Andrew, On Fri, Sep 03, 2004 at 04:22:50PM +1000, Andrew Clausen wrote: > On Thu, Sep 02, 2004 at 09:02:20AM +0200, Sven Luther wrote: > > I had some discussion with Steve Langasek, one of debian's release manager, > > about including parted 1.6.12 in sarge. Since parted is ipart of base, and > > has > > thus been frozen since more or less amonth, this cause problem, and we were > > wondering if your fix could not be backported.
> > Steve (vorlon on irc) suggested : > > 08:45 < vorlon> svenl: it looks to me like the parted bug could be > > backported w/o ABI breakage by hiding the new geom info in the > > arch_specific > > member. > > I have some reservation, given that libparted really has no concept of > > hidden > > data structures, and mostly exports everything, but i suppose that the two > > new > > geom stuff could be added at the end of the structure and still keep > > backward > > compatibility. Not sure though. > > Do you have any insight on this question ? > I agree that putting stuff in arch_specific is "hiding", and you could > maintain backward compatability using this trick. It is conceptually > very ugly, and I don't want to do this work myself. It would be painful > for you Debian people to maintain the Parted package like this > long-term. > To be honest, I don't see why you can't just recompile packages to use > 1.6.12. I doubt there will be any compilation problems. If there were > any, they would be minor - things like replacing dev->sectors with > dev->bios_geom.sectors. How many packages are we talking about anyway? > I think this is a time where you need to look at the code and say: > these are really trivially changes, and aren't going to break anything. > I think you should be far more concerned with the possibility of new > bugs in 1.6.12. Any change that requires an ABI change and a subsequent soname bump and recompile of all related software is not one that I would call "trivial". We are in the late stages of a freeze in preparation for the next stable release of Debian, and parted is heavily used by our installer -- which is what makes this particular bug so important at present, but it also makes a forced recompile awkward and rather irritating. I defer to your expertise when it comes to decisions of what the libparted ABI should look like in the long term; this particular ABI change just happens to be particularly ill-timed from our POV. I'm willing to do the work necessary for a backport, but it seems the public CVS revision history of libparted is unfortunately rather spotty. :/ -- Steve Langasek postmodern programmer
signature.asc
Description: Digital signature
