Commit access to the new git ? Uploader status ?

Ian Jackson · External Since: Sep 08, 2005 Posts: 70

Would it be OK for me to have full maintainer access to dpkg ? Ie,
access to the primary vcs repository and permission to upload ?

The proximate reason is that there are a number of patches that I've
submitted in bug reports with my Ubuntu hat on which haven't yet been
applied, and it would be Ubuntu's merges easier if the diff was
smaller.

Of course with my Debian hat on it would be nice for me to be able to
commit minor changes as and when it seems appropriate.

I will of course be careful about controversial changes. For example,
I'll refrain from committing my formatting fixup from #375711 until
we've come to a conclusion. I hope you can trust my judgement about
what would be a controversial change.

Thanks,
Ian.

--
To UNSUBSCRIBE, email to debian-dpkg-REQUEST.RemoveThis@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster.RemoveThis@lists.debian.org

Guillem Jover · External Since: Nov 13, 2004 Posts: 429

Hi Ian,

On Wed, 2007-07-11 at 17:17:59 +0100, Ian Jackson wrote:
> Would it be OK for me to have full maintainer access to dpkg ? Ie,
> access to the primary vcs repository and permission to upload ?

Ok. I've added you to the alioth project, so you should have commit
access now to the svn repo. About the uploads, please read Frank's
mail about the organization, those need to be coordinated, but I
don't think we have been lacking in that regard anyway.

> The proximate reason is that there are a number of patches that I've
> submitted in bug reports with my Ubuntu hat on which haven't yet been
> applied, and it would be Ubuntu's merges easier if the diff was
> smaller.

I'm sorry that you have felt the need for this request due to those
patches not being applied, I'll take the blame, as I've been meaning
to merge most of the stuff from Ubuntu (as a side note I'm quite happy
that you have been filing those bug reports with the patches).

The fact that I've not done so has been actually the bug report that
you mention below (which also influenced a bit in me going silent
during last few weeks, the other reason for the silence has been that
I went on holidays and expected I would spare more time for Debian
related stuff, which I didn't).

Some weeks ago I started considering applying them anyway, although
by reverting the formatting changes (which I think should be done
regardless, otherwise the diff gets quite messy).

> Of course with my Debian hat on it would be nice for me to be able to
> commit minor changes as and when it seems appropriate.

Sure.

> I will of course be careful about controversial changes. For example,
> I'll refrain from committing my formatting fixup from #375711 until
> we've come to a conclusion. I hope you can trust my judgement about
> what would be a controversial change.

Sure.

regards,
guillem

--
To UNSUBSCRIBE, email to debian-dpkg-REQUEST DeleteThis @lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster DeleteThis @lists.debian.org

Ian Jackson · External Since: Mar 01, 2005 Posts: 199

Guillem Jover writes ("Re: Commit access to the new git ? Uploader status ?"):
> On Wed, 2007-07-11 at 17:17:59 +0100, Ian Jackson wrote:
> > Would it be OK for me to have full maintainer access to dpkg ? Ie,
> > access to the primary vcs repository and permission to upload ?
>
> Ok. I've added you to the alioth project, so you should have commit
> access now to the svn repo. About the uploads, please read Frank's
> mail about the organization, those need to be coordinated, but I
> don't think we have been lacking in that regard anyway.

Thanks. I'll take a look. (Thanks also for your earlier excellent
mail!)

> > The proximate reason is that there are a number of patches that I've
> > submitted in bug reports with my Ubuntu hat on which haven't yet been
> > applied, and it would be Ubuntu's merges easier if the diff was
> > smaller.
>
> I'm sorry that you have felt the need for this request due to those
> patches not being applied, I'll take the blame, as I've been meaning
> to merge most of the stuff from Ubuntu (as a side note I'm quite happy
> that you have been filing those bug reports with the patches).

Well, there's always an effort problem.

> Some weeks ago I started considering applying them anyway, although
> by reverting the formatting changes (which I think should be done
> regardless, otherwise the diff gets quite messy).

I'll send a separate mail about that I think.

Ian.

--
To UNSUBSCRIBE, email to debian-dpkg-REQUEST DeleteThis @lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster DeleteThis @lists.debian.org

Ian Jackson · External Since: Mar 01, 2005 Posts: 199

Frank Lichtenheld writes ("[long] Dpkg Team Organisation/Status"):
> TODO:
> Ian and Guillem (and maybe others) need to solve their disagreements
> about coding style in the C part of dpkg

I agree that we need to come to a conclusion about this. But there
are some general principles involved which aren't specific to dpkg so
I think debian-devel would be the right venue. I'll post there when I
have a moment.

However ...

Guillem Jover writes ("Re: Commit access to the new git ? Uploader
status ?"):
> [re Breaks:]
>
> Some weeks ago I started considering applying them anyway, although
> by reverting the formatting changes (which I think should be done
> regardless, otherwise the diff gets quite messy).

I think failing to apply that change immediately is a mistake.

The patch I sent in #375711 is purely the reversion of something that
is clearly an earlier mistake. The only change I made there is to
return the indent width from actual tabs implying 8-column-per-indent
to the 2-column-per-indent which is used in all of the rest of the C
and C++ code in the whole of dpkg and which was used in previous
versions of these very same files.

Surely it is clear that the change in those files (and the relevant
part of src/archives.c) compared to earlier versions (cf 1.4.0 for
example) is just a mistake caused by erroneous editor settings or
tab/space conversion ?

This change should be made immediately _even if you think the
currently predominant style ought to be changed_ because it increases
the consistency and because it undoes a previous unintentional and
undesirable change.

Even if one were to want to choose to change the style later it would
make it easier if the style started out more consistent.

So, I hope you'll agree that #375711 should be applied immediately.

Ian.

--
To UNSUBSCRIBE, email to debian-dpkg-REQUEST DeleteThis @lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster DeleteThis @lists.debian.org

Raphael Hertzog · External Since: May 28, 2005 Posts: 461

Hi,

some third-party input on this issue, in the hope that it can help solve
this little different.

On Tue, 17 Jul 2007, Ian Jackson wrote:
> > Some weeks ago I started considering applying them anyway, although
> > by reverting the formatting changes (which I think should be done
> > regardless, otherwise the diff gets quite messy).
>
> I think failing to apply that change immediately is a mistake.
>
> The patch I sent in #375711 is purely the reversion of something that
> is clearly an earlier mistake. The only change I made there is to
> return the indent width from actual tabs implying 8-column-per-indent
> to the 2-column-per-indent which is used in all of the rest of the C
> and C++ code in the whole of dpkg and which was used in previous
> versions of these very same files.

I agree that it's best if the indentation in all the C code is consistent.
In that regard, I'd suggest to apply the patch.

However I clearly dislike the 2-char indent (4 chars are the minimum IMO
to be readable) and wouldn't oppose a mass reformat later on.

> This change should be made immediately _even if you think the
> currently predominant style ought to be changed_ because it increases
> the consistency and because it undoes a previous unintentional and
> undesirable change.

I think the changes were intentionnal. It's just that Guillem finds more
reasonable to convert slowly the formatting as he works through the code.
As it happens, I disagree with this point of view.

Cheers,
--
Raphaël Hertzog

Premier livre français sur Debian GNU/Linux :
http://www.ouaza.com/livre/admin-debian/

--
To UNSUBSCRIBE, email to debian-dpkg-REQUEST.TakeThisOut@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster.TakeThisOut@lists.debian.org