Re: New lsbappchk package and arch strategy
On Fri, Sep 12, 2003 at 06:03:36PM -0600, Matt Taggart wrote:
> The new lsbappchk package
> will FTBFS on several architectures due to the following code in
> src/tests/elfchk/proginterp.c,
I'm inclined towards (a) trying to support all our arches, but not
necessarily succeeding and (b) arch:any and having unsupported arches
FTBFS or otherwise listing the supported arches specifically.
> a.) only support the architectures that upstream claims to support
> * all lsb packages should list only those archs in Architecture:
> * the lsb package should only create symlinks for only those archs
> * the above lsbappchk file above, and other such cases, should
> support only those archs
Having an lsb package on all arches (even ones that don't comply with the
arch-spec because there /isn't/ an arch spec yet) seems reasonable. Having
the tests not exist on arches where they don't build isn't unreasonable.
> b.) support all debian architectures
> * the lsb packages should be Architecture: any or all
> * the lsb package should create symlinks for all archs(this might
> mean making decisions that upstream normally makes, where the
> symlink would point and the official arch name)
> * other lsb packages would need Debian specific changes to add
> support for architectures that upstream does not (yet) support
> c.) some combination
>
> I'm leaning towards option b, but I haven't thought through all the
> implications.
I don't think it much matters?
Cheers,
aj
--
Anthony Towns <aj@humbug.org.au> <http://azure.humbug.org.au/~aj/>
I don't speak for anyone save myself. GPG signed mail preferred.
Australian DMCA (the Digital Agenda Amendments) Under Review!
-- http://azure.humbug.org.au/~aj/blog/copyright/digitalagenda
Reply to: