July 1998 - XEmacs-Beta - XEmacs Mailing List Archives

Re: What option should be tested ? 26 years, 4 months

SL Baur

Marcus Thiessel <thiessel(a)hpbbse.bbn.hp.com> writes in xemacs-beta(a)xemacs.org: > Any option to configure that should be tested? I could do a number of > builds on my box. I just need to know which are the most important. Default builds and minimal Mule builds are probably the most important. Build reports with default build options `./configure' or `./configure --with-mule' are nonexistent.

2 participants
1 comments

Re: key bindings from the user's point-of-view 26 years, 4 months

Jan Vroonhof

Marc Aurel <4-tea-2(a)bong.saar.de> writes: > since running X on Red Hat LiNUX 5.1 with the de-latin1 keymap, I get > this boring "dead-circumflex not defined" whenever I touch the "^"-key > (and I don't mean Control) in XEmacs. The key works fine in other > applications (e.g. pressing "^", then space will give me the circumflex > in xterm) - so I imagine I should correct the key binding in XEmacs. You should... Wheee...:-( Martin and I already spent days on this. Everytime we fix it it breaks again. The problem is there are too many different deadkeys setup all to test them all :-( Could you eval the following (x-keysym-on-keyboard-p 'dead-circumflex) and mail me the result. Then apply the following patch, redump XEmacs and see if this works Index: x-init.el =================================================================== RCS file: /usr/CVSroot/XEmacs/xemacs-20/lisp/x-init.el,v retrieving revision 1.8 diff -u -u -r1.8 x-init.el --- x-init.el 1998/05/16 05:06:55 1.8 +++ x-init.el 1998/07/20 17:22:19 ＠＠ -208,6 +208,7 ＠＠ (x-define-dead-key dead-cedilla compose-cedilla-map) (x-define-dead-key dead-diaeresis compose-diaeresis-map) (x-define-dead-key dead-circum compose-circumflex-map) + (x-define-dead-key dead-circumflex compose-circumflex-map) (x-define-dead-key dead-tilde compose-tilde-map) ) > I can see two ways to do this: a) binding the circumflex character to the > dead-circumflex key, b) using dead-circumflex the same way dead-anything > is used (with X-compose). b) is what XEmacs is supposed to do already. > I would've been *so* happy if I would've been able to figure out how > method a) works. But I wasn't. The info manual has at least a dozen > examples on how to use global-set-key to set a key to a command. But I > want to set it to a character sequence. The FAQ talks quite a bit about > key bindings - but, you might've guessed - no hint on how to redefine a > key to insert a simple character. [For the records: the last thing I > tried was: global-set-key ^ self-insert-command] The trick is to use a string instead of a command M-x global-set-key RET [dead-circumflex] RET "^" RET you will have to paste the string "^" in from an Xterm. > OTOH: I think I shouldn't see a message like "dead-circumflex not > defined" after a fresh installation. Has Red Hat messed up something or > is it something in my local setup? Difficult question it is partly an omission in XEmacs, a mess from XFree86 and our friends from the late X consortium and you being you scared to scared to go for MULE :-) (Note the smiley, we are in dire need of poeple with a setup like you (the rest of Germany has long since disabled deadkeys in XFree). If you want to know, MULE enables XIM, which makes the deadkey processing be handled by the X libraries where it shuuld). Keep on testing (I am afraid it so addictive you'll start to learn elisp anyway. Jan

1 participants
0 comments

Re: A bug with TeX \$ special sequence 26 years, 4 months

Christoph Wedler

>>>>>"BM" == Bartosz Maruszewski <B.Maruszewski(a)jtz.org.pl> writes: BM> I've been writing a lot in LaTeX last time and noticed that XEmacs BM> interprets incorrectly the "\$" sequence. It treats it as if it was a math BM> formula starting while it's just a shell-variable name's starting. Like BM> here : The shell varaible {\tt \$blah} ...". XEmacs starts to highlight BM> everything green. And it even doesn't end with next "\$" sequence, but BM> sometimes earlier (on a newline for example). BM> It refers to XEmacs v20.4 on a Linux 2.0.34 on an Pentium box. I got BM> the precompiled version from a mirror site. Same for 21.0. The right patch would be to fix the syntax class `character quote'... Here is a workaround... - Christoph 1998-07-20 Christoph Wedler <wedler(a)fmi.uni-passau.de> * tex-mode.el (tex-common-initialization): Use syntax class `escape' for \. Syntax class `character quote' only works for `forward-sexp' and friends. diff -c /home/clstaff/wedler/xemacs/xemacs-packages/lisp/texinfo/tex-mode.el~ /home/clstaff/wedler/xemacs/xemacs-packages/lisp/texinfo/tex-mode.el *** /home/clstaff/wedler/xemacs/xemacs-packages/lisp/texinfo/tex-mode.el~ Mon Jul 20 11:01:31 1998 --- /home/clstaff/wedler/xemacs/xemacs-packages/lisp/texinfo/tex-mode.el Mon Jul 20 11:01:31 1998 *************** *** 406,412 **** (modify-syntax-entry ?\f ">") (modify-syntax-entry ?$ "$$") (modify-syntax-entry ?% "<") ! (modify-syntax-entry ?\\ "/") (modify-syntax-entry ?\" ".") (modify-syntax-entry ?& ".") (modify-syntax-entry ?_ ".") --- 406,412 ---- (modify-syntax-entry ?\f ">") (modify-syntax-entry ?$ "$$") (modify-syntax-entry ?% "<") ! (modify-syntax-entry ?\\ "\\") (modify-syntax-entry ?\" ".") (modify-syntax-entry ?& ".") (modify-syntax-entry ?_ ".")

1 participants
0 comments

Some questions about packages 26 years, 4 months

Olivier Galibert

I eventually played with package-* instead of untarring where needed and I think the low-level routines are quite nice. The only missing thing is a point-and-drool interface for normal users[1] and a remote package-get-base. Some questions though: 1- I retrieve a package with package-get. How can I "activate" it ? For instance, I find-file a Makefile. The editor yells at me that makefile-mode is unknown (normal). I (package-get 'makefile). Now, can I use it without restarting the editor ? - (require 'makefile) and (load-library 'makefile) do not find the files - package-*-autoloads DOC strings include blinking "don't use this or Steven will personally come and bite you where it hurts" comments. 2- Is there a magical incantation to know why a given package is loaded by (package-get-update-all)? I'd like to know why vm has suddendly[2] been considered necessary[3]. 3- If a package includes a dumped-lisp file and is retrived after having compiled and installed xemacs, will it work anyway? Of course, I'm not talking about packages like "xemacs-base" or "mule-base". 4- Is there another magical incantation to check for the internal consistency of the installed packages? The question can be reformulated as "were installed packages all up-to-date, is there a function which says which packages package-get-update-all would retrieve?". That should be enough for a start :-) OG. [1] Easier to whine than to code :-) [2] I guess I should have used package-get-all instead of package-get [3] I'm on a E1, so it's not *really* a problem, but it sucks anyway. My MUA is mutt, damnit' :-)

2 participants
1 comments

21.0 release checklist (was Re: Iceberg, right ahead) 26 years, 4 months

SL Baur

[Redirected to xemacs-beta] Andy Piper <andyp(a)parallax.co.uk> writes in xemacs-review(a)xemacs.org: > At 04:24 17/07/98 -0700, SL Baur wrote: >> If you can do without binaries, or can wait for binaries for a couple >> of weeks until Jason comes back we can still try to hit late July. >> Jason Mastaler the binary coordinator is going on vacation. > I think binaries are an essential part of the release. > Apart from them what remains to be done? This is my list. If I've forgotten anything, please refresh my memory. 1. More package documentation. 2. Package deletion procedure. 3. Documentation of how single-file packages are supposed to work.[1] 4.0 Finalize Lisp packages to go out with 21.0. We have pending integration of bbdb 2 and possibly MEW. Where does bbdb live now? 4.1 Creation and testing of the SUMO lisp package. > The activity on xemacs-beta has been fairly low of late. It'll pick up again when there are advertised places to put patches for the future. Summer is a slow time for development anyway as many people go on vacation. Footnotes: [1] There is little likelihood I'm going to be able implement this in time for 21.0.

2 participants
1 comments

Re: 21.0 release checklist (was Re: Iceberg, right ahead) 26 years, 4 months

Oscar Figueiredo

>>>>> "sb" == SL Baur <steve(a)xemacs.org> writes: [...] sb> We have pending integration of bbdb 2 and possibly MEW. Where sb> does bbdb live now? There's anonymous CVS access to the main BBDB repository: From: Matt Simmons <simmonmt(a)acm.org> Date: 24 Feb 1998 23:01:16 -0800 As so rather unsubtly hinted at on Sunday night, the BBDB is now available via anonymous CVS. To check out the sources, you first have to log in. Use the following command: cvs -d :pserver:bbdb＠cvs.xemacs.org:/usr/CVSroot login CVS password: bigbrother To actually check out the sources, use the following command: cvs -d :pserver:bbdb＠cvs.xemacs.org:/usr/CVSroot checkout bbdb Please address any questions about it to me. Enjoy! Matt Oscar

1 participants
0 comments

What option should be tested ? 26 years, 4 months

Marcus Thiessel

Any option to configure that should be tested? I could do a number of builds on my box. I just need to know which are the most important. --Marcus

1 participants
0 comments

Re: Common Lisp vs. Scheme 26 years, 4 months

Raymond Toy

>>>>> "Glynn" == Glynn Clements <glynn(a)sensei.co.uk> writes: Glynn> Raymond Toy wrote: >> One question, though, on the tail-recursive feature. It's nice to >> have but is it really, really important? Elisp isn't (right?), and >> we've been doing fine without proper tail recursion. Glynn> We've been *surviving* without tail recursion optimisation. Glynn> Ultimately, tail recursion can always be converted to explicit Glynn> iteration, which is what you need to do in elisp if you don't want to Glynn> take a performance hit. Of course. I'm just curious how bad it really is. I assume that the performance hit is not the lack of tail-recursivion, but the exceedingly slow funcall's in elisp. Some discussions in comp.lang.lisp (or scheme?) mentioned that even though tail-recursion is equivalent to iteration, it still takes a toll on speed. 20% maybe or more? I don't remember the numbers from the tests that were done. Glynn> However, it's effectively a bug (IMHO) which we've had to work around. Glynn> It would be nice if algorithms which would naturally be implemented Glynn> using tail recursion could be written in their natural form. I guess my question is, if things were written recursively, how a big a stack would we use and would that be big enough to matter? Not theoretical questions, but real world numbers for xemacs. (No, I'm not asking anyone to actually do this with xemacs. I wouldn't do it. :-) ) Ray

3 participants
2 comments

Re: Common Lisp vs. Scheme 26 years, 4 months

Raymond Toy

>>>>> "Michael" == Michael Sperber <sperber(a)informatik.uni-tuebingen.de> writes: Michael> Many thanks to everyone who have participated in the discussion so Michael> far. Michael> I've done a pass over my web pages at Michael> http://www-pu.informatik.uni-tuebingen.de/users/sperber/xemacs/next-gener... Michael> I've removed a lot of the noise (hopefully), and added stuff on Common Michael> Lisp from the recent posts on xemacs-beta. My aim is, at this point A good overview. One question, though, on the tail-recursive feature. It's nice to have but is it really, really important? Elisp isn't (right?), and we've been doing fine without proper tail recursion. Also, is CLISP tail-recursive? That fact that ANSI CL doesn't require it doesn't mean we can't have it. I don't have great preferences to either Scheme or CL. I think it would be great to replace elisp with one of these. Ray

4 participants
3 comments

Re: scheme - my opinion 26 years, 4 months

Justin Sheehy

Martin Buchholz <martin(a)xemacs.org> writes: > I actually go further and claim that call/cc is too difficult even for > your run-of-the-mill XEmacs implementer, like me, to understand. > Maybe I'm in need of some 180 degree neck-twisting. I think that the reason why everyone is having so much trouble understanding call/cc is that they're trying to do so without first understanding the basic concept of continuations. When I first started looking at Scheme code that used call/cc, I was baffled. I was totally unable to wrap my brain around it. Then, I read the chapter on continuations in _Essentials of Programming Languages_ (Wand, et al) two or three times. Once I understood continuations in the general sense, call/cc was no big deal. To show how much of a difference this makes: I used this approach in a class that I taught. At the outset, I had ten people who were familiar with the basics of Scheme, but none of them were very proficient. Two hours later, they all undertsood the wonders of continuation passing. I attribute most of this to the fact that I didn't even mention call/cc for at least an hour. :-) -- Justin Sheehy In a cloud bones of steel.

1 participants
0 comments

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

XEmacs-Beta July 1998