[xml/sgml] second request for review and upload of xerces packages

[Jay Berkenbilt]

Hello all --

It's been just over a week now since I prepared libxml-xerces-perl,
xerces23, xerces24, and xerces25 for upload.  My original
announcements included copious information about exactly what I did
and where to download the .orig.tar.gz files from.

Here is a message I had previously sent to Ivo (who is on the list)
with a CC to the list:

http://lists.alioth.debian.org/pipermail/debian-xml-sgml-devel/2004-April/000148.html

I'd be quite grateful if someone could upload these.  (I hope I'm not
out of line by going back to the list after a week to ask that someone
take care of this, but I don't think I am -- after all, that's why we
are using collaborative maintenance.)

Of course, as I am not (yet?) a DD, you will probably want to review
my work.  The above referenced message gives plenty of information
that should help with that process, and my debian/rules files are
well-commented.  I'd suggest the following review procedure:

 * libxml-xerces-perl is identical except for changelog, control,
   copyright, and possibly whitespace changes

 * xerces23 has been completely repackaged but verified to achieve the
   same results except for fixing lintian warnings.  Review
   debian/rules and read the comments

 * xerces24 and xerces25 are essentially identical to xerces23 -- look
   at the diffs between debian/rules in the three packages

I've already verified that all the packages are linda- and
lintian-clean and build in pbuilder, and I've done a file-by-file
comparison (as described in other messages) of the xerces23 packages.
I'm quite confidence that the packaging is correct.

Please let me know if there's anything else I should do to help with
this process.  Thanks!

Oh, one more thing: It is not at all clear to me how to make
svn-buildpackage find upstream tarballs in the repository, hence my
use of the local tarballs.  The use of tarballs.mk and
simple-patchsys.mk is fundamental to how the package creation works
for xerces (since it configures the system multiple ways), so I don't
want to just check the whole upstream source into subversion.
Besides, it's very big and the packaging is very small.

--Jay

-- 
Jay Berkenbilt <ejb@ql.org>
http://www.ql.org/q/