Hi 1034744 team: please see #1024695 as a bug that could get fixed at
the same time...
⛔ Warning (comp): debian-el.el:90:26: Warning: reference to free variable
‘dired-mode-map’
⛔ Warning (comp): debian-el.el:95:35: Warning: docstring has wrong usage of
unescaped single quotes (use \= or different quoting)
⛔ Warning (comp): debian-ispell.el:420:16: Warning: reference to free variab
X-Debbugs-Cc: Peter S. Galbraith
Package: elpa-debian-el
Version: 37.10
Severity: minor
Warning (comp): deb-view.el:303:28: Warning: reference to free variable
‘dired-mode-map’, and:
Use ‘with-current-buffer’ rather than save-excursion+set-buffer
assignment to free variable ‘deb-view-find-minor-
reopen 1030394
retitle 1030394 1.20 not cleaned up
thanks
Wait, on my other computer there is the exact same problem:
$ ls -log /usr/share/emacs/site-lisp/elpa/
drwxr-xr-x 2 4096 2018-08-07 circe-2.6
drwxr-xr-x 2 4096 2022-05-26 csv-mode-1.20
drwxr-xr-x 2 4096 01-25 12:31 csv-mode-1.21
drwxr-xr
Package: elpa-csv-mode
Version: 1.21-1
Severity: minor
We note the ugly double slashes. Yes, cosmetic problem only.
ls -og ...
/usr/share/emacs/site-lisp/elpa/csv-mode-1.21:
-rw-r--r-- 132 01-18 18:15 Install.log.gz
lrwxrwxrwx 172 01-18 18:15 csv-mode-autoloads.el ->
/usr/share/emacs/site
Package: elpa-csv-mode
Version: 1.21-1
When I open a .csv file, I see:
Warning (comp): Cannot look-up eln file as no source file was found for
/usr/share/emacs/site-lisp/elpa/csv-mode-1.20/csv-mode.elc Disable showing
Disable logging
That's odd.
*needs .el, so perhaps ship it.
*1.20 vs. 1.21 p
I have no idea.
Please reassign it.
Thanks.
---
On 2023-01-29 05:17, Sven Joachim wrote:
On 2023-01-28 07:24 -0400, David Bremner wrote:
Dan Jacobson writes:
Package: dh-elpa-helper
Version: 2.0.16
Severity: minor
I saw this:
Warning (comp): Cannot look-up eln file as no source file was
Package: dh-elpa-helper
Version: 2.0.16
Severity: minor
I saw this:
Warning (comp): Cannot look-up eln file as no source file was found for
/usr/share/emacs/site-lisp/elpa/csv-mode-1.20/csv-mode.elc
Package: elpa-debian-el
Version: 37.10
File: /usr/share/emacs/site-lisp/elpa-src/debian-el-37/debian-el.el
I see in the *Warning* bfuffer
Warning (comp): debian-el.el:90:26: Warning: reference to free variable
‘dired-mode-map’
# apt-show-versions |grep emacs
emacs:all/unstable 1:28.2+1-6 uptodate
Well on a vanilla system I still get:
# aptitude -s purge debian-el
The following packages will be REMOVED:
debian-el{p} elpa-debian-el{u}
aptitude show says:
Package: elpa-debian-el
Version: 37.10
State: installed
Automatically installed: yes
Package: debian-el
Version: 37.10
State: installe
Package: debian-el
Version: 37.10
When I try to remove this dummy package, aptitude says
The following packages will be REMOVED:
debian-el{p} elpa-debian-el{pu} (D: debian-el)
But other dummy packages don't trigger this in aptitude.
Package: dh-elpa-helper
Please make
https://elpa.gnu.org/packages/tramp.html
into a elpa-tramp package.
Package: dh-elpa
Severity: wishlist
There is no documentation on the proper way to request that
there also be a "elpa-tramp" package available among the 290 elpa-
packages on Debian.
Yes, one can find it in M-x list-packages, but I wish someone would put
it into apt too. Thanks.
> "DB" == David Bremner writes:
DB> To be honest I'm not sure supporting local variables in
DB> /etc/apt/sources.list is a good idea...
OK, but at least have emacs say "Local variables detected in important
file. Not allowed!"
Else when they silently don't work people will think it is a bug.
Package: elpa-debian-el
Version: 37.8
File: /usr/share/emacs/site-lisp/elpa-src/debian-el-37/apt-sources.el
Don't disable local variables set in files please.
# cat /etc/apt/sources.list.d/jidanni.list
deb http://opensource.nchc.org.tw/debian/ experimental main contrib non-free
deb http://opensou
Well I'll send a new bug if I run into one if I use that package again.
Seems like there is a spam a day on this list, at least reading from
gmane. Can't anything be done?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
>> $ emacs-snapshot
>> Warning: Missing charsets in String to FontSet conversion
>> How might I take steps to alleviate this?
RF> Try installing xfs.
I did and unix/:7100 is on the FontPath according to
/var/log/Xorg.0.log. No help.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject o
$ emacs-snapshot
Warning: Missing charsets in String to FontSet conversion
How might I take steps to alleviate this?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
The emacs I start from xdm's .xsession is one line shorter than an
emacs I start later from a command line. Seen now with emacs-snapshot
but earlier too. Must be a complex interaction between the fonts and
resolutions I choose in .emacs and ~/.Xresources and timings. Should I
just add more and more
If I installed emacs-snapshot, then should I purge gnus, as
emacs-snapshot's gnus is newer than gnus package's gnus?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Today I discovered a Help page titled:
Characters in the charset chinese-sisheng.
Finally single width accented U+01CE etc.!
(Who can stand mixed single/doublewidth Pinyin?)
OK, how does one correct emacs-version 21.4.1 to instead use them when
encountering U+01CE etc. instead of doublewidth?
Sid is out of phase?!?:
$ egrep 'Depen|Packa|Vers'
Package: emacs21-bin-common
Version: 21.4a-3
Depends: emacs21-common (= 21.4a-3), libc6 (>= 2.3.5-1)
Package: emacs21-common
Version: 21.4a-3.1
Depends: emacsen-common (>= 1.4.10), dpkg (>= 1.9.0)
Package: emacs21-el
Version: 21.4a-3.1
Depends: ema
Odd, the bugs I report upstream seem to never get fixed. Ah:
$ zcat /usr/share/doc/emacs21-common/changelog.gz|head
2003-03-18 Francesco PotortAl <[EMAIL PROTECTED]>
* Version 21.3 released.
Where's the "excitement of sid"? Or maybe it's upstream that are slow
to release stuff. (I don't
'((font . "-*-fixed-medium-r-normal-*-16-*-*-*-*-*-fontset-standard"))
looks great but now I am older and want bigger fonts. One cannot just
s/16/18/ it turns out, perish the thought of just hitting control-plus
like in some browsers.
I somehow determined what looked great was
-sony-fixed-medium-
I've just got to get this emacs-snapshot package that I've been
hearing about. Imagine me sending rms bug reports for long fixed bugs.
But wait,
# apt-cache show emacs-snapshot
Nothing.
# apt-get install emacs-snapshot
Package emacs-snapshot is not available, but is referred to by another package.
Fellas, how good should C-h H (view-hello-file) be looking these days?
The only lines that still have blank boxes are
Amharic
Arabic
Tibetan
Tigrigna
Just for a test of JISX0212
Oops, works when I do
$ DEB_MULEUCS_UNICODE=on emacs -f gnus
(I wish there would be more dynamic ways to load that bulky thing only
when needed. So far I use it only once a year and always forget that
I turned it off.) Versions emacs21 21.3+1-5 gnus 5.10.6-6
In gnus I should have no problem seeing
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: base64
messages, no?
All I see is the \346\262\222 stuff for the non ascii parts.
I did
(and
(eq window-system 'x)
(progn
(defun browse-url-firefox (url &optional ignored)
"Browse URL using a new tab in firefox."
(interactive "sURL: ")
(let ((process-environment (cons "DISPLAY=:0.0" process-environment)))
(call-process "firefox" nil 0 nil
Gentlemen, these didn't open a new window
(setq browse-url-generic-program "mozilla")
(setq browse-url-netscape-program "mozilla")
but they do with firefox. Tips?
>>>>> "M" == Matt Hodges <[EMAIL PROTECTED]> writes:
>>>>> Dan Jacobson writes:
>> Just for fun, I started two emacs -q's, and then ran top(1).
>> In one I typed M-x apt-utils-show-package RET
>> PID USER PR NI VIRT
Tried to save file as gb2312 instead of big5
$ emacs -q --no-site-file /usr/lib/cce/pinyin.map
C-x f g b C-x C-w / t m p /
g C-x C-v
But couldn't see it in gb2312... it just stays in big5,
maybe due to
$ env|grep -i tw
LANG=zh_TW.Big5
LC_CTYPE=zh_TW.Big5
or paltry instructions in
(describe-fun
Annoyed by the word "Success" in times of failure,
young Dan writes the pros, who respond:
> "RMS" == Richard Stallman <[EMAIL PROTECTED]> writes:
DJ> emacsclient: can't stat /tmp/esrv1000-jidanni.org: Success
RMS> I can't see anything in the source code of emacsclient to produce
RMS> th
Ben> Sure. I am doing so now. Internet. Internet. Internet. I use
Ben> all of the above modes, by the way.
>> I see if I turn off Abbrev mode, all is ok. Turning off Flyspell mode
>> was not enough.
Ben> Do you have an abbrev for "internet"? Do you have something
Ben> funny set up to expan
I find with these modes
7 lines matching "mode.*:" in buffer *Help*.
4:Message mode:
6:Like Text Mode but with these additional commands:
157:Mml minor mode (indicator MML):
201:Mc-Write minor mode (indicator MC-w):
215:Flyspell minor mode (indicator Fly):
244:Abbrev min
Here is a problem that has bugged me for a long time in gnus. I can
work around it, with ^u g, but it is mainly a curiosity. When I get a
message from a certain person, and hit g to read it, I get
mail-decode-encoded-word-region: Args out of range: 483, 494
etc.
http://jidanni.org/comp/gnus_el.
>>>>> "R" == Reiner Steib <[EMAIL PROTECTED]> writes:
R> On Thu, Nov 27 2003, Dan Jacobson wrote:
>> Info node:
>> If you set `gnus-keep-backlog' to a number N, Gnus will store
>> This variable is `nil' by default.
R> This has
Matt, thanks for sending me the new version of apt-utils.el, but guess
what, last time I had an apt-utils.el in my private dir so long, it
started to mask the official debian version, so this time I dare not
put it in my private dir, lest I forget about it and thus block
updates for years.
Indeed,
Why do I have to tell emacs this file coding is euc-japan, why can't
it tell automatically? file2 can.
# apt-get install file-kanji
$ cd /usr/share/doc/file-kanji/
$ file2 HEADER
HEADER: EUC text
$ emacs -q HEADER
DEB_MULEUCS_UNICODE=off makes emacs startup many times faster, but
when once a month, someone sends me a utf-8 email, the mail looks
broken. I wish there was a DEB_MULEUCS_UNICODE=auto that would
activate it the first time we encounter unicode this session.
Or, with DEB_MULEUCS_UNICODE=off, there
Where is /usr/lib/emacs/21.3/i386-linux/emacsserver etc. documented?
find /usr/share/doc/emacs*|xargs zgrep emacsserver finds nothing.
Tatsuya> Debconf interface cannot help non-admin users. The current
Tatsuya> answer is `/usr/bin/env DEB_MULEUCS_UNICODE=off emacs'. See also
Tatsuya> `/usr/share/doc/mule-ucs/README.Debian'.
Thanks. Emacs now starts 6 times faster for me.
$ time emacs -q -eval '(kill-emacs)'
real0m6.523s
us
I probably use the mule-ucs package once a year, but to have it
installed means that it gets loaded every time I start emacs.
Perhaps dpkg-reconfigure mule-ucs [and other hefty packages] could ask
if one wants to have the package on one's system, but not loaded.
One might say one could just insta
I noticed upon upgrading from the emacs of about 8 weeks ago, if the
old emacs is still running during the upgrade, one might get 'invalid
byte code' messages upon various commands [e.g. w in dired, q in gnus
group buffer...) Therefore it would seem best to exit emacs before
upgrades.
Gentlemen, do
$ find /usr/local/|grep emacs|xargs ls -ld
drwxrwsr-x 4 root staff 4096 2003-07-16 14:49 /usr/local/share/emacs
drwxrwsr-x 3 root staff 4096 2003-07-16 14:49 /usr/local/share/emacs/21.3
drwxrwsr-x 2 root staff 4096 2003-07-16 14:49
/usr/local/share/emacs/21.3/site-lisp
drwxrwsr-x 2 r
Gentlemen, dro(o)p what you are doing and hit C-h h (view-hello-file).
Are there still squares there indicating unknown characters, even
though we are well into the electronic age?
Not only do I see squares for many esoteric languages, but even
several big-name eastern European languages have squ
I'll just run
# dpkg-reconfigure emacs21
to see what questions it might ask me and woah.
emacs-remove emacs21
remove/records-gnuemacs: Handling removal of emacsen flavor emacs21
records: purging...
Hey wait a second, can't you give some reassuring message here so we
don't hit C-c as
Say, would gnus user me benefit from apt-getting mailutils-el?
Description: ... an interface between Mailutils and Emacs using mh-e
module.
$ dlocate -L emacs21-el|sed -n /mh/s/$/c/p|xargs -i find {} -atime -10
No output, i.e. I don't use mh I see. OK, never mind.
Would byte-compiling these speed things up?
Loading 50gnuserv (source)...done
Loading 50html-helper-mode (source)...done
Loading 50mmm-mode (source)...done
etc.
While the gnus maintainer is now hopefully making the gnus package
installable even on my system, I would recommend that if installation
fails, there be a messages emitted to see
http://groups.google.com/groups?selm=u1y713qu2.fsf%40sdm.de
wherein we are told to manually delete the line "(setq
gnus-
One even gets
http://debian.linux.org.tw/debian/pool/main/g/gnus/gnus_5.10.2.orig.tar.gz
and follows the instructions
$ make
cd lisp && make EMACS="emacs" lispdir="/usr/share/emacs/site-lisp" all
make[1]: Entering directory `/home/jidanni/tmp/gnus-5.10.2/lisp'
rm -f *.elc gnus-load.el
URLDIR=/usr/
Gentlemen, you are at risk of me canceling my subscription to
gmane.linux.debian.devel.emacsen. I mean how could _I_ be fed spam?
I'm a vegetarian. Furthermore, a simple run thru SpamAssassin finds
FROM_NUM_AT_WEBMAIL (2.9 points) From address is webmail, but starts with a
number
LOSE_POUNDS
P> If I do `apt-get remove emacs-goodies-el' which is a pretty method these
P> days, or if I `dpkg --remove emacs-goodies-el', the _config_ file
P> stays in place. The next time I start Emacs, it loads the file. Since
P> I have not added a condition in this file about `emacs-goodies-el'
P> actual
Peter> browse-url-browser-function
But when I get a multipart message, and hit K b and then right click
View Externally on text/html here:
1. (*) text/plain ( ) text/html
I still get
Displaying konqueror '/tmp/emm.3522FtC/mm.3522S3I'...
even though I do
(setq browse-url-browser-function '
Speaking about scrollbars, I've mentioned on gmane.emacs.bugs that
often I look up to find my scrollbars gone. They were there when I
started, but somewhere during the several hours that I was using
emacs, they went bye bye...
Dan> 1. (*) text/plain ( ) text/html
Dan> and then middle mouse click on the text/html button choosing
Dan> "View externally" it starts konqueror whereas I want to make
Dan> this mozilla.
J> Do M-x customizebrowse-url and look at the value of
J> "Browser Function".
After the first RET I am swept
Looking for hours in
/usr/share/emacs/site-lisp/gnus/mm-decode.el
I still can't figure out why
when I hit K b and then see the buttons
1. (*) text/plain ( ) text/html
and then middle mouse click on the text/html button
choosing "View externally"
it starts konqueror
whereas I want to make
Miles> would be a good idea to have an `emacs21-for-slow-machines' package.
Naw. Do
$ emacs21 --no-site-file -q -nw
$ emacs21 -q -nw
$ emacs21 --no-site-file -nw
$ emacs21 -nw
$ emacs21
etc. to see where the problem lies. The first should be lightning fast.
Miles> I expect the latter group could largely be satisfied by a
guide to Miles> turning off the various new features of emacs 21
that they dislike What happened to the anti news file? I
distinctly remember an anti news file. P.S. in
/usr/share/emacs/21.2/etc/enriched.doc the line Faces: def
>> I used to be able to paste Chinese texts from Emacs to other X apps
>> until upgrading to KDE3 last year via setting
>> "set-selection-coding-system" to "chinese-iso-8bit-with-esc".
>>
>> I get western characters (ASCII >127) rather than correct Chinese
>> characters when copying Chinese text i
Wow look at all the stuff in /usr/share/emacs/21.2/etc/ that could be
gzipped, like is done for /usr/share/doc/*/* We're talking about
kilobytes of savings.
Wait a second. That might break some stuff that reads them like ^H^H
(help-for-help). Or maybe it would automatically read them anyway
sti
Excuse me, just curious, if they are all linked together,
$ ls -l /usr/bin/emacs* /etc/alternatives/emacs*|sed '[EMAIL PROTECTED]/]*@@'
/etc/alternatives/emacs -> /usr/bin/emacs21
/usr/bin/emacs -> /etc/alternatives/emacs
/usr/bin/emacs21 -> emacs21-x
/usr/bin/emacs21-x
Then why does
$ emacs21 -q
$
>> Hi all. I ran into a slight difficulty. In emacs, i realise you can
>> do C-w and C-SPC to cut tetxt from a buffer which can then be yanked
>> into a buffer with C-Y. However, if you are reviewing a buffer that
>> is read only, you cannot grab the text out in this way.
Ben> Actually, you c
Ian> The thing is, Debian does something a bit unexpected with elisp files:
Ian> it stores the compiled .elc files in a different place from the
Ian> sources. It does that to make it possible to upgrade one's Emacs
Ian> without upgrading each and every unbundled lisp package.
Ian> Only the direct
Gentlemen, I find that the load-path has oh, 60 directories on
it. Though only one exact repeat is found,
/usr/share/emacs21/site-lisp/dictionaries-common , but still, things
look a little inefficient. Here we see the common trailing names:
sed s#.*/## the-load-path.list |sort|uniq -c|sort -nr|etc
>> Odd, the apparently juicy new fat one isn't the one being daily read,
>> $ dlocate message.el|awk '{print $2}'|xargs ls -logtu
>> 159716 2003-03-18 09:02 /usr/share/emacs/21.2/lisp/gnus/message.el
>> 213487 2003-03-09 03:26 /usr/share/emacs/site-lisp/gnus/message.el
>> 125129 2003-03-03 05:03 /u
Odd, .el's newer than .elc's; and two .el's
$ dlocate message.el|awk '{print $2}'|xargs ls -logt
213487 2002-12-24 08:21 /usr/share/emacs/site-lisp/gnus/message.el
125129 2002-10-16 06:03 /usr/share/emacs/21.2/lisp/gnus/message.elc
159716 2002-03-23 01:38 /usr/share/emacs/21.2/lisp/gnus/message.
Why don't you guys include
http://www.netfort.gr.jp/~dancer/software/downloads/apt-el-20030128.tar.gz
hmm, I see. It isn't polished enough yet. OK, never mind.
--
http://jidanni.org/ Taiwan(04)25854780
> "Ian" == Ian Zimmerman <[EMAIL PROTECTED]> writes:
Ian> The following message is a courtesy copy of an article
Ian> that has been posted to gnu.emacs.sources as well.
Ian> ;; this will be the last revision for a while, I promise.
Ian> ;;; dig-browser.el --- a dired-style DNS zone browser
I was about to report this bug, but it happened too for debian-bug too
I do describe-function, and get
apt-sources-mode is an interactive compiled Lisp function in
`apt-sources'.
but when i middle click to see apt-sources.el,:
find-function-search-for-symbol: The library `apt-sourc
Ian> Does list-load-path-shadows turn up anything suspicious?
Dan> Yes!
Ian> It's a bug in the package, then. It should _not_ add the plain
Ian> /site-lisp directory to load-path, only the 21.2 one.
OK, then I hope folks will run list-load-path-shadows and report the
bugs... this is already wel
Ian> Does list-load-path-shadows turn up anything suspicious?
Yes! [I removed the common parts]:
/site-lisp/html-helper-mode/visual-basic-mode hides 21/site-lisp/...
/site-lisp/html-helper-mode/html-helper-mode hides 21/
/site-lisp/html-helper-mode/tempo hides 21/
/site-lisp/html-helper-mode/hhm-
Fellas. I do (load "hhm-config") and end up loading an .el not the .elc:
$ locate hhm-config|xargs ls -ltu|sed s/.*20//
03-01-03 10:15 /usr/share/emacs/site-lisp/html-helper-mode/hhm-config.el
02-12-07 16:10 /usr/share/emacs21/site-lisp/html-helper-mode/hhm-config.elc
Indeed this may be the tip of
Dan> The bbdb I got from sid CD is corrupting any big5 chinese I enter
Dan> and refusing to save the database in big5!
H> Try setting the bbdb-file-coding-system variable.
Thanks, (setq bbdb-file-coding-system 'chinese-big5) fixed it.
--
http://jidanni.org/ Taiwan(04)25854780
The following message is a courtesy copy of an article
that has been posted to gmane.linux.debian.user.chinese.big5 as well.
The bbdb I got from sid CD is corrupting any big5 chinese I enter and refusing
to save the database in big5! Is there anything i can do in a hook to
alleviate the problem?
> "M" == Manoj Srivastava <[EMAIL PROTECTED]> writes:
Dan> What is the justification for a standalone package for gnus if one
Dan> need emacs to run gnus, and that emacs has a newer gnus inside?
M> There is none for emacs21. There is if you run emacs 20. Or
then maybe have the gnus pack
The following message is a courtesy copy of an article
that has been posted to gnu.emacs.gnus as well.
Dan> (gnus-version)"Gnus v5.9.0" is seen in gnus, but the debian version
Dan> number is 5.8.8
Manoj> emacs 21 comes with gnus 5.9.0; and the stand alone gnus
Manoj> package version 5.8.8 does
The following message is a courtesy copy of an article
that has been posted to gnu.emacs.bug as well. wow.
This message consists of a debian problem and an emacs problem. First
the emacs problem.
lets say one has
* Emacs: (emacs-20/emacs). The extensible self-documenting text editor.
* Emacs: (
M-x man ddd gives \267's for the bullets unless one does LC_CTYPE=C.
Is that the price one must pay for having LC_CTYPE=zh_TW.Big5?
Remember of course that sometimes I read Chinese man pages too.
Emacs21.2 "reply to the list"
--
http://jidanni.org/ Taiwan(04)25854780
--
To UNSUBSCRIBE, email t
The following message is a courtesy copy of an article
that has been posted to gnu.emacs.bug as well. Great.
Sorry for the irrelevant groups and even mangled address, but I just
wanted to say I found the solution to folks on the places where I
posted my original question. Solution =
You guys go
me:
#Why in debian woody (under icewm), does galeon, mozilla have nifty
#icons in the toolbar of current windows, but emacs just has a plain X
#like any other plain X window. OK, I cc'd this to the debian dudes.
RMS says in a mail to me:
#If you find out what we should do, please let u
The following message is a courtesy copy of an article
that has been posted to gnu.emacs.bug as well.
This may be a debian specific problem, hence the CC.
It may be "user lack of understanding". If so, "then beef up the
FAQs". Or maybe I didn't look. Anyways, I solved the problem. I put
The following message is a courtesy copy of an article
that has been posted to gnu.emacs.bug as well.
Why in debian woody (under icewm), does galeon, mozilla have nifty
icons in the toolbar of current windows, but emacs just has a plain X
like any other plain X window. OK, I cc'd this to the debi
The following message is a courtesy copy of an article
that has been posted to gnu.emacs.bug as well.
>> I got debian woody and emacs is 20.7. I thought debian was supposed
>> to real advanced or something, but why old emacs?
Chris> Um... install the emacs21 package instead of emacs20. Problem
85 matches
Mail list logo
