[Failure] The Package Smoketest
16 years, 9 months
Stephen J. Turnbull
C'mon guys, we need to do something about this. I think that the
people who are working to provide more advanced structures in Emacs
Lisp should be supported, and we've promised to do that in the stable
line and for some months after a new release gets promoted, which
realistically is a year away (at minimum).
Even though the monkey-patch hack I suggested is nowhere near as
general as I thought (I think it was Mike Kupfer who made me
understand that, thanks, Mike!), I still like it better (ie, think it
is as clean and more generalizable) than having to patch stable
versions to keep up with advances in packages' technology. But I'll
go with whatever Vin and Norbert think is best.
The alternatives on the table are
- My monkey-patch, which installs 21.5's autoloads builder into 21.4
during the package build for packages which need it, and will work
for all versions of 21.4 AFAIK
- Mike's patch to 21.4, which only works for some future version of
21.4
- both, so that we can support building the packages with earlier
versions of XEmacs, but have the relevant technology in 21.4 for
possible future use
The XEmacs Package Smoketest writes:
> These are the results from the XEmacs Package Smoketest:
>
> 21.5-nomule -- ok
> 21.5-mule -- ok
> 21.4-mule -- errors
> 21.4-nomule -- errors
>
> For details check "http://labb.contactor.se/~matsl/smoketest/logs"
> --
> This is generated by the XEmacs Package Smoketest.
>
>
> _______________________________________________
> XEmacs-Buildreports mailing list
> XEmacs-Buildreports(a)xemacs.org
> http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-buildreports
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
Re: MacOS?
16 years, 9 months
Rodney Sparapani
Didier Verna wrote:
> skip(a)pobox.com wrote:
>
>> dvl> To be precise, I think XEmacs should consider:
>> dvl> ~/Library/Application Support/XEmacs as the alternative to
>> dvl> ~/.xemacs and also /Library/Application Support/XEmacs as the
>> dvl> alternative to /usr/local/share/xemacs.
>>
>
> I believe these options already exist (although I never managed to
> understand the --with-{early/late/whatever}-package mess). To me, the
> question is more about the default values, which should conform to
> whatever "platform standards" exist.
>
> IMHO, the native carbon port should use the paths I mentioned because it
> is the Mac Way. However, I still believe that an XEmacs compiled for X11
> on MacOS X is more Unix oriented and should stick to
> /usr/local/share/xemacs.
>
Sorry for stripping email addresses, but, for me, tbird has problems
with x-posting between newsgroups and email (to my chagrin, often either
nothing gets sent or blank messages get sent).
My 2 cents. I think we have already discussed this, but I'm too lazy to
search for it right now. I believe the consensus was that we should
have both. Let me elaborate it a little more since I don't think it was
spelled out clearly.
So, suppose we are talking about ~/.xemacs and ~/Library/Application
Support/XEmacs. One of these is the real directory, say ~/.xemacs, and
the other is a symlink to it. This has the advantage that ~/.xemacs
will not show up under the Finder with the default settings (and the
non-default settings have other annoyances that I'm not going to get
into), but ~/Library/Application\ Support/XEmacs does which is a nice
convenience. Similar arguments apply for /usr/local/share/xemacs (or is
it /usr/local/lib/xemacs?) and /Library/Application Support/XEmacs.
This type of symlinking between Mac OS X and Unix directories is quite a
common customization for those of us who are really Unix users, but have
found a happy home on Mac OS X.
Rodney
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
Re: sync auctex 11.84: good news.
16 years, 9 months
Uwe Brauer
>>>>> "Stephen" == Stephen J Turnbull <stephen(a)xemacs.org> writes:
> Do you mean *installation* or *build*? It's not a problem to
> require LaTeX for compiling any AUCTeX-specific .tex files etc,
> which is part of the *build*. But if you need to build in the
> location of TeX binaries or libraries outside of the AUCTeX
> package, it's a showwtopper IMO.
Well I meant *build*, when I am running make.
> I don't think this is a problem in principle (the JDE support is much
> more horrifying :-), but I have some questions about the patch.
> EXTRA_AUTOLOADS_DEPENDENCIES is independent of
> GENERATED_ELC_DEPENDENCIES, right? If so, you shouldnt substitute the
> first for the second. Both should be present.
> Also, where are EXTRA_AUTOLOADS_DEPENDENCIES and
> GENERATED_ELC_DEPENDENCIES set? This should be be documented.
Ok I discuss this with Mats.
>>
>> cvs add newfile
>> cvs remove oldfile
> This is much preferable. Although CVS won't preserve history, when we
> move to a more capable version control system the migration script
> will see the remove/add pair and do the right thing. You can automate
> this with a shell function
> function cvs-move () {
> oldfile=$1
> newfile=$2
> mkdir -p `dirname $newfile`
> mv $oldfile $newfile
> cvs remove $oldfile
> cvs add $newfile
> cvs commit -m "Move $oldfile to $newfile." $oldfile $newfile
> }
Oops that is (t)csh?
And what about
cvs remove:
find . \( -name '*' -o -name '.[a-z]*' \) -print -exec cvs remove {} \;
Or something of the sort and then after the copy the same for add
> The reason for doing the actual commit in this way is to help
> migration scripts, which often look at the log message to identify
> changesets that go together. It is also a very good idea to do all
> the mv'ing in one session, based on moving the current versions of the
> files, and tag the workspace both before and after. Eg,
Oops I would have preferred David suggestion.
> cvs tag -c pre_Great_AUCTeX_11_84_Move
> cvs-move A B/C
> cvs-move D B/E
> cvs-move F G/H
> .
> .
> .
> cvs tag -c post_Great_AUCTeX_11_84_Move
> Then update to the new versions and add new files
> rsync -av AUCTeX-11.84 AUCTeX-XEmacs
> cvs add any new files not part of the move
> cvs commit -m "Sync to AUCTeX 11.84."
> cvs tag -c post_AUCTeX_11_84_Sync
Again oops looks again complicated.
Uwe
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
feedmail.el - Fix typo
16 years, 9 months
Dr. Volker Zell
2008-02-15 Dr. Volker Zell <Dr.Volker.Zell(a)oracle.com>
* feedmail.el (feedmail-fill-to-cc-function): Fixes typo which
prevents BCC adresses to be processed.
--- feedmail.el.orig 2007-03-06 14:47:45.001000000 +0100
+++ feedmail.el 2007-12-30 16:31:49.014870400 +0100
＠＠ -3169,7 +3169,7 ＠＠
(feedmail-say-debug ">in-> feedmail-fill-to-cc-function")
(let ((case-fold-search t)
(headers (mapconcat #'regexp-quote
- '("FROM:" "REPLY-TO:" "TO:" "CC:" "BCC"
+ '("FROM:" "REPLY-TO:" "TO:" "CC:" "BCC:"
"RESENT-TO:" "RESENT-CC:" "RESENT-BCC:")
"\\|"))
this-line
Ciao
Volker
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
Dired problem reports
16 years, 9 months
Ville Skyttä
Hello,
I'm forwarding a couple of dired related bug reports here, perhaps someone who
uses it could have a look and/or comment?
https://bugzilla.redhat.com/show_bug.cgi?id=432101
1) After using XEmacs for a while (C, C++ modes), auto-save triggers
minibuffer messages about "dired alist being mangled, and then the cursor
jumps around in the buffer and some newlines are inserted in the middle of
existing lines". IIRC something related to the alist mangling thing was
semi-recently discussed here, anyone remember more details?
2) The new pre-release dired package reportedly causes the following issues:
"
- Hitting enter while on top of a directory did not descend into the director
(reported that buffer 'dirname' was read-only)
- hitting 's' to resort by date would actually change to a different dired
buffer, etc
"
I don't think I've ever seen either issue.
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
Re: (temp-directory) needs some (regexp-quote) love in gnuserv.el
16 years, 9 months
Skip Montanaro
>> Error: Required issue property module not supplied
Stephen> Couldn't you select one? Or did you just ignore it?
A quick glance around the screen didn't reveal anything. What is an "issue
property module"? The error message read as if there was a missing Python
module, not that I forgot to fill out some required field.
Stephen> If you think it's entirely unobvious what "module" means, can
Stephen> you suggest a term that would be more meaningful to you? I've
Stephen> thought about "component".
A better user interface would also highlight the title of the field which
wasn't filled in (maybe render it in red text). Also, now that I go back
and study the submission form more closely, I see that the explanation of
the title highlighting is at the bottom of the form. Shouldn't that be at
the top?
Skip
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
ACL problem under Mac OS X 10.5.2
16 years, 9 months
robert delius royar
This is just a brief explanation. I would first like to know if anyone
is seeing the same problem as I under Mac OS X 10.5.2 using
macports-installed mercurial (0.9.5).
The two problems I see are
1)
hg 0.9.5 (installed from macports) will add ACLs to all the files and
prevent changes by group members if the directory which contains the
source does not have group write access. The ACLs override the owner's
write access. Note, that hg status (which should change nothing) also
causes the change; isn't that a violation of some UI dogma?
The solution is to make the directory containing the tree group
writable which does not lead to hg adding ACLs. That is not a real
problem in OS X 10.4 because the user's group is the same as the user.
But in 10.5 it can be a problem. My default group is staff. On a
multi-user machine, there could be other members of the group.
2)
hg pull causes any differences to be processed through Xcode's
FileMerge program which is an interactive GUI and not automatic--new
development with 10.5.2
--
Dr. Robert Delius Royar Associate Professor of English
Morehead State University Morehead, Kentucky
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
(temp-directory) needs some (regexp-quote) love in gnuserv.el
16 years, 9 months
Skip Montanaro
I tried posting this ticket through the new tracker but was rebuffed with
this error:
Error: Required issue property module not supplied
so here's the bug report.
Skip
At the moment I have a very wacky (temp-directory):
/var/folders/5q/5qTPn6xq2RaWqk+1Ytw3-U+++TI/-Tmp-/
This is on a Mac OS X 10.5.1 system using some XEmacs.app wrapper I found on
the net.
All the plusses wreak havoc with (string-match). In particular,
gnuserv-temp-file-regexp needs a hug from (regexp-quote):
diff -u gnuserv.el.~1~ gnuserv.el
--- gnuserv.el.~1~ 2004-12-14 03:56:48.000000000 -0600
+++ gnuserv.el 2008-02-14 10:43:10.000000000 -0600
＠＠ -214,7 +214,7 ＠＠
:group 'gnuserv)
(defcustom gnuserv-temp-file-regexp
- (concat "^" (temp-directory) "/Re\\|/draft$")
+ (concat "^" (regexp-quote (temp-directory)) "/Re\\|/draft$")
"*Regexp which should match filenames of temporary files deleted
and reused by the programs that invoke the Emacs server."
:type 'regexp
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
XEmacs Packages have been pre-released (2008-02-11-10)
16 years, 9 months
Norbert Koch
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hey there everyone.
I have just added the following packages to the 'Pre-Releases'
directory:
New Packages in Pre-Release:
===========================
prog-modes-2.16-pkg.tar.gz upstream version: none
xetla-1.02-pkg.tar.gz upstream version: steve(a)eicq.org--2005/xetla--main--1.1--version-0
Previously Announced Packages Still in Pre-Release:
==================================================
apel-1.33-pkg.tar.gz upstream version: 10.6
auctex-1.48-pkg.tar.gz upstream version: 11.55
calendar-1.34-pkg.tar.gz upstream version: none
dired-1.19-pkg.tar.gz upstream version: 7.16
easypg-1.02-pkg.tar.gz upstream version: 0.0.16
ediff-1.74-pkg.tar.gz upstream version: 2.75
edit-utils-2.38-pkg.tar.gz upstream version: none
edt-1.14-pkg.tar.gz upstream version: none
efs-1.34-pkg.tar.gz upstream version: 1.24
elib-1.13-pkg.tar.gz upstream version: 1.0
eshell-1.12-pkg.tar.gz upstream version: 2.4.1
eudc-1.40-pkg.tar.gz upstream version: 1.32
general-docs-1.05-pkg.tar.gz upstream version: none
gnus-1.92-pkg.tar.gz upstream version: 5.10.8
guided-tour-0.51-pkg.tar.gz upstream version: none
hm--html-menus-1.24-pkg.tar.gz upstream version: 5.9
hyperbole-1.17-pkg.tar.gz upstream version: 5.0
igrep-1.16-pkg.tar.gz upstream version: 2.111
latin-euro-standards-1.08-pkg.ta upstream version: 1.08
leim-1.27-pkg.tar.gz upstream version: none
locale-1.27-pkg.tar.gz upstream version: none
mh-e-1.31-pkg.tar.gz upstream version: 7.4.2
mule-base-1.52-pkg.tar.gz upstream version: none
mule-ucs-1.16-pkg.tar.gz upstream version: 0.84
net-utils-1.54-pkg.tar.gz upstream version: N/A
oo-browser-1.05-pkg.tar.gz upstream version: 4.08
os-utils-1.40-pkg.tar.gz upstream version: none
pcl-cvs-1.68-pkg.tar.gz upstream version: R-2_9_9
perl-modes-1.11-pkg.tar.gz upstream version: none
psgml-1.45-pkg.tar.gz upstream version: 1.3.2
python-modes-1.10-pkg.tar.gz upstream version: 426
scheme-1.17-pkg.tar.gz upstream version: none
sh-script-1.23-pkg.tar.gz upstream version: 2.0f
skk-1.24-pkg.tar.gz upstream version: 10.62a
text-modes-1.95-pkg.tar.gz upstream version: none
tramp-1.40-pkg.tar.gz upstream version: 2.0.56
vc-1.45-pkg.tar.gz upstream version: none
viper-1.61-pkg.tar.gz upstream version: 3.09
vm-7.26-pkg.tar.gz upstream version: 7.19
w3-1.34-pkg.tar.gz upstream version: 4.0pre47
xemacs-base-2.15-pkg.tar.gz upstream version: none
xemacs-devel-1.77-pkg.tar.gz upstream version: none
Detailed Changes:
================
- ------- ChangeLog Entries from xemacs-packages/prog-modes/ChangeLog -------
2008-02-11 Norbert Koch <viteno(a)xemacs.org>
* Makefile (VERSION): XEmacs package 2.16 released.
2008-02-10 Ville Skyttä <scop(a)xemacs.org>
* rpm-spec-mode.el (rpm-increase-release-tag): Increase last digit
of the release tag instead of the first (eg. Y of X.Y, not X), with
some heuristics in deciding what the desired "last" digit is.
(rpm-spec-mode-version): Set to 0.12.1x.
- ------- ChangeLog Entries from xemacs-packages/xetla/ChangeLog -------
2008-02-11 Norbert Koch <viteno(a)xemacs.org>
* Makefile (VERSION): XEmacs package 1.02 released.
2007-11-30 Michael Sperber <mike(a)xemacs.org>
* xetla-browse.el (jde-tree-widget): tree-widget.el has been
renamed to jde-tree-widget.el.
Installing These:
================
Manually:
- --------
1) Download the packages that you want to install from:
/ftp.xemacs.org:/pub/xemacs/beta/experimental/packages/
2) Unpack them to: [1]
/usr/local/lib/xemacs/xemacs-packages/
3) Re-start XEmacs.
Using XEmacs Package Tools (XEmacs 21.[245].x):
- ----------------------------------------------
1) Tools -> Packages -> Add Download Site -> Pre-Releases
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 -> Pre-Releases
2) Options -> Manage Packages -> List and Install
3 - 5) As per XEmacs 21.[245].x.
norbert - XEmacs Package Release Manager.
Footnotes:
[1] Note: Mule packages should be installed into:
/usr/local/lib/xemacs/mule-packages/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
iD8DBQFHsFJVgu3ywdHdhM0RApDtAJ0SEtvtkZ6hPKxy+60mCVoaqNc0eQCfXa7d
K+3iJmzrcpLcBrU3tjUdsq0=
=aDek
-----END PGP SIGNATURE-----
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
WTF: (string-match "\\S-" "\n") => t
16 years, 9 months
Stephen J. Turnbull
And on the contrary, (string-match "\\s-" "\n") => nil
That seems backwards to me.
More speculatively, I wonder if we shouldn't special-case ?\n and have
*both* of those evaluate to nil. The use case is that in a line-
oriented parser for shell scripts (specifically version.sh), I'd like
to write
(looking-at #r"\(\w+\)=\(\S-\(.*\S-\)?\)?\s-*$")
The problem is that if the value assigned is the empty string (not
whitespace), the first \S- sucks up the newline, and then the rest of
the next line!
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta