Re: RFS: lbzip2

[ERSEK Laszlo <lacos@elte.hu>]

Paul Wise wrote:

> The reason is "that's the way we do it in Debian". The policy manual
> probably has a rationale.

[...]

> Hmm, which Debian architectures do you test on?
> 
> Perhaps /dev/urandom could be a good source of data.

[...]

> Yeah, always run lintian in sid and build/test your packages on a sid
> system (chroot, vm, etc).

[...]

> Use quilt or dpatch. This isn't nessecary (P = pedantic), especially
> when you as upstream are also maintaining the package.

I uploaded a new build of the package:

http://mentors.debian.net/debian/pool/main/l/lbzip2

Changes since last upload:
- package is built in a sid chroot
- added "debian-sanity" target to Makefile which executes basic
compression/decompressin checks, using 10^7 bytes from /dev/urandom
- "debian-sanity" depends on "lbzip2"
- the "install" target depends on "debian-sanity"
- the "clean" target removes "debian-sanity"'s temporary files
- since "debian-sanity" uses uuencode (to give lbzip2 a bit more work),
added "sharutils" as a new build dependency
- added missing comma (,) before ${misc:Depends}
- CFLAGS contains -g3 unconditionally
- updated and reformatted debian/changelog
- the only remaining lintian lament is the pedantic
"direct-changes-in-diff-but-no-patch-system" which I don't have to fix,
according to the last citation above.

Thanks,
lacos