[parted-devel] please support m68k atari partitioning

[Stephen R Marenka]

On Wed, Aug 26, 2009 at 04:44:57PM +0200, Jim Meyering wrote:
> Stephen R Marenka wrote:
> 
> > Please support m68k atari partitioning. I'm sure this isn't high
> > on anyone's list, but it would be nice if we could get it into
> > upstream before it bitrots beyond my poor abilities of repair.
> >
> > The patch seems quite well tested at this point and it's pretty
> > isolated from everything else, so it should be low risk. You can
> > pull from the atari branch of my git repo at [1].
> >
> > The patch was originally written by "K.G." <k_guillaume at libertysurf.fr>
> > almost 4 years ago. It's history is in the bug report at [2].
> >
> > I can also provide a straight patch if that is desired.
> 
> Thanks for posting.
> 
> A git patch series would be great, either via a repository
> or simply "git format-patch" output.

I'll give that a try. The full patch was longer than the list would
allow, so I'll have to try -n I guess.

> Are the changes mentioned above already relative to parted's "next" branch?

They were forward ported to master.

> If you're still interested in pursuing this, please rebase to next
> and ensure that "make distcheck" still passes.

That will be something new to learn.

> That runs all tests with a few different simulated sector sizes.
> Also, please attribute carefully (in commit logs) any changes that you
> have not written yourself.  We'll need to arrange to get a copyright
> assignment from anyone who has contributed more than a few lines.
> See this link
>   http://git.sv.gnu.org/cgit/coreutils.git/plain/HACKING
> for details on filling out copyright assignment forms.

At this point I haven't actually written anything, I've just moved stuff
around to make it compile for the 1.8 series.

> Also, in order to expedite acceptance, it would be good to include a
> test script that creates and manipulates atari partition tables.
> For a most basic example, see tests/t0250-gpt.sh.
> Ideally, your tests would also add a few empty partitions,
> print/check results, remove a few,
> print/check results again.

I guess I can give that a shot.

> Be sure not to actually create file systems in those partitions.
> We're trying to avoid using parted for FS-manipulation, because
> that code has bit-rotted so much as to be unmaintainable.

Will do.

Thanks,

Stephen

-- 
Stephen R. Marenka     If life's not fun, you're not doing it right!
<stephen at marenka.net>