7:59 a.m.
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hello.
The following XEmacs packages have just been released.
Notes
there are 4 new packages in this release
fortran-modes, perl-modes, python-modes and ruby-modes
they have been split out of prog-modes so if you used any of those
modes you will have to install the new packages
Updated or New Packages:
=======================
calendar-1.19-pkg.tar.gz
debug-1.16-pkg.tar.gz
ecrypto-0.12-pkg.tar.gz
ediff-1.42-pkg.tar.gz
fortran-modes-1.00-pkg.tar.gz *** New ***
gnus-1.64-pkg.tar.gz
idlwave-1.26-pkg.tar.gz
jde-1.36-pkg.tar.gz
liece-1.08-pkg.tar.gz
locale-1.19-pkg.tar.gz *** Mule ***
lookup-1.13-pkg.tar.gz *** Mule ***
mail-lib-1.52-pkg.tar.gz
perl-modes-1.00-pkg.tar.gz *** New ***
prog-modes-1.66-pkg.tar.gz
python-modes-1.00-pkg.tar.gz *** New ***
ruby-modes-1.00-pkg.tar.gz *** New ***
skk-1.23-pkg.tar.gz *** Mule ***
viper-1.35-pkg.tar.gz
w3-1.26-pkg.tar.gz
xemacs-base-1.69-pkg.tar.gz
Detailed Changes:
================
calendar-1.19-pkg.tar.gz
- ------------------------
2002-09-26 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.19 released.
2002-09-25 A. Guru <a.guru(a)sympatico.ca>
* solar.el (solar-equinoxes-solstices): Fix off by one day bug
affecting eastern US autumnal equinox this year.
debug-1.16-pkg.tar.gz
- ---------------------
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.16 released.
2002-09-22 Scott Douglass <sdouglass(a)arm.com>
* gdb.el (gdb-filter-insert): Fix broken comint-kill-output
behavior in gdb-mode.
2002-07-15 Adrian Aichner <adrian(a)xemacs.org>
* pdb.el: It's XEmacs, not Xemacs.
ecrypto-0.12-pkg.tar.gz
- -----------------------
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 0.12 released.
2002-09-17 Ville Skyttä <ville.skytta(a)xemacs.org>
* md5-el.el (md5-update): Comment typo fix.
ediff-1.42-pkg.tar.gz
- ---------------------
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.42 released.
2002-09-23 Michael Kifer <kifer(a)cs.stonybrook.edu>
* ediff-init.el (ediff-get-next-window): bug fix.
* ediff-hook.el (ediff-cond-compile-for-xemacs-or-emacs): Put the
definition straight into loaddefs.el as well as executing it here
(by RMS)
2002-09-18 Michael Kifer <kifer(a)cs.stonybrook.edu>
* ediff-hooks.el: Put back the autoloads (for compatibility with XEmacs).
* ediff-init.el: Use defalias instead of fset.
* ediff-util.el: Use defalias instead of fset.
* ediff-ptch.el (ediff-test-patch-utility): catch errors
* ediff-wind.el (ediff-window-setup-function): better choices
* ediff-backup.el (ediff-backup): new function
* ediff.texi: typo
2002-07-16 Adrian Aichner <adrian(a)xemacs.org>
* .cvsignore: Add generated .info* and .html files.
fortran-modes-1.00-pkg.tar.gz *** New ***
- -----------------------------
2002-09-30 Ville Skyttä <ville.skytta(a)xemacs.org>
* fortran.el: Add autoloaded auto-mode-alist associations.
* f90.el: Ditto.
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.00 released.
2002-09-18 Ville Skyttä <ville.skytta(a)xemacs.org>
* Package created, moved f90, fortran-mode and fortran-misc
from the prog-modes package here.
gnus-1.64-pkg.tar.gz
- --------------------
2002-09-26 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.64 released.
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.63 released.
2002-09-17 Ville Skyttä <ville.skytta(a)xemacs.org>
* Makefile (REQUIRES): Add ecrypto, we'll use md5 from there.
(EXTRA_SOURCES): New, add gnus/lisp/md5.el.upstream.
* package-info.in (provides): Remove md5.
idlwave-1.26-pkg.tar.gz
- -----------------------
2002-09-30 Ville Skyttä <ville.skytta(a)xemacs.org>
* idlwave.el: Add autoloaded auto-mode-alist association.
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.26 released.
2002-09-16 Carsten Dominik <dominik(a)astro.uva.nl>
* idlwave.el: Upgraded to IDLWAVE version 4.15
2002-07-16 Adrian Aichner <adrian(a)xemacs.org>
* .cvsignore: Add generated .info* and .html files.
jde-1.36-pkg.tar.gz
- -------------------
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.36 released.
2002-09-13 Andy Piper <andy(a)xemacs.org>
* Makefile: sync with jde 2.2.9beta12. Register moiself as the
XEmacs maintainer.
* ReleaseNotes.txt: move to standard jde location.
* jde-checkstyle.el (jde-checkstyle-exec): use jde-build-classpath
so that things work under cygwin XEmacs.
* jde-java-font-lock.el (jde-java-font-lock-html-tag-regexp):
required during compilation.
(jde-java-font-lock-def-html-keyword): keyword is not used.
* jde-wiz.el (semantic-util): Require semantic-util.
2002-08-14 Andy Piper <andy(a)xemacs.org>
* ClassPathEntry.java (addClass): map fully qualified names and
cope with classes that don't have a packaged.
2002-07-15 Adrian Aichner <adrian(a)xemacs.org>
* lisp/jde-java-font-lock.el (jde-setup-syntax-coloring): It's
XEmacs, not Xemacs.
liece-1.08-pkg.tar.gz
- ---------------------
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.08 released.
2002-09-19 Adrian Aichner <adrian(a)xemacs.org>
* doc/Makefile: New file to generate html documentation.
2002-09-19 Adrian Aichner <adrian(a)xemacs.org>
* Makefile (html): New.
locale-1.19-pkg.tar.gz *** Mule ***
- ----------------------
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.19 released.
2002-09-07 Mike Fabian <mfabian(a)suse.de>
* start-files/ja/locale-start.el (command-line-do-help):
Use the correct encoding for the output of
'LC_ALL=ja_JP.eucJP xemacs --help',
'LC_ALL=ja_JP.UTF-8 xemacs --help', and
'LC_ALL=ja_JP.SJIS xemacs --help'.
2002-01-23 Stephen J. Turnbull <stephen(a)xemacs.org>
* Makefile (.el.elc): Remove special .el.elc rules.
lookup-1.13-pkg.tar.gz *** Mule ***
- - ----------------------
rebuild info pages with xemacs 21.4
mail-lib-1.52-pkg.tar.gz
- ------------------------
2002-09-30 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.52 released.
2002-09-28 Simon Josefsson <jas(a)extundo.com>
* smtpmail.el (smtpmail-via-smtp): Only negotiate starttls on
streams that were opened using starttls.
2002-09-26 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.51 released.
2002-09-24 Simon Josefsson <jas(a)extundo.com>
* mail/mail-extr.el (mail-extr-ignore-single-names): Change default.
(mail-extract-address-components): Doc fix.
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.50 released.
2002-09-17 Simon Josefsson <jas(a)extundo.com>
* smtpmail.el: Sync with Emacs CVS (following ChangeLog entries
from Emacs, until 2002-09-16 change).
2002-09-12 Simon Josefsson <jas(a)extundo.com>
* mail/smtpmail.el (smtpmail-send-it): Don't use : in filenames
(for cygwin). Suggested by Andrew Senior <aws(a)watson.ibm.com>.
Use expand-file-name. Also don't require time-stamp.
2002-09-11 Simon Josefsson <jas(a)extundo.com>
* mail/smtpmail.el (smtpmail-default-smtp-server): Doc fix.
2002-08-30 Simon Josefsson <jas(a)extundo.com>
* mail/smtpmail.el (smtpmail-via-smtp): Delete process buffer
unless s-debug-info is enabled.
2002-05-12 Richard M. Stallman <rms(a)gnu.org>
* mail/smtpmail.el (smtpmail-mail-address): New variable.
(smtpmail-send-it): Bind and use that instead of `mail-address'.
(smtpmail-via-smtp): Likewise.
2002-05-09 Simon Josefsson <jas(a)extundo.com>
* mail/smtpmail.el (smtpmail-send-it): Use user-mail-address from
calling buffer.
2002-05-08 Richard M. Stallman <rms(a)gnu.org>
* mail/smtpmail.el (netrc-machine, netrc-get): Add autoloads.
2002-09-16 Simon Josefsson <jas(a)extundo.com>
* Makefile (REQUIRES): Add ecrypto.
perl-modes-1.00-pkg.tar.gz *** New ***
- --------------------------
2002-09-30 Ville Skyttä <ville.skytta(a)xemacs.org>
* cperl-mode.el: Add autoloaded auto-mode-alist and
interpreter-mode-alist associations.
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.00 released.
2002-09-20 Ville Skyttä <ville.skytta(a)xemacs.org>
* Package created, moved perl-mode and cperl-mode from the
prog-modes package here.
prog-modes-1.66-pkg.tar.gz
- --------------------------
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.66 released.
2002-09-20 Ville Skyttä <ville.skytta(a)xemacs.org>
* Makefile (ELCS): Remove cperl-mode.elc and perl-mode.elc.
(REQUIRES): Remove ps-print, was needed by cperl-mode.
* package-info.in (provides): Remove cperl-mode and perl-mode.
* cperl-mode.el: Remove, move to perl-modes package.
* perl-mode.el: Ditto.
2002-09-18 Ville Skyttä <ville.skytta(a)xemacs.org>
* Makefile (ELCS): Remove ruby-mode.elc and inf-ruby.elc.
* package-info.in (provides): Remove inf-ruby and ruby-mode.
* inf-ruby.el: Remove, move to ruby-modes package.
* inf-ruby.el.upstream: Ditto.
* ruby-mode.el: Ditto.
* ruby-mode.el.upstream: Ditto.
2002-09-18 Ville Skyttä <ville.skytta(a)xemacs.org>
* Makefile (ELCS): Remove f90.elc, fortran.elc and
fortran-misc.elc.
* package-info.in (provides): Remove f90 and fortran.
* f90.el: Remove, move to fortran-modes package.
* fortran.el: Ditto.
* fortran-misc.el: Ditto.
2002-09-18 Ville Skyttä <ville.skytta(a)xemacs.org>
* Makefile (ELCS): Remove python-mode.elc and pydoc.elc.
(EXTRA_SOURCES): Removed.
(MAINTAINER): Change to xemacs-beta.
* package-info.in (provides): Remove python-mode and pydoc.
* pydoc.el: Removed, moved to python-modes package.
* pydoc.el.upstream: Ditto.
* pydoc-el-README: Ditto.
* pydoc_lisp.py: Ditto.
* python-mode.el: Ditto.
* python-mode.el.upstream: Ditto.
2002-09-11 Ville Skyttä <ville.skytta(a)xemacs.org>
* javascript-mode.el (javascript-mode): Set menu name to
JavaScript (it's not C++), thanks to Geert Ribbers.
(javascript): Case tweak for custom group in menus.
python-modes-1.00-pkg.tar.gz *** New ***
- ----------------------------
2002-09-30 Ville Skyttä <ville.skytta(a)xemacs.org>
* python-mode.el: Add autoloaded auto-mode-alist and
interpreter-mode-alist associations.
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.00 released.
2002-09-18 Ville Skyttä <ville.skytta(a)xemacs.org>
* Package created, moved pydoc and python-mode from the
prog-modes package here.
ruby-modes-1.00-pkg.tar.gz *** New ***
- --------------------------
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.00 released.
2002-09-18 Ville Skyttä <ville.skytta(a)xemacs.org>
* Package created, moved Ruby modes from the prog-modes
package here.
skk-1.23-pkg.tar.gz *** Mule ***
- -------------------
rebuild info pages with xemacs 21.4
viper-1.35-pkg.tar.gz
- ---------------------
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.35 released.
2002-09-18 Michael Kifer <kifer(a)cs.stonybrook.edu>
* viper-util.el (viper-chars-in-region): simplification.
* viper.el (viper-emacs-state-mode-list): added modes.
* viper-cmd.el, viper-ex.el, viper.texi: typos.
2002-07-16 Adrian Aichner <adrian(a)xemacs.org>
* .cvsignore: Add generated .info* and .html files.
w3-1.26-pkg.tar.gz
- ------------------
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.26 released.
2002-09-17 Ville Skyttä <ville.skytta(a)xemacs.org>
* Makefile (REQUIRES): Add ecrypto, we'll use md5 from there.
(EXTRA_SOURCES): Add lisp/md5.el.upstream.
* package-info.in (provides): Remove md5.
* lisp/md5.el: Renamed to lisp/md5.el.upstream.
xemacs-base-1.69-pkg.tar.gz
- ---------------------------
2002-09-30 Ville Skyttä <ville.skytta(a)xemacs.org>
* add-log.el: Add autoloaded auto-mode-alist associations.
2002-09-25 Rendhalver [Peter Brown] <rendhalver(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.69 released.
2002-09-11 Evgeny Roubinchtein <zhenya(a)sdf.lonestar.org>
* easy-mmode.el (define-minor-mode): Use add-minor-mode to add
the keymap to minor-mode-alist and add the modeline indicator.
2002-09-11 Simon Josefsson <jas(a)extundo.com>
* time-stamp.el (time-stamp-hhmmss): New function.
Installing These:
================
Manually:
- --------
1) Download the packages that you want to install from: [1]
/ftp.xemacs.org:/pub/xemacs/packages/
2) Unpack them to: [2]
/usr/local/lib/xemacs/xemacs-packages/
3) Re-start XEmacs.
Using XEmacs Package Tools (XEmacs 21.[245].x):
- ----------------------------------------------
1) Tools -> Packages -> Add Download Site -> choose site
2) Tools -> Packages -> List and Install
3) Select the packages you wish to install (there are brief
instructions at the bottom of the packages buffer).
4) Packages -> Install/Remove Selected
5) Re-start XEmacs.
Using XEmacs Package Tools (XEmacs 21.1.14):
- -------------------------------------------
1) Options -> Manage Packages -> Add Download Site -> choose site
2) Options -> Manage Packages -> List and Install
3 - 5) As per XEmacs 21.[245].x.
Footnotes:
[1] Or from a mirror near you.
[2] Note: Mule packages should be installed into:
/usr/local/lib/xemacs/mule-packages/
- --
XEmacs Advocate | Do not try the patience of Wizards,
FreeBSD Devote | for they are subtle and quick to anger.
Perl Hacker | - Elric (Technomage) , Babylon 5.
Apache God | <mailto:rendhalver at xemacs.org> <GnuPG KeyID: AE51D190>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (FreeBSD)
Comment: Signed by Rendhalver, part of the XEmacs team.
iD8DBQE9oTDQkNWo665R0ZARArD+AJwNS1+DL5hChvktm2Q7EGXC1JkFXwCdEQs5
hc2u0yOsd7oAk96xAx91u1I=
=7NbA
-----END PGP SIGNATURE-----
10:43 a.m.
You don't need to CC xemacs-beta, I'm pretty sure it's on the
xemacs-announce distribution list.
>>>> "R" == rendhalver
<rendhalver(a)xemacs.org> writes:
R> Hello. The following XEmacs packages have just been released.
R> Notes there are 4 new packages in this release fortran-modes,
R> perl-modes, python-modes and ruby-modes they have been split
R> out of prog-modes so if you used any of those modes you will
R> have to install the new packages
This is going to cause a lot of pain and agony. Please monitor c.e.x
as well as xemacs-beta for the probable FAQ. (Note: people who simply
use M-x list-packages RET are _not_ going to see your announcement.)
Also, has this been set up properly for the netinstaller?
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
My nostalgia for Icon makes me forget about any of the bad things. I don't
have much nostalgia for Perl, so its faults I remember. Scott Gilbert c.l.py
11 a.m.
>>>> "Stephen" == Stephen J Turnbull
<stephen(a)xemacs.org> writes:
Stephen> You don't need to CC xemacs-beta, I'm pretty sure it's on the
Stephen> xemacs-announce distribution list.
i don't cc xemacs-beta
>>>> "R" == rendhalver
<rendhalver(a)xemacs.org> writes:
R> Hello. The following XEmacs packages have just been released.
R> Notes there are 4 new packages in this release fortran-modes,
R> perl-modes, python-modes and ruby-modes they have been split
R> out of prog-modes so if you used any of those modes you will
R> have to install the new packages
Stephen> This is going to cause a lot of pain and agony. Please monitor c.e.x
Stephen> as well as xemacs-beta for the probable FAQ. (Note: people who simply
Stephen> use M-x list-packages RET are _not_ going to see your announcement.)
oki cool will do
ville can you watch comp.emacs.xemacs as well so you can catch the
questions that popup while im not online
Stephen> Also, has this been set up properly for the netinstaller?
i believe so
i put the setup-packages.ini file in the windows directory and ran
sh ./makeini.sh
from there
steve informs me that this rebuilds the netinstaller files
--
XEmacs Advocate | Do not try the patience of Wizards,
FreeBSD Devote | for they are subtle and quick to anger.
Perl Hacker | - Elric (Technomage) , Babylon 5.
Apache God | <mailto:rendhalver at xemacs.org> <GnuPG KeyID: AE51D190>
11:34 a.m.
>>>> "R" == rendhalver
<rendhalver(a)xemacs.org> writes:
>>>> "Stephen" == Stephen J Turnbull
<stephen(a)xemacs.org> writes:
Stephen> You don't need to CC
xemacs-beta, I'm pretty sure it's on
Stephen> the xemacs-announce distribution list.
R> i don't cc xemacs-beta
Oops, sorry, there's a reply-to. Now did you add that, or did
mailman?
R> i believe so
Your word is good enough for me ;-)
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
My nostalgia for Icon makes me forget about any of the bad things. I don't
have much nostalgia for Perl, so its faults I remember. Scott Gilbert c.l.py
4:42 p.m.
>>>> "Stephen" == Stephen J Turnbull
<stephen(a)xemacs.org> writes:
>>>>> "R" == rendhalver <rendhalver(a)xemacs.org> writes:
>>>> "Stephen" == Stephen J Turnbull
<stephen(a)xemacs.org> writes:
Stephen> You don't need to CC
xemacs-beta, I'm pretty sure it's on
Stephen> the xemacs-announce distribution list.
R> i don't cc xemacs-beta
Stephen> Oops, sorry, there's a reply-to. Now did you add that, or did
Stephen> mailman?
looks like it got added by mailman
i just checked my sent email and its not there
R> i believe so
Stephen> Your word is good enough for me ;-)
cool
steve taught me well :)
--
XEmacs Advocate | Do not try the patience of Wizards,
FreeBSD Devote | for they are subtle and quick to anger.
Perl Hacker | - Elric (Technomage) , Babylon 5.
Apache God | <mailto:rendhalver at xemacs.org> <GnuPG KeyID: AE51D190>
5:36 p.m.
On Mon, 2002-10-07 at 13:00, Rendhalver [Peter Brown] wrote:
Stephen> This is going to cause a lot of pain and agony.
Please monitor c.e.x
Stephen> as well as xemacs-beta for the probable FAQ. (Note: people who simply
Stephen> use M-x list-packages RET are _not_ going to see your announcement.)
They're not going to upgrade any packages without first downloading the
new package index, right?
<rant>
I'd appreciate reports about anticipated problems with packages earlier
in the process, preferably before they're released. IIRC I've mailed
several times to xemacs-beta and at least 4 commit messages to
xemacs-patches about splitting prog-modes.
</rant>
ville can you watch comp.emacs.xemacs as well so you can catch the
questions that popup while im not online
Will try to.
--
\/ille Skyttä
ville.skytta at xemacs.org
2:32 a.m.
>>>> "Ville" == Ville Skytt <Ville>
writes:
Ville> On Mon, 2002-10-07 at 13:00, Rendhalver [Peter Brown] wrote:
Stephen> This is going to cause a lot of pain and agony. Please monitor c.e.x
Stephen> as well as xemacs-beta for the probable FAQ. (Note: people who simply
Stephen> use M-x list-packages RET are _not_ going to see your announcement.)
Ville> They're not going to upgrade any packages without first downloading the
Ville> new package index, right?
Ville> <rant>
Ville> I'd appreciate reports about anticipated problems with packages earlier
Ville> in the process, preferably before they're released. IIRC I've
mailed
Ville> several times to xemacs-beta and at least 4 commit messages to
Ville> xemacs-patches about splitting prog-modes.
Ville> </rant>
i will second your rant
please people we are new at this and if we do something that you dont
like please tell us and we can discuss it
> ville can you watch comp.emacs.xemacs as well so you can catch
the
> questions that popup while im not online
Ville> Will try to.
thanks dude
--
XEmacs Advocate | Do not try the patience of Wizards,
FreeBSD Devote | for they are subtle and quick to anger.
Perl Hacker | - Elric (Technomage) , Babylon 5.
Apache God | <mailto:rendhalver at xemacs.org> <GnuPG KeyID: AE51D190>
6:38 a.m.
>>>> "R" == rendhalver
<rendhalver(a)xemacs.org> writes:
Ville> <rant> I'd appreciate reports about anticipated problems
They weren't anticipated. Strictly speaking, it's your job to
anticipate this kind of problem; the rest of us will help if we can,
but realistically our primary attention is not going to be on
packages. You guys have been doing a good job[1], we're not going to
wrack our brains to project problems. :-/
Ville> with packages earlier in the process, preferably before
Ville> they're released. </rant>
R> i will second your rant
R> please people we are new at this and if we do something that
R> you dont like please tell us and we can discuss it
Well, in the future, consider bug reports like the ones from
Pre-Release testers whose cperl-mode, etc, stopped working a sign that
you should slow down on the release. _Any_ problem related to upgrade
or installation is a red flag. Backwards compatibility is extremely
hard to achieve, and harder to plan in advance, for things like this.
I just assumed that you would take some corrective action before
releasing the split packages. When you didn't, well, as you say
you're new at it, so all I could do was to suggest watching c.e.x.
Footnotes:
[1] In particular, I think general relations with package maintainers
is better even than when Steve Y was handling them.
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
My nostalgia for Icon makes me forget about any of the bad things. I don't
have much nostalgia for Perl, so its faults I remember. Scott Gilbert c.l.py
7:56 a.m.
>>>> "Stephen" == Stephen J Turnbull
<stephen(a)xemacs.org> writes:
>>>> "R" == rendhalver
<rendhalver(a)xemacs.org> writes:
Ville> <rant> I'd appreciate
reports about anticipated problems
Stephen> They weren't anticipated. Strictly speaking, it's your job to
Stephen> anticipate this kind of problem; the rest of us will help if we can,
Stephen> but realistically our primary attention is not going to be on
Stephen> packages. You guys have been doing a good job[1], we're not going to
Stephen> wrack our brains to project problems. :-/
well as a general rule i put a note in my announce email's about
major changes to the packages so people who are on announce will find
out if there is anything the need to do to make those new packages
work, if thats not enough we might need to hack the package index to
include those messages, i assume it would be possible to stick a
message at the top of the pui buffer ? we could then add functionality
to alow me to include a message about drastic changes to the packages
i WILL need some help doing this if thats what we want to do
Ville> with packages earlier in the process, preferably before
Ville> they're released. </rant>
R> i will second your rant
R> please people we are new at this and if we do something that
R> you dont like please tell us and we can discuss it
Stephen> Well, in the future, consider bug reports like the ones from
Stephen> Pre-Release testers whose cperl-mode, etc, stopped working a sign that
Stephen> you should slow down on the release. _Any_ problem related to upgrade
Stephen> or installation is a red flag. Backwards compatibility is extremely
Stephen> hard to achieve, and harder to plan in advance, for things like this.
Stephen> I just assumed that you would take some corrective action before
Stephen> releasing the split packages. When you didn't, well, as you say
Stephen> you're new at it, so all I could do was to suggest watching c.e.x.
Stephen> Footnotes:
Stephen> [1] In particular, I think general relations with package maintainers
Stephen> is better even than when Steve Y was handling them.
cool thaks but ville can probably take the credit for that :)
--
XEmacs Advocate | Do not try the patience of Wizards,
FreeBSD Devote | for they are subtle and quick to anger.
Perl Hacker | - Elric (Technomage) , Babylon 5.
Apache God | <mailto:rendhalver at xemacs.org> <GnuPG KeyID: AE51D190>
10:52 a.m.
>>>> "R" == rendhalver
<rendhalver(a)xemacs.org> writes:
R> well as a general rule i put a note in my announce email's
R> about major changes to the packages so people who are on
R> announce will find out if there is anything the need to do to
R> make those new packages work, if thats not enough we might need
R> to hack the package index to include those messages, i assume
R> it would be possible to stick a message at the top of the pui
R> buffer ?
Sure. But a lot of this could be handled automatically if we had a
better dependency database. That work could go hand in hand with the
new messages.
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
My nostalgia for Icon makes me forget about any of the bad things. I don't
have much nostalgia for Perl, so its faults I remember. Scott Gilbert c.l.py
12:10 p.m.
>>>> "Stephen" == Stephen J Turnbull
<stephen(a)xemacs.org> writes:
>>>> "R" == rendhalver
<rendhalver(a)xemacs.org> writes:
R> well as a general rule i put a note
in my announce email's
R> about major changes to the packages so people who are on
R> announce will find out if there is anything the need to do to
R> make those new packages work, if thats not enough we might need
R> to hack the package index to include those messages, i assume
R> it would be possible to stick a message at the top of the pui
R> buffer ?
Stephen> Sure. But a lot of this could be handled automatically if we had a
Stephen> better dependency database. That work could go hand in hand with the
Stephen> new messages.
ok what do we need to do to make the dependancy database better
what files does the current one live in ??
--
XEmacs Advocate | Do not try the patience of Wizards,
FreeBSD Devote | for they are subtle and quick to anger.
Perl Hacker | - Elric (Technomage) , Babylon 5.
Apache God | <mailto:rendhalver at xemacs.org> <GnuPG KeyID: AE51D190>
1:06 p.m.
>>>> "R" == rendhalver
<rendhalver(a)xemacs.org> writes:
R> ok what do we need to do to make the dependancy database better
Have one ...
R> what files does the current one live in ??
... ./*-packages/*/Makefile.
I'm _not_ joking.
If you do a search on "dependency" and "python" in XEmacs beta a
couple years back, somebody (I think it was Charles Waldman) had a
script which checked the dependencies. It wasn't close to perfect,
but at least it would give you some idea.
I think that the basic things to do would be (1) make World, so you
have all the auto-autoloads files, then (2) grep for provides and
requires to link up the dependencies, and (3) grep for all the
autoloaded functions, and link up those dependencies. Maybe you
should also annotate links for macros and libraries defining macros,
because those are the ones that are really troublesome (compiled code
can't contain calls to macros, because they are open-coded).
Once we've got the graph database, it should be relatively cheap
(time-wise) to update, package by package.
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
My nostalgia for Icon makes me forget about any of the bad things. I don't
have much nostalgia for Perl, so its faults I remember. Scott Gilbert c.l.py
8:11 a.m.
>>>> "Stephen" == Stephen J Turnbull
<stephen(a)xemacs.org> writes:
>>>> "R" == rendhalver
<rendhalver(a)xemacs.org> writes:
R> ok what do we need to do to make the
dependancy database better
Stephen> Have one ...
R> what files does the current one live in ??
Stephen> ... ./*-packages/*/Makefile.
Stephen> I'm _not_ joking.
Stephen> If you do a search on "dependency" and "python" in
XEmacs beta a
Stephen> couple years back, somebody (I think it was Charles Waldman) had a
Stephen> script which checked the dependencies. It wasn't close to perfect,
Stephen> but at least it would give you some idea.
ok will have a look
Stephen> I think that the basic things to do would be (1) make World, so you
Stephen> have all the auto-autoloads files, then (2) grep for provides and
Stephen> requires to link up the dependencies, and (3) grep for all the
Stephen> autoloaded functions, and link up those dependencies. Maybe you
Stephen> should also annotate links for macros and libraries defining macros,
Stephen> because those are the ones that are really troublesome (compiled code
Stephen> can't contain calls to macros, because they are open-coded).
sounds like a job for a perl script
Stephen> Once we've got the graph database, it should be relatively cheap
Stephen> (time-wise) to update, package by package.
we could stick the update mecanism in the build process, yes?
--
XEmacs Advocate | Do not try the patience of Wizards,
FreeBSD Devote | for they are subtle and quick to anger.
Perl Hacker | - Elric (Technomage) , Babylon 5.
Apache God | <mailto:rendhalver at xemacs.org> <GnuPG KeyID: AE51D190>
3:28 p.m.
|--==> "R" == rendhalver <rendhalver(a)xemacs.org> writes:
R> ok what do we need to do to make the dependancy database better
Invent one, because we ain't got one really.
R> what files does the current one live in ??
find *-packages/ -regex "^.*\(Makefile\|package-info\.in\)$" -print
--
|---<Steve Youngs>---------------<GnuPG KeyID: 10D5C9C5>---|
| XEmacs - It's not just an editor. |
| It's a way of life. |
|------------------------------------<youngs(a)xemacs.org>---|
8:13 a.m.
>>>> "SY" == Steve Youngs
<youngs(a)xemacs.org> writes:
SY> |--==> "R" == rendhalver <rendhalver(a)xemacs.org> writes:
R> ok what do we need to do to make the dependancy database better
SY> Invent one, because we ain't got one really.
any one have any preferences for what to store it in ?
R> what files does the current one live in ??
SY> find *-packages/ -regex "^.*\(Makefile\|package-info\.in\)$" -print
ok then
--
XEmacs Advocate | Do not try the patience of Wizards,
FreeBSD Devote | for they are subtle and quick to anger.
Perl Hacker | - Elric (Technomage) , Babylon 5.
Apache God | <mailto:rendhalver at xemacs.org> <GnuPG KeyID: AE51D190>
3:21 p.m.
|--==> "R" == rendhalver <rendhalver(a)xemacs.org> writes:
Ville> <rant> I'd appreciate reports about anticipated problems
This is probably my fault for not keeping a closer eye on things.
Ville, Rendhalver, please accept my apologies. But in hindsight, I
think the only thing I would have done differently would have been a
longer lead-in time from Pre-Release to Release.
Stephen> They weren't anticipated.
I haven't read c.e.x for quite some time, are many FAQs coming in
about this package split? I'd imagine not.
R> well as a general rule i put a note in my announce email's about
R> major changes to the packages so people who are on announce will find
R> out if there is anything the need to do to make those new packages
R> work, if thats not enough we might need to hack the package index to
R> include those messages, i assume it would be possible to stick a
R> message at the top of the pui buffer ? we could then add functionality
R> to alow me to include a message about drastic changes to the packages
Sounds like a pretty good idea.
R> i WILL need some help doing this if thats what we want to do
Yep. Nothing that copious amounts of beer and pizza couldn't fix. :-)
Ville> with packages earlier in the process, preferably before
Ville> they're released. </rant>
R> i will second your rant
Noted. I'll try to keep a closer eye on things.
R> please people we are new at this and if we do something that
R> you dont like please tell us and we can discuss it
And don't forget that I'm always available via email, ICQ, phone, or
in person to help you with any package issues.
--
|---<Steve Youngs>---------------<GnuPG KeyID: 10D5C9C5>---|
| XEmacs - It's not just an editor. |
| It's a way of life. |
|------------------------------------<youngs(a)xemacs.org>---|
8:20 a.m.
>>>> "SY" == Steve Youngs
<youngs(a)xemacs.org> writes:
SY> |--==> "R" == rendhalver <rendhalver(a)xemacs.org> writes:
Ville> <rant> I'd appreciate reports about anticipated problems
SY> This is probably my fault for not keeping a closer eye on things.
SY> Ville, Rendhalver, please accept my apologies. But in hindsight, I
SY> think the only thing I would have done differently would have been a
SY> longer lead-in time from Pre-Release to Release.
its cool dude
on this note i am going to finish my increnental release script
and then we can do individual releases for packages instead of waiting
till all the packages in pre-release are stable enough
on a related note
whats the status on the bug-report database ??
having that up and running will be helpful for package testing
Stephen> They weren't anticipated.
SY> I haven't read c.e.x for quite some time, are many FAQs coming in
SY> about this package split? I'd imagine not.
R> well as a general rule i put a note in my announce email's about
R> major changes to the packages so people who are on announce will find
R> out if there is anything the need to do to make those new packages
R> work, if thats not enough we might need to hack the package index to
R> include those messages, i assume it would be possible to stick a
R> message at the top of the pui buffer ? we could then add functionality
R> to alow me to include a message about drastic changes to the packages
SY> Sounds like a pretty good idea.
R> i WILL need some help doing this if thats what we want to do
SY> Yep. Nothing that copious amounts of beer and pizza couldn't fix. :-)
i will provide the net access :)
Ville> with packages earlier in the process, preferably before
Ville> they're released. </rant>
R> i will second your rant
SY> Noted. I'll try to keep a closer eye on things.
thanks dude
R> please people we are new at this and if we do something that
R> you dont like please tell us and we can discuss it
SY> And don't forget that I'm always available via email, ICQ, phone, or
SY> in person to help you with any package issues.
i know dude thanks
--
XEmacs Advocate | Do not try the patience of Wizards,
FreeBSD Devote | for they are subtle and quick to anger.
Perl Hacker | - Elric (Technomage) , Babylon 5.
Apache God | <mailto:rendhalver at xemacs.org> <GnuPG KeyID: AE51D190>
2:47 p.m.
|--==> "SJT" == Stephen J Turnbull <stephen(a)xemacs.org> writes:
SJT> In particular, I think general relations with package maintainers
SJT> is better even than when Steve Y was handling them.
Crushed ego. :-(
Hey Rendhalver, Ville! I don't remember saying that you guys were
allowed to be better than me. :-)
Congrats guys! You're both doing a terrific job!
--
|---<Steve Youngs>---------------<GnuPG KeyID: 10D5C9C5>---|
| XEmacs - It's not just an editor. |
| It's a way of life. |
|------------------------------------<youngs(a)xemacs.org>---|
6:51 p.m.
>>>> "SY" == Steve Youngs
<youngs(a)xemacs.org> writes:
SY> |--==> "SJT" == Stephen J Turnbull <stephen(a)xemacs.org>
writes:
SJT> In particular, I think general relations with package maintainers
SJT> is better even than when Steve Y was handling them.
SY> Crushed ego. :-(
SY> Hey Rendhalver, Ville! I don't remember saying that you guys were
SY> allowed to be better than me. :-)
it could be cause there are two of us and we have rigidly defined
areas of doubt and uncertanty
SY> Congrats guys! You're both doing a terrific job!
thanks dude :)
--
XEmacs Advocate | Do not try the patience of Wizards,
FreeBSD Devote | for they are subtle and quick to anger.
Perl Hacker | - Elric (Technomage) , Babylon 5.
Apache God | <mailto:rendhalver at xemacs.org> <GnuPG KeyID: AE51D190>
7:08 p.m.
On Mon, 2002-10-21 at 20:51, Rendhalver [Peter Brown] wrote:
SY> Hey Rendhalver, Ville! I don't remember saying that
you guys were
SY> allowed to be better than me. :-)
it could be cause there are two of us and we have rigidly defined
areas of doubt and uncertanty
LoL! Well put!
SY> Congrats guys! You're both doing a terrific job!
thanks dude :)
Yeah, thanks!
--
\/ille Skyttä
scop at xemacs.org
7:38 p.m.
>>>> "Ville" == Ville Skytt <Ville>
writes:
Ville> On Mon, 2002-10-21 at 20:51, Rendhalver [Peter Brown] wrote:
SY> Hey Rendhalver, Ville! I don't remember saying that you guys were
SY> allowed to be better than me. :-)
>
> it could be cause there are two of us and we have rigidly defined
> areas of doubt and uncertanty
Ville> LoL! Well put!
dont thank me thats a Douglas Adams quote from Hitch Hikers Guide to
the Galaxy
--
XEmacs Advocate | Do not try the patience of Wizards,
FreeBSD Devote | for they are subtle and quick to anger.
Perl Hacker | - Elric (Technomage) , Babylon 5.
Apache God | <mailto:rendhalver at xemacs.org> <GnuPG KeyID: AE51D190>
12:42 a.m.
|--==> "R" == rendhalver <rendhalver(a)xemacs.org> writes:
R> dont thank me thats a Douglas Adams quote from Hitch Hikers Guide
R> to the Galaxy
A trilogy in 4 parts.
--
|---<Steve Youngs>---------------<GnuPG KeyID: 10D5C9C5>---|
| XEmacs - It's not just an editor. |
| It's a way of life. |
|------------------------------------<youngs(a)xemacs.org>---|
1:49 p.m.
"Stephen J. Turnbull" <stephen(a)xemacs.org> writes:
You don't need to CC xemacs-beta, I'm pretty sure it's on
the
xemacs-announce distribution list.
>>>>> "R" == rendhalver <rendhalver(a)xemacs.org> writes:
R> Hello. The following XEmacs packages have just been released.
R> Notes there are 4 new packages in this release fortran-modes,
R> perl-modes, python-modes and ruby-modes they have been split
R> out of prog-modes so if you used any of those modes you will
R> have to install the new packages
This is going to cause a lot of pain and agony. Please monitor c.e.x
as well as xemacs-beta for the probable FAQ. (Note: people who simply
use M-x list-packages RET are _not_ going to see your announcement.)
Maybe prog-modes can REQUIRE the 4 new packages for some time?
5:46 p.m.
Ville Skyttä <ville.skytta(a)iki.fi> writes:
On Mon, 2002-10-07 at 15:49, Simon Josefsson wrote:
> Maybe prog-modes can REQUIRE the 4 new packages for some time?
I don't think will help too much since pui won't enforce the
requirements anyway unless explicitly told so, right?
Ouch, in that case never mind. It would be a good thing to fix that
pui bug though, as I know at least mail-lib and sieve won't work
properly if the REQUIREd packages aren't installed.
6:25 p.m.
On Mon, 2002-10-07 at 19:46, Simon Josefsson wrote:
>> Maybe prog-modes can REQUIRE the 4 new packages for some
time?
>
> I don't think will help too much since pui won't enforce the
> requirements anyway unless explicitly told so, right?
Ouch, in that case never mind. It would be a good thing to fix that
pui bug though, as I know at least mail-lib and sieve won't work
properly if the REQUIREd packages aren't installed.
Yes, there are lots of packages that won't work if let's say xemacs-base
isn't installed. But AFAIK, the current REQUIREs are build time
requirements, and the lack of enforcement is more like a missing
functionality (in both the package system in general and pui) rather
than a bug. But I agree that it should be fixed :)
--
\/ille Skyttä
ville.skytta at iki.fi
2:35 p.m.
|--==> "VS" == Ville Skytt <Ville> writes:
VS> Yes, there are lots of packages that won't work if let's say
VS> xemacs-base isn't installed. But AFAIK, the current REQUIREs
VS> are build time requirements, and the lack of enforcement is more
VS> like a missing functionality (in both the package system in
VS> general and pui) rather than a bug. But I agree that it should
VS> be fixed :)
If you want to do this, Ville, great (I've been meaning to for a
while). Just two things: only for 21.5 (unless Steve T wants it for
21.4); and make it customisable so people can switch it off if they
want to. I don't mind if it defaults to being on.
--
|---<Steve Youngs>---------------<GnuPG KeyID: 10D5C9C5>---|
| XEmacs - It's not just an editor. |
| It's a way of life. |
|------------------------------------<youngs(a)xemacs.org>---|
10:15 a.m.
On Wed, 2002-10-16 at 16:35, Steve Youngs wrote:
VS> Yes, there are lots of packages that won't work if
let's say
VS> xemacs-base isn't installed. But AFAIK, the current REQUIREs
VS> are build time requirements, and the lack of enforcement is more
VS> like a missing functionality (in both the package system in
VS> general and pui) rather than a bug. But I agree that it should
VS> be fixed :)
If you want to do this, Ville, great (I've been meaning to for a
while). Just two things: only for 21.5 (unless Steve T wants it for
21.4); and make it customisable so people can switch it off if they
want to. I don't mind if it defaults to being on.
Go ahead Steve if you find time for that, I lack the necessary time and
knowledge:(
--
\/ille Skyttä
scop at xemacs.org
8066
days inactive
8082
days old
27 comments
7 participants
participants (7)
-
Rendhalver [Peter Brown] -
Simon Josefsson -
Stephen J. Turnbull -
Steve Youngs -
Ville Skyttä -
Ville Skyttä -
Ville Skyttä