May 2000 - XEmacs-Beta - XEmacs Mailing List Archives

XEmacs info.el and PGnus info files 23 years, 10 months

Didier Verna

There's a problem: - our info.el assumes that all info files have a .info extention (modulo gzip compression and stuff). - PGnus calls its info files gnus-[0-9]+ As a result, these files can never be seen in the Info buffer. I discovered that because the PGnus info files are currently the only ones I have in my site-packages directory, and this directory suddenly disapeared from my Info buffer. Personaly, I think that our info.el tries to be too clever, but I also think that PGnus shouldn't mess around with makeinfo's output file name. Waddyathink ? -- / / _ _ Didier Verna http://www.inf.enst.fr/~verna/ - / / - / / /_/ / EPITA / LRDE mailto:didier＠lrde.epita.fr /_/ / /_/ / /__ / 14-16 rue Voltaire Tel. +33 (1) 44 08 01 77 94276 Kremlin-Bicêtre cedex Fax. +33 (1) 44 08 01 99

1 participants
0 comments

Re: Is anyone interested in old dump code? 23 years, 10 months

Giacomo Boffi

Jan Vroonhof writes: > > Does applying Yoshiki's patch work? do you mean? 2000-05-24 Yoshiki Hayashi <yoshiki(a)xemacs.org> * unexelf.c (unexec): Search for ".data" section. Initialize new_data2_offset from old_data_index. Remove redundant check for ElfW. yes, i've applied, and it fails > > '--with-system-malloc' \ > > '--rel-alloc' \ > > These should be automatic... in fact, removing them gave me a working xemacs... > Does 21.1.10 build correctly? i tried it, and it builds (without passing _any_ option to configure) if you want, i'll try removing --with-system-malloc and --rel-alloc one at a time afaict, i have succesfully built xemacs with the above options, and i have built it at least once on RH6.2, the 1st dumped xemacs failing to run was on Sat 20

1 participants
0 comments

Re: Is anyone interested in old dump code? 23 years, 10 months

Jan Vroonhof

giacomo boffi <giacomo.boffi(a)polimi.it> writes: > '--with-system-malloc' \ > '--rel-alloc' \ These should be automatic... Does applying Yoshiki's patch work? Does 21.1.10 build correctly? Jan

1 participants
0 comments

cvs-edit-done calls (ring-insert nil ...) 23 years, 10 months

Adrian Aichner

Using the recently release pcl-cvs 1.46 CVS frontend. I get this *Backtrace* when I try to commit a file: Signaling: (wrong-type-argument number-char-or-marker-p nil) ring-insert(nil "make expiration year 4 digits.\n") #<compiled-function nil "...(120)" [1 ?\n cvs-edit-require-final-newline t y-or-n-p format "Buffer %s does not end in newline. Add one? " buffer-name boundp vc-comment-ring ring-insert buffer-string get-buffer-window cvs-edit-files-buf win cvs-edit-confirm changed cvs-edit-files cvs-edit-initial-files cvs-edit-show-files "Really commit ? " cvs-edit-hide-buf message "Oh, well! Later maybe?" run-hooks cvs-edit-done-hook cvs-edit-keep-buffer cvs-bury-buffer cvs-buffer call-interactively cvs-edit-callback] 4 ("c:\\XEmacs\\xemacs-packages\\lisp\\pcl-cvs\\cvs-edit.elc" . 6054) nil>() call-interactively(cvs-edit-done) ring.el comes from xemacs-base 1.39 Fundamental XEmacs support, you almost certainly need this. Best regards, Adrian

1 participants
0 comments

Re: Is anyone interested in old dump code? 23 years, 10 months

Jan Vroonhof

giacomo boffi <giacomo.boffi(a)polimi.it> writes: > Jan Vroonhof writes: > > [...] Are there any platforms where we are currently failing? > > linux --- please see my previous posts note that `currently' here means: In the CVS version with Yoshiki's patch for the Debian woody problem applied. Jan

2 participants
1 comments

Re: Is anyone interested in old dump code? 23 years, 10 months

Jan Vroonhof

Yoshiki Hayashi <yoshiki(a)xemacs.org> writes: > These are diff between FSF Emacs 20.6.90 and current XEmacs > 21.2. Just by looking at the patch.. > +/* Return the index of the section named NAME. > + SECTION_NAMES, FILE_NAME and FILE_H give information > + about the file we are looking in. > + > + If we don't find the section NAME, that is a fatal error > + if NOERROR is 0; we return -1 if NOERROR is nonzero. */ > + > +static int > +find_section (name, section_names, file_name, old_file_h, old_section_h, noerror) Although defining this function is a good idea, it does not change the effect of the code... > + > + if (old_sbss_index != -1) > + if (!strcmp (old_section_names + NEW_SECTION_H (nn).sh_name, ".sbss")) > + { > + NEW_SECTION_H (nn).sh_offset = > + round_up (NEW_SECTION_H (nn).sh_offset, > + NEW_SECTION_H (nn).sh_addralign); > + NEW_SECTION_H (nn).sh_type = SHT_PROGBITS; > + } This seems to be the only new part. Are there any ChangeLog entries describing the changes? Are there any platforms where we are currently failing? Given the problems we had with the "Linux processor type problem" I am a bit weary of syncing just for syncing sake. Jan

3 participants
2 comments

XEmacs packages? 23 years, 10 months

Daniel Pittman

I have a couple of questions about packages for XEmacs, as I would like to submit one in the not too distant future.[1] Simply put, I have two questions. Firstly, is there documentation on how to put together a package for XEmacs? I can go trolling through the existing packages and stuff, but that's more work than reading a document and all. :) Secondly, what requirements are there on the code and stuff to get it accepted? Daniel Footnotes: [1] Modulo, of course, that it actually does get to a stable(ish) release soon. -- A man's worst difficulties begin when he is able to do as he likes. -- Thomas Henry Huxley

2 participants
1 comments

adding /usr/shlib to X_LIBS 23 years, 10 months

Falk Hueffner

Hi, configure.in contains this line: dnl Avoid version mismatch for shared library libXm.so on osf4 if test "$GCC" = yes -a -d /usr/shlib; then XE_APPEND(-L/usr/shlib, X_LIBS); fi which makes compiling fail on Alpha Linux if you have the Compaq CC installed. I don't when exactly it *is* needed, so perhaps it should be guarded by a check for !linux. Falk

1 participants
0 comments

Re: [PATCH 21.[12]] Fix package checksum calculation 23 years, 10 months

Adrian Aichner

>>>>> "APA" == Adrian Aichner <aichner(a)ecf.teradyne.com> writes: Here's the ChangeLog I forgot to include ... APA> lisp/ChangeLog: 2000-05-28 Adrian Aichner <aichner(a)ecf.teradyne.com> * package-info.el (pi-md5sum): Calculate MD5 sum just like `package-get' does in package-get.el, instead of using external "md5sum" program. * package-get.el (package-get): Insert package file literally to make checksum calculation work. -- Adrian Aichner <adrian(a)xemacs.org>

1 participants
0 comments

Re: untabify works no more 23 years, 10 months

Jan Vroonhof

sperber(a)informatik.uni-tuebingen.de (Michael Sperber [Mr. Preprocessor]) writes: > (indent-tabs-mode nil)) > ^^^^^^^^^^^^^^^^^^^^ > > But `indent-tabs-mode''s C pendant from indent.c: > > DEFVAR_BOOL ("indent-tabs-mode", &indent_tabs_mode /* > *Indentation can insert tabs if this is non-nil. > Setting this variable automatically makes it local to the current buffer. > */ ); > indent_tabs_mode = 1; > > appears unruffled during the execution of `indent-to'. I read through > the huge fuckup of code off specpdl and gave up at some point. Maybe it has somewthing to with the fact that indent-tabs-mode is buffer local? Jan

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 May 2000