Bug#159286: tetex-bin is not a native Debian package

["C.M. Connelly" <c@eskimo.com>]

"AK"  == Atsuhito Kohda <kohda@pm.tokushima-u.ac.jp>
"CMC" == C.M. Connelly <cmc@debian.org>
"TE"  == Thomas Esser <te@informatik.uni-hannover.de>


    CMC> It can be fixed pretty easily by just rebuilding the
    CMC> package with everything accessible, which I did, but the
    CMC> ftp admins rejected the packages because they were the
    CMC> same version as the previous package.

[I was wrong here; see below.]

    AK> Only for sake of preciseness, I would like to know even if
    AK> you upgraded to 1.0.7+20011202-8 (NOT -7) with the correct
    AK> tetex-bin_1.0.7+20011202.orig.tar.gz, it failed to upload?

I'm not sure.  It turns out that I was a bit confused -- my
attempts to change the tar file were related to tetex-base, not
tetex-bin.  (And were rejected because the MD5 sums wouldn't
match, so older versions couldn't be built from source.)

However, the problem was still caused by the cvs-buildpackage
thinko.

    AK> # And why a native Debian tetex-bin was accepted?

Couldn't tell you.


    CMC> So the most sensible thing to do, I guess, is to update
    CMC> our tetex-bin to match Thomas's latest beta, then build
    CMC> brand new packages with a brand new .orig.tar.gz.

    AK> I checked beta version a bit and it seemed to be
    AK> teTeX-src-beta-20020901.tar.gz at present.

    AK> Then we should buid the package with version
    AK> 1.0.7+20020901 ?  I'm afraid that 1.0.7 is not so correct
    AK> in this case...

Dunno what the best bet is.  If all Thomas has for a version
number is 20020901, then maybe we're stuck with 1.0.7+20020901.
But maybe we could get more information from him about what
version number he prefers.


    TE> this was fixed in the teTeX-beta from may. Please, make
    TE> sure not to use the latest teTeX-texmf tarball in the
    TE> debian packages.

    AK> And from the above, I understood we only would update
    AK> tetex-bin to the beta version and did not touch tetex-base
    AK> at all, is this right?

My interpretation of what Thomas was saying was that we should use
the beta release of the texmf tree rather than his current
development tree, which makes sense.  (Presumably the beta is
roughly what ended up in TeX Live.)

Now that I've actually looked at the source archives, though, I
think he's saying *don't use the 20020901 release*, which makes
sense given that it's brand new, and there's been little or no
feedback.

He hasn't said anything about tetex-bin, but sticking with the
beta there is probably sound advice, as well.

If you're not sure what the best files to use are, I would just
ask him!

   CMC

+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+
 Man cannot be civilised, or be kept civilised by what he does in his
	    spare time; only by what he does as his work.
			     W.R. Lethaby
+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+
  C.M. Connelly               c@eskimo.com                   SHC, DS
+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+