9:34 p.m.
On Tue, 2003-07-29 at 23:20, Adrian Aichner wrote:
Ville> MAINTAINERS should be updated. Dunno about the
website, is it
Ville> autogenerated from $something (package-index?) nowadays?
[snip replies]
Ok, thanks for the clarifications, folks. I'm trying to remember all
the places that need to be touched when a new package is added.
One thing not mentioned is etc/PACKAGES in core. What's the status of
that? Seems terribly outdated to me (both 21.[45] say "Up-to-date as of
May 15, 2001" which probably is true). Should it just be removed?
man/xemacs/packages.texi is missing both hyperbole and oo-browser in
21.[45].
Also, README.(HYPERBOLE|OO-BROWSER) in 21.4 is not up to date with
21.5. But it is very much possible that I've missed an already sent
patch.
etc/package-index.LATEST* is updated just before a core release, I
guess.
Eek. Far too many places to update.
--
\/
10:03 p.m.
New subject: [COMMIT] Final Patches Before Initial Package Release
>>>> "Ville" == Ville Skytt <Ville>
writes:
Ville> On Tue, 2003-07-29 at 23:20, Adrian Aichner wrote:
Ville> MAINTAINERS should be updated. Dunno about the website, is it
Ville> autogenerated from $something (package-index?) nowadays?
Ville> [snip replies]
Ville> Ok, thanks for the clarifications, folks. I'm trying to
Ville> remember all the places that need to be touched when a new
Ville> package is added.
Ville> One thing not mentioned is etc/PACKAGES in core. What's
Ville> the status of that? Seems terribly outdated to me (both
Ville> 21.[45] say "Up-to-date as of May 15, 2001" which probably
Ville> is true). Should it just be removed?
Ville> man/xemacs/packages.texi is missing both hyperbole and
Ville> oo-browser in 21.[45].
Ville> Also, README.(HYPERBOLE|OO-BROWSER) in 21.4 is not up to date with
Ville> 21.5. But it is very much possible that I've missed an already sent
Ville> patch.
Ville> etc/package-index.LATEST* is updated just before a core
Ville> release, I guess.
Ville> Eek. Far too many places to update.
Yep, I genuinely hate duplication of information.
Ville> --
Ville> \/
--
Adrian Aichner
mailto:adrian＠xemacs.org
http://www.xemacs.org/
2:44 p.m.
New subject: [COMMIT] Final Patches Before Initial Package Release
>>>> "VS" == Ville Skytt <Ville> writes:
VS> Also, README.(HYPERBOLE|OO-BROWSER) in 21.4 is not up to date with
VS> 21.5. But it is very much possible that I've missed an already sent
VS> patch.
Stephen submitted a patch for both 21.4 and 21.5. to update those two
readmes.
--
Jake Colman -- OO-Browser Package Maintainer
4:25 p.m.
New subject: [COMMIT] Final Patches Before Initial Package Release
Ville Skyttä <scop(a)xemacs.org> writes:
One thing not mentioned is etc/PACKAGES in core. What's the
status of
that? Seems terribly outdated to me (both 21.[45] say "Up-to-date as of
May 15, 2001" which probably is true). Should it just be removed?
seconded ... *drool* :-)
man/xemacs/packages.texi is missing both hyperbole and oo-browser in
21.[45].
See below.
etc/package-index.LATEST* is updated just before a core release, I
guess.
I update the 21.5 branch after each official package release and
submit a patch against the 21.4 version. I think packages.texi should
be handled alike.
norbert.
4:17 a.m.
New subject: [COMMIT] Final Patches Before Initial Package Release
>>>> "viteno" == Norbert Koch
<viteno(a)xemacs.org> writes:
viteno> Ville Skyttä <scop(a)xemacs.org> writes:
> One thing not mentioned is etc/PACKAGES in core. What's the
> status of that? Seems terribly outdated to me (both 21.[45]
> say "Up-to-date as of May 15, 2001" which probably is true).
> Should it just be removed?
viteno> seconded ... *drool* :-)
If you're not going to update it, Norbert, submit a patch to kill it.
I think that is sufficiently important documentation that it should be
discussed, but when you and Ville both say it's out-of-date and
redundant, I don't think there will be much objection.
viteno> I update the 21.5 branch after each official package
viteno> release and submit a patch against the 21.4 version.
Right. The basic principles we operate on are
(1) Nobody except the maintainer is trusted with stable code == the
anointed core branch, currently 21.4, and packages with explicit
maintainers.
(2) Reviewers can commit or authorize commit to everything else.
People occasionally make private branches (eg, the unsupported
tree in packages or the Carbon and Qt branches of the core), and
generally that should be (and is) respected.
Everything has exceptions, of course, but that's the idea.
viteno> I think packages.texi should be handled alike.
I disagree. I think the documentation should go into the packages.
Also, ISTR texinfo has a texi2txt option. We should texinfo-ize
INSTALL, README etc and use that to generate it. It doesn't really
matter how ugly, right, as long as <xemacs-beta(a)xemacs.org> comes
through clearly. ;-)
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Ask not how you can "do" free software business;
ask what your business can "do for" free software.
5:31 a.m.
New subject: [COMMIT] Final Patches Before Initial Package Release
"Stephen J. Turnbull" <stephen(a)xemacs.org> writes:
If you're not going to update it, Norbert, submit a patch to kill
it.
I think that is sufficiently important documentation that it should be
discussed, but when you and Ville both say it's out-of-date and
redundant, I don't think there will be much objection.
If you think, it's valuable to have this file around, I simply bring
it up to date and keep it in sync, no big deal.
I only seconded the notion about the too many places to look for (and
update) the same kind of information.
viteno> I think packages.texi should be handled alike.
I disagree. I think the documentation should go into the packages.
This is true. I was only referring to the "when" of updating the
information. IMO it's sufficient to update this file after official
releases that contain new packages (eg hyperbole last time and
oo-browser next time). It's not necessary to put new packages right
in when they get first pre-released. YMMV.
Also, ISTR texinfo has a texi2txt option. We should texinfo-ize
INSTALL, README etc and use that to generate it. It doesn't really
matter how ugly, right, as long as <xemacs-beta(a)xemacs.org> comes
through clearly. ;-)
What would you suggest? Add a texi directory to the package tree top
and put all this stuff in there? Would you like the generated files,
ie the text versions also keep in CVS?
norbert.
7574
days inactive
7576
days old
5 comments
5 participants
participants (5)
-
Adrian Aichner -
Jake Colman -
Norbert Koch -
Stephen J. Turnbull -
Ville Skyttä