1.8.39, which will be the version in the next upload.  i 
suggest 1.8.38ubuntublah0 or 1.8.38.0ubuntublah or similar.

> The fix itself works and is actually nicer than adding "|| true" to the
> rmdir, like it was done before for the older entries.

good to know.

> I couldn't find a link to the SVN repo from
> http://alioth.debian.org/projects/dbconfig-common/ - but Google helped
> me out.
> You may want to add Vcs-headers to debian/control on the next upload, too.

indeed.  i've been slowly adding this in for the packages i maintain, but i 
guess dbconfig-common hasn't had that done yet.

> > looking at the open bugs, i think the following two would be wise to have
> > fixed for a stable release:
> >
> > #476946: dbconfig-common does not honor preseeded values, if priority is
> > too low
> >
> > (haven't verified this one)
>
> Is this a regression? it is the only other bug tagged "Version 1.3.38"
> IIRC. As far as I understand this report, this is the desired behavior -
> during setup of dbconfig-common you're asked if you want to have
> questions related to remote DB servers and it tells you that this
> influences the default set of questions.
> That should probably be different with preseeded values though.

like i said, i haven't verified (or perhaps "reviewed" would be a better word) 
this bug, so i can't comment on it yet.

> Do you think it's sensible to push 1.8.38 (with the additional fix) for
> Ubuntu, or is 1.8.36 safer (despite the number of bugs fixed in 1.8.38)?

i would say definitely go with the latest and greatest.  while the diff might 
be fairly large, almost all of it is work i've been doing on a new test suite 
and is not "live" code.  every other change is a translation update, doc fix, 
lintian fix, or bug fix (including one security-related fix).


	sean
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part.
Url : http://lists.alioth.debian.org/pipermail/dbconfig-common-devel/attachments/20080423/e05e421f/attachment.pgp