September 1998 - XEmacs-Beta - XEmacs Mailing List Archives

Re: package annoyance 26 years, 2 months

Karl M. Hegbloom

Also, what if I copy one .el file out of the system-wide directories, and put it in my ~/.xemacs/packages/** directory, in a mirror to the system-wide file, then make modifications and byte compile it. I want my version to be loaded rather than the site-wide version. Should I have to copy down the entire package, or just one file?

2 participants
1 comments

Re: Proposal: defface.d 26 years, 2 months

Karl M. Hegbloom

>>>>> "Greg" == Greg Klanderman <greg(a)alphatech.com> writes: >>>>> "sb" == SL Baur <steve(a)xemacs.org> writes: sb> sb> Karl M Hegbloom <karlheg(a)inetarena.com> writes in sb> xemacs-beta(a)xemacs.org: >>> What if we pulled the defface stuff out of everything, made >>> them into separate files, and had each package drop that >>> defface file into a directory, like: sb> sb> How much of a problem is this really? Greg> I don't think it is. Some time ago I was trying to figure Greg> out where all my color cells were going (8 bit display). Greg> From what I recall, it seemed that for example the billions Greg> of Gnus faces (which I do not use but are defined in my Greg> XEmacs) did not actually cause color cells to be allocated Greg> until they were used. How much space does a face that's been deffaced use, even if it's never used? What prompted me to tap down that late night note was that I'd used `scheme-lock' to set some extra highlights, and I used `font-lock-emphasized-face', which is defined by cperl mode. When I restarted XEmacs, it failed to finish initializing becuase that's not a standard face. I thought that if the faces are all kept in one place, and got autoloaded when referenced, it would solve that. Maybe I should just say "never mind" and go back to my book reading. :-)

1 participants
0 comments

"finnish landrace" fail to detect PNG 26 years, 2 months

Greg Klanderman

This is the first configure failure I can remember. On Solaris 2.5.1, PNG 1.0.2, png.h says: #define PNG_LIBPNG_VER 10002 /* 1.0.2 */ ./configure gives: checking for pow... yes checking for png.h... yes checking for png_read_image in -lpng... yes checking for workable png version information... no checking for tiffio.h... yes Corsican correctly detected this same PNG. I don't see why I'm not getting this message from configure.h: *** WARNING *** $png_problem I'm not touching that with a 10-foot pole! If you really want to use the installed version of libPNG, rerun configure and add '--with-png=yes', but don't blame me if XEmacs crashes!" Jareth, it looks like you recently patched PNG detection - let me know what further info you need. greg

1 participants
0 comments

package annoyance 26 years, 2 months

Greg Klanderman

If you have 2 versions of a package (eg, the one that the sys admin downloaded with your SUMO binary and a newer one you've installed in ~/.xemacs/packages) you get these irritating warnings on startup: (1) (warning/warning) Autoload error in: /home/greg/.xemacs/packages/lisp/pcl-cvs/auto-autoloads: Already loaded Isn't this an expected usage, one which should not spew warnings? greg

3 participants
2 comments

Re: Portability and usage statistics (was Re: Seppuku and x_IO_error_handler() ) 26 years, 2 months

SL Baur

Michael Sperber <sperber(a)informatik.uni-tuebingen.de> writes in xemacs-beta(a)xemacs.org: >>>>>> "sb" == SL Baur <steve(a)xemacs.org> writes: sb> We do not build on latest AIX. > Not to comment on your posting in general, but we do now. I just sent > patches. Excellent! I am most pleased to stand corrected on this issue.

1 participants
0 comments

Re: Portability and usage statistics (was Re: Seppuku and x_IO_error_handler() ) 26 years, 2 months

SL Baur

Marcus Thiessel <thiessel(a)bbn.hp.com> writes in xemacs-beta(a)xemacs.org: > SL Baur writes: > [...] > | > You forgot to mention HP-UX. Actually, I didn't. ... > Just for the my record, are there any open problems with HP platforms? I'm only aware of the problem about being unable to successfully configure due to /usr/dt being stuffed at the top of the load paths. There's a patch in the incoming queue to address it.

2 participants
1 comments

[gnu.emacs.bug] Bug in byte code optimizer 26 years, 2 months

Barry A. Warsaw

This bug still exists in XEmacs 21.0-b53. -Barry ------- Start of forwarded message ------- Subject: Bug in byte code optimizer Mime-Version: 1.0 (generated by tm-edit 7.106) Content-Type: text/plain; charset=US-ASCII From: mast(a)lysator.liu.se (Martin Stjernholm) Date: 01 Sep 1998 03:39:31 +0200 Message-ID: <7m3eacr4q4.fsf(a)starbug.rydnet.lysator.liu.se> Newsgroups: gnu.emacs.bug There seems to be a bug in the byte code optimizer. The following is a minimal reproducible case under both Emacs 20.2 and XEmacs 20.4. Create a file foo.el with this content (note that char-before is a built-in function): (if whatever (condition-case nil (char-before)) (defsubst char-before () nil)) Start (X)Emacs with -q -no-site-file. Byte compile foo.el with M-x byte-compile-file. I get this in the compilation log: Compiling file /home/mast/elisp/cc-mode/foo.el at Tue Sep 1 03:33:36 1998 ** reference to free variable whatever !! Wrong type argument ((listp #<subr char-before>)) No byte code file is produced. I get no error when I set byte-optimize to nil. /Martin ------- End of forwarded message -------

1 participants
0 comments

Install shield? 26 years, 2 months

Pete Ware

A few days ago someone mentioned they had a version of xemacs that installed with install shield on NT. Can I give it a try? --pete

1 participants
0 comments

Re: Portability and usage statistics (was Re: Seppuku and x_IO_error_handler() ) 26 years, 2 months

Hrvoje Niksic

Olivier Galibert <galibert(a)pobox.com> writes: > *Looking at the ultra in front of him* Hrrrm, what is EOS? Era On Sparc; an old interface to the Sun dbx tools, I think. It used to be supported by Sun, and some people allegedly still use it. > What are annotations, too? Annotations are supposed to be a high-level interface to glyphs and extents. Most of the Lisp authors, including myself, never saw the benefit of using annotations instead of simple extents and glyphs. -- Hrvoje Niksic <hniksic(a)srce.hr> | Student at FER Zagreb, Croatia --------------------------------+-------------------------------- Predestination was doomed from the start.

1 participants
0 comments

Re: Portability and usage statistics (was Re: Seppuku and x_IO_error_handler() ) 26 years, 2 months

SL Baur

Gary D Foster <Gary.Foster(a)Corp.Sun.COM> writes in xemacs-beta(a)xemacs.org: > This talk of only supporting linux disturbs me deeply. I cannot possibly > believe that anyone is seriously considering this? I'm not seriously considering this. End of discussion. Except, see below ... I was pointing out an observation based on what (flawed?) statistical data I had and my belief that usage of Unix systems will (continue to) tend towards Free variants. Nothing more, nothing less. O.K. We are having serious problems keeping XEmacs properly implemented on Unix. No, `serious' doesn't begin to describe the situation. We do not build on latest AIX. You, you and you go fix it. The problem, of course, is that other than Michael Sperber we don't appear to have anyone support AIX. Let's move on to DG/UX and ... forget DG/UX because noone has ever turned in a build report for that system in my tenure as XEmacs maintainer. When I tried configuring on such a system it bombed miserably. O.K. Let's move on to SGI/Irix. Are we passed the traditional XEmacs Fuckage or aren't we? I don't know. Darren Stalder was kindly enough to donate access to a system to test on and ... I never heard heard anything since from the SGI volunteer. O.K. So let's try Solaris. It's the most popular commercial Unix in the market, surely it is getting developer support? *Wrong*! We continue to get crash reports with annotations tickled by the EOS support. There have been hundreds if not thousands of these reports since they've been aimed at me. Is there any developer running with Sun Solaris? Hello? DEC OSF/1 on Alpha has had a long history of breakage with noone working on it. The amount of people working on it has decreased since I became maintainer. If you're concerned that your company's hardware supports XEmacs, send me equipment to test it on. OMRON sent me Wnn6 for testing purposes, and thus XEmacs will *always* support it. (I'm linking their ancient static library against glibc-2.0.7-pre5 with the latest and greatest? Linux 2.1 kernel and it *still* works. The Wnn6 server is the only libc5 binary I usually have running). That is my world view, and that is why I consider commercial Unix support falling apart at the seams. If *you* don't like it, do something about it or send me equipment and software with appropriate licenses and I can take a look at it. Don't just complain about it. It doesn't do anyone any good.

4 participants
4 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 September 1998