Re: gnome-1.4.1 beta1

[Laszlo PETER <Laszlo Peter ireland sun com>]

Hi,

I'm not a maintainer, so I'm just guessing. I can think of 3 reasons
why the tarball contents may differ from the CVS tags:
  - the maintainer simply forgets to create the tag right after
    creating the tarball.
  - the maintainer creates the tarball and the tag, then finds that
    something's wrong, fixes the tarball but leaves the tag as it is
  - horribile dictum: the tarball is not created from a clean cvs checkout
    but a workspace with uncommitted changes.

Maintainers, please correct me if I'm wrong.

The question is: would you guys use a script like the one Colm proposed?

Laca

Colm Smyth wrote:

> Warning: "obvious" statement follows...
> 
> It might be easier if tarballs were created directly from CVS as
> the CVS tree (mostly) does not contain generated files. People
> have to "do the right thing" to put their files in CVS, why
> not use it?
> 
> Maybe if we had a script that automated tarball creation
> using CVS, people would be more inclined to do it this way.
> 
> Something like:
> 
> usage: cvsball <module> <cvs-tag>
> creates a gzipped tar archive (aka tarball) of a CVS module using
> the cvs-tag to identify file versions.
> 
> notes:
> * uses $TMPDIR as working disk space while getting the files from CVS
> * standard CVS environment variables apply
> 
> Of course if 'make clean' worked, commits to CVS would be easier
> (and people with CVS-aversion could still make portable tarballs).
> 
> Colm.

> >> I don't always trust that a tag in CVS with the same name as the release
> >> tarball is *exactly* the same as what's in the tarball. Try doing some
> >> diffs and you'll see that people don't always do that right.

> >So is there a reason why maintainers can't do The Right Thing when releasing
> >a tarball?