December 1998 - XEmacs-Beta - XEmacs Mailing List Archives

XEmacs 21.2.6 "Apollo" is released 25 years, 11 months

Andy Piper

This is nothing specical in that there are no new changes to CVS I am simply taging, updating changelogs and fixing the version so that people can get back at a stable version before we start another patch-fest. I am unable to upload tarballs. to 21.2 beta6 "Apollo" -- mswindows compile fixes from Martin Buchholz, Andy Piper, Greg Klanderman and Adrian Aichner -- Synch with XEmacs 21.0.60 -- mega-patch fixes from Martin Buchholz -- md5 fixes and testsuite from Hrvoje Niksic -- database fix from Hrvoje Niksic andy --------------------------------------------------------------------------- " .sigs are like your face - rarely seen by you and uglier than you think" Dr Andy Piper, Technical Architect, Parallax Solutions Ltd mail: andyp(a)parallax.co.uk web: www.parallax.co.uk/~andyp

1 participants
0 comments

Re: [PATCH] x-symbol should have a _real_ binary package 25 years, 11 months

Jim Radford

> Upfront I want to thank Christoph for his work on X-symbol since he > seems to move onto other things soon. I, of course, would like to second this. Thanks Christoph. > > Third, I really think that XEmacs should handle the font path > > issues itself without asking the user (or administrator) to set it > > up. If this means adding X(Set|Get)FontPath primitives to XEmacs > > then so be it. I'm not sure if this is reasonable. I.e. does a > > "nice X citizen" mess with the font path? > The problem is that the X server does not necessarily run on the > same host as the prorgam is running. The proper way to do it (at > least that is what programs like LyX etc that need their own fonts > seem to do) seem to be to act as a font server yourself. I can see that this might be a problem, but I suspect that often the machines "see" the same filesystem, so it shouldn't be a problem in practice. > > + while (ndirs_return--) > > + font_path = Fcons (build_string (directories[ndirs_return]), font_path); > Please read the 'Rules When Writing New C Code' section of the > "Internals" manual. Especially the "Coding for Mule" section. > You need to make sure the strings are property converted to the > internal MULE string format. I think in this case you need > build_ext_string with a FORMAT_FILENAME argument. > We have been busy stamping out bugs like this in the last month so it > would be unwise to introduce new ones. > > + EXTERNAL_LIST_LOOP (path_entry, font_path) > > + { > > + GET_C_STRING_CTEXT_DATA_ALLOCA (XCAR (path_entry), directories[i++]); > Conversely you probably want > GET_C_CHARPTR_EXT_FILENAME_DATA_ALLOCA Thanks for the code suggestions. I'll send a patch to xemacs-patches. -Jim

1 participants
0 comments

Success/Problems: XEmacs 21.0-b60 "Poitou60" configured for `i586-pc-cygwin32' 25 years, 11 months

Dr. Volker Zell

Hi I successfully compiled Xemacs on Win-NT 4.0 SP3 under cygwin B20.1, see below: But I see the following two problems until now: o gnuclient is not working for me. I get the following errors vzell:/src> gnuclient spell-1.0.diff gnuclient: Connection reset by peer gnuclient: unable to read the reply from the server vzell:/src> o I don't get any bold and italic fonts. I'm running Exceed as the X-Server Any hints ??? ------- uname -a: CYGWIN_NT-4.0 VZELL 20.1 (0.3/1/1) 1998-12-3 20:39:18 i586 unknown ./configure '--package-path=/usr/local/lib/xemacs/site-packages:/usr/local/lib/xemacs/xemacs-packages' '--cppflags=-I/usr/local/include' XEmacs 21.0-b60 "Poitou60" configured for `i586-pc-cygwin32'. Where should the build process find the source code? /src/xemacs-21.0.60 What installation prefix should install use? /usr/local What operating system and machine description files should XEmacs use? `s/cygwin32.h' and `m/intel386.h' What compiler should XEmacs be built with? gcc -g -O2 -Wall -Wno-switch Should XEmacs use the GNU version of malloc? yes Should XEmacs use the relocating allocator for buffers? default What window system should XEmacs use? x11 Where do we find X Windows header files? /usr/X11/include Where do we find X Windows libraries? /usr/X11/lib Compiling in support for XAUTH. Compiling in support for XPM images. Compiling in support for PNG image handling. Compiling in support for (builtin) GIF image handling. Compiling in support for JPEG image handling. Compiling in support for TIFF image handling. Compiling in support for X-Face message headers. Compiling in support for Berkeley DB. Compiling in support for GNU DBM. Compiling in support for LDAP (UMich libs). Compiling in support for ncurses. Compiling in File coding support. Compiling in EXPERIMENTAL support for Drag'n'Drop ( msw ). Compiling in support for proper session-management. Using Lucid menubars. Using Lucid scrollbars. Using Athena dialog boxes. movemail will use "dot-locking" for locking mail spool files. Using Lisp_Objects with minimal tagbits. Using indexed lrecord implementation. Compiling in extra code for debugging. Compiling in code for checking XEmacs memory usage. WARNING: --------------------------------------------------------- WARNING: Compiling in support for runtime error checking. WARNING: XEmacs will run noticeably more slowly as a result. WARNING: Error checking is on by default for XEmacs beta releases. WARNING: --------------------------------------------------------- Ciao Volker -- ___________________________________________________________________________ "Is a dream a lie if it don't come true or is it something worse ..." Dr. Volker Zell __ / _ / __ / ___/ / ___/ -- Senior Dozent -- / / /__/ /_ / / / __/ Oracle Germany ____/__/ _|__/ __/____/ ____/____/ Paul-Thomas-Str.58 D-40599 D|sseldorf e-Mail: mailto:vzell＠de.oracle.com WWW: http://trgsund4.de.oracle.com/ Tel: (+49) 211-74839-414 Fax: (+49) 211-74839-44 ___________________________________________________________________________ The statements and opinions expressed here are my own and do not necessarily represent those of Oracle Corporation.

2 participants
3 comments

Re: [FIX for crash] in Aphrodite when displaying Gnus-picon 25 years, 11 months

Gunnar Evermann

Martin Buchholz <martin(a)xemacs.org> writes: > >>>>> "Gunnar" == Gunnar Evermann <ge204(a)eng.cam.ac.uk> writes: > > Gunnar> I can only make the obvious wild guess here: it's a GC problem! :-) > > One trick to try to get reproducible GC crashes: run xemacs under the > debugger, wait till it comes up, then do > > set always_gc = 1 > > then continue and execute your suspect code. XEmacs will run VERY ^^^^^ > slowly, and it will crash as early as possible. ^^^^^^^ no kidding! this didn't help as apparently only sometimes Lisp stuff is called. But I can trigger the crash if I call garbage_collec_1 from the debugger right before the call to tagged_vector_to_alist in glyphs.c:432. The problem is that normalize_image_instantiator potentially calls simple_image_type_normalize, which has the comment (/* This function can call lisp */). the fix is appended below. One could make the gcprotection local to the 'STRINGP' case, but I was to lazy, should I change that? 1998-12-15 Gunnar Evermann <ge204(a)eng.cam.ac.uk> * glyphs.c (normalize_image_instantiator): GCPRO instantiator Index: src/glyphs.c =================================================================== RCS file: /usr/CVSroot/XEmacs/xemacs/src/glyphs.c,v retrieving revision 1.23.2.2 diff -u -r1.23.2.2 glyphs.c --- glyphs.c 1998/12/05 16:55:54 1.23.2.2 +++ glyphs.c 1998/12/15 22:59:26 ＠＠ -466,8 +466,12 ＠＠ Lisp_Object contype, Lisp_Object dest_mask) { + struct gcpro gcpro1; + + GCPRO1(instantiator); + if (IMAGE_INSTANCEP (instantiator)) - return instantiator; + RETURN_UNGCPRO (instantiator); if (STRINGP (instantiator)) instantiator = process_image_string_instantiator (instantiator, contype, ＠＠ -484,9 +488,9 ＠＠ struct image_instantiator_methods * meths = decode_image_instantiator_format (XVECTOR_DATA (instantiator)[0], ERROR_ME); - return IIFORMAT_METH_OR_GIVEN (meths, normalize, + RETURN_UNGCPRO (IIFORMAT_METH_OR_GIVEN (meths, normalize, (instantiator, contype), - instantiator); + instantiator)); } } -- Gunnar Evermann Speech, Vision & Robotics Group Engineering Department Cambridge University

3 participants
2 comments

Re: [CRASH] 21.2-b5 with Jonathan's gc crash fix applied 25 years, 11 months

Adrian Aichner

>>>>> "Jonathan" == Jonathan Harris <jhar(a)tardis.ed.ac.uk> writes: Jonathan> Adrian Aichner wrote: >> >> This seems to be a new kind... Jonathan> I haven't played with 21.2 since Martin's changes. Could Jonathan> you provide a debug_backtrace() next time you hit this Jonathan> and a description of how to reproduce the problem, or Jonathan> failing that a description of what you were doing. Jonathan, since this last crash I have been using pgnus-0.66 heavily and avoided no attempt at crashing XEmacs; withoout success so far! Thanks for the patch! I'm not familiar with using debug_backtrace(). Do I call it form the debugger after distater stroke? Adrian Jonathan> Jonathan. Jonathan> -- Jonathan> Jonathan Harris | jhar(a)tardis.ed.ac.uk Jonathan> London, England | Jonathan.Harris(a)symbian.com -- Adrian Aichner European Design Center Teradyne GmbH Semiconductor Test Group Telephone +49/89/41861(0)-208 Dingolfinger Strasse 2 Fax +49/89/41861-217 (What is a Fax?) D-81673 MUENCHEN E-mail adrian.aichner(a)teradyne.com

1 participants
0 comments

[CRASH] 21.2-b5 with Jonathan's gc crash fix applied 25 years, 11 months

Adrian Aichner

This seems to be a new kind... Adrian NTDLL! 77f76148() error_check_cons(Lisp_Object {...}) line 717 + 80 bytes condition_case_unwind(Lisp_Object {...}) line 1521 + 9 bytes unbind_to_hairy(int 0x00000018) line 4625 + 18 bytes unbind_to(int 0x00000018, Lisp_Object {...}) line 4600 + 181 bytes garbage_collect_1() line 4548 + 16 bytes Feval(Lisp_Object {...}) line 2911 Fprogn(Lisp_Object {...}) line 761 + 9 bytes Flet(Lisp_Object {...}) line 926 + 9 bytes Feval(Lisp_Object {...}) line 2959 + 13 bytes Fprogn(Lisp_Object {...}) line 761 + 9 bytes funcall_lambda(Lisp_Object {...}, int 0x00000000, Lisp_Object * 0x0082eaf4) line 3508 + 9 bytes Feval(Lisp_Object {...}) line 3092 + 20 bytes Fif(Lisp_Object {...}) line 676 + 9 bytes Feval(Lisp_Object {...}) line 2959 + 13 bytes Fprogn(Lisp_Object {...}) line 761 + 9 bytes funcall_lambda(Lisp_Object {...}, int 0x00000000, Lisp_Object * 0x0082ee74) line 3508 + 9 bytes Feval(Lisp_Object {...}) line 3092 + 20 bytes Fprogn(Lisp_Object {...}) line 761 + 9 bytes funcall_lambda(Lisp_Object {...}, int 0x00000000, Lisp_Object * 0x0082f0e8) line 3508 + 9 bytes Ffuncall(int 0x00000001, Lisp_Object * 0x0082f0e4) line 3229 + 17 bytes execute_optimized_program(const unsigned char * 0x0082f2d8, int 0x00000004, Lisp_Object * 0x01225b3c) line 750 + 16 bytes Fbyte_code(Lisp_Object {...}, Lisp_Object {...}, Lisp_Object {...}) line 2400 + 38 bytes Feval(Lisp_Object {...}) line 2986 + 195 bytes condition_case_1(Lisp_Object {...}, Lisp_Object (Lisp_Object)* 0x01058ae6 Feval(Lisp_Object), Lisp_Object {...}, Lisp_Object (Lisp_Object, Lisp_Object)* 0x0105641e run_condition_case_handlers(Lisp_Object, Lisp_Object), Lisp_Object {...}) line 1640 + 7 bytes condition_case_3(Lisp_Object {...}, Lisp_Object {...}, Lisp_Object {...}) line 1721 + 27 bytes execute_rare_opcode(Lisp_Object * 0x0082f630, const unsigned char * 0x01225e12, int 0x0000008f) line 1275 + 19 bytes execute_optimized_program(const unsigned char * 0x01225d4c, int 0x00000008, Lisp_Object * 0x01225994) line 660 + 17 bytes funcall_compiled_function(Lisp_Object {...}, int 0x00000001, Lisp_Object * 0x0082f92c) line 530 + 53 bytes Ffuncall(int 0x00000002, Lisp_Object * 0x0082f928) line 3221 + 17 bytes execute_optimized_program(const unsigned char * 0x01226764, int 0x00000005, Lisp_Object * 0x01226708) line 750 + 16 bytes funcall_compiled_function(Lisp_Object {...}, int 0x00000001, Lisp_Object * 0x0082fc24) line 530 + 53 bytes Ffuncall(int 0x00000002, Lisp_Object * 0x0082fc20) line 3221 + 17 bytes call1(Lisp_Object {...}, Lisp_Object {...}) line 3834 + 11 bytes execute_internal_event(Lisp_Object {...}) line 3044 + 19 bytes Fdispatch_event(Lisp_Object {...}) line 4689 + 9 bytes Fcommand_loop_1() line 578 + 9 bytes command_loop_1(Lisp_Object {...}) line 494 condition_case_1(Lisp_Object {...}, Lisp_Object (Lisp_Object)* 0x0103a8fe command_loop_1(Lisp_Object), Lisp_Object {...}, Lisp_Object (Lisp_Object, Lisp_Object)* 0x0103a38b cmd_error(Lisp_Object, Lisp_Object), Lisp_Object {...}) line 1640 + 7 bytes command_loop_3() line 255 + 35 bytes command_loop_2(Lisp_Object {...}) line 267 internal_catch(Lisp_Object {...}, Lisp_Object (Lisp_Object)* 0x0103a4f3 command_loop_2(Lisp_Object), Lisp_Object {...}, int * volatile 0x00000000) line 1315 + 7 bytes initial_command_loop(Lisp_Object {...}) line 304 + 25 bytes STACK_TRACE_EYE_CATCHER(int 0x00000001, char * * 0x10105680, char * * 0x101ce100, int 0x00000000) line 1715 + 9 bytes main(int 0x00000001, char * * 0x10105680, char * * 0x101ce100) line 2144 + 21 bytes XEMACS! mainCRTStartup + 180 bytes _start() line 165 KERNEL32! 77f1b304() -- Adrian Aichner European Design Center Teradyne GmbH Semiconductor Test Group Telephone +49/89/41861(0)-208 Dingolfinger Strasse 2 Fax +49/89/41861-217 (What is a Fax?) D-81673 MUENCHEN E-mail adrian.aichner(a)teradyne.com

2 participants
1 comments

Re: [PATCH] x-symbol should have a _real_ binary package 25 years, 11 months

Jan Vroonhof

Christoph Wedler <wedler(a)fmi.uni-passau.de> writes: > JV> there. [...] Now if you really want to be a bad boy you can put > JV> the generated files in the MANIFEST file but NOT include them in > JV> the distribution. > > What about the created lex files and executables for conversion in > etc/x-symbol/bin/SYSCONFIG/ > > They get created after "M-x x-symbol-exec-create" under XEmacs/no-Mule. > I cannot make them appear in the MANIFEST because SYSCONFIG is the value > of `system-configuration' and users might have defined their own token > languages (BTW, language "Isabelle Symbol" is in preparation, > to be distributed with the Emacs package ProofGeneral). These file are major PITA [1] for the package system anyway since one day maybe the package tree will move to /usr/..../share and you do not want executables there :-( > BTW, it would be only 67k if faces would have a face property > `baseline-shift'... You mean XEmacs faces? Maybe that wouldn't even that difficult to add. > OK, I will include them. Although the real benefit (no user interaction > during installation) is only there with sth like Jim's code for an Elisp > function set-font-path... Of course nothing forbids you to run a (call-process "xset" nil nil nil "+fp" "path") :-) Jan Footnotes: [1] I like greek Pita bread a lo so I find this acronym to be somewhat doubtful. -- Jan Vroonhof http://www.math.ethz.ch/~vroonhof/ Mathematik, vroonhof ＠ math.ethz.ch HG E16, ETH-Zentrum, Tel: +41-1-6325456/25154 Raemistrasse 101, CH-8092 Zuerich. Fax: +41-1-6321085

1 participants
0 comments

Re: [PATCH] x-symbol should have a _real_ binary package 25 years, 11 months

Christoph Wedler

>>>>>"JV" == Jan Vroonhof <vroonhof(a)math.ethz.ch> writes: JV> Christoph Wedler <wedler(a)fmi.uni-passau.de> writes: >> Hm, if you want to remove everything, why not use >> rm {etc,lisp,...}/PACKAGE/* ? JV> That is what we used to do but no longer. All the files belonging JV> to the package should be in the MANIFEST file. That's why it JV> there. [...] Now if you really want to be a bad boy you can put JV> the generated files in the MANIFEST file but NOT include them in JV> the distribution. What about the created lex files and executables for conversion in etc/x-symbol/bin/SYSCONFIG/ They get created after "M-x x-symbol-exec-create" under XEmacs/no-Mule. I cannot make them appear in the MANIFEST because SYSCONFIG is the value of `system-configuration' and users might have defined their own token languages (BTW, language "Isabelle Symbol" is in preparation, to be distributed with the Emacs package ProofGeneral). Note that this is not a problem of X-Symbol alone: M-x TeX-auto-generate-global creates files in etc/auctex/auto/ which don't appear in the MANIFEST. IMHO, the best thing would be that packages can have their own private directory in the package hierarchy which they could use at will, e.g. data/{auctex,x-symbol}. >> Thus, if more people think that putting the pcf font files with the >> binary package, this would be fine with me. Hm, I've though of >> that. But this would add 200k (tarred) to the binary package >> (versus 1x "make" taking 5s)... BTW, it would be only 67k if faces would have a face property `baseline-shift'... JV> I think with the current state of the package system this is the JV> best solution. OK, I will include them. Although the real benefit (no user interaction during installation) is only there with sth like Jim's code for an Elisp function set-font-path... JR> It might be better to have a separate font package JR> x-symbol-fonts-pkg.tar.gz to help save down-loading time during an JR> upgrade. >> I have though of that, too. (Jan, was it you who said it is better not >> to split the distribution?) JV> I am not sure. It does create some hassle (mainly because the package JV> system does not support dependencies now). However the hassle is not JV> that BIG. OK, if there is a general way how to handle additional fonts with the package system, I will probably split the distribution (but not now). JV> Maybe you can make the x-symbol package signal a warning if there JV> is no fonts package installed. It does already. - Christoph

1 participants
0 comments

Re: List of packages, maintainers and dev sites? 25 years, 11 months

Adrian Aichner

>>>>> "Jari" == Jari Aalto+list xemacs <jari.aalto(a)poboxes.com> writes: Jari> Hi, Jari> I'm looking for single point of entry to the Jari> - list of packages that can be installed to 21.x. Jari> Alternatively to a list of XEmacs 20.4 features (that Jari> is, eg. 'w3 for bunch of W3 files) - Contact address to Jari> the maintainer of the feature - Development site URI Jari> from where to get latest version or beta for the Jari> feature. Hello Jari, in 21.0 there is `package-get-base' variable describing the list of installed packages. The latest version of all available packages can be found at /ftp.xemacs.org:pub/xemacs/packages/package-index.LATEST.pgp This file contains additional information (e.g. maintainer). It has all the information you asked for. Regards, Adrian `package-get-base' (buffer: *Hyper Apropos*, mode: Hyper-Apropos) Variable: value: see below List of packages that are installed at this site. For each element in the alist, car is the package name and the cdr is a plist containing information about the package. Typical fields kept in the plist are: version - version of this package provides - list of symbols provided requires - list of symbols that are required. These in turn are provided by other packages. filename - name of the file. size - size of the file (aka the bundled package) md5sum - computed md5 checksum description - What this package is for. type - Whether this is a 'binary (default) or 'single file package More fields may be added as needed. An example: '( (name (version "<version 2>" file "filename" description "what this package is about." provides (<list>) requires (<list>) size <integer-bytes> md5sum "<checksum" type single ) (version "<version 1>" file "filename" description "what this package is about." provides (<list>) requires (<list>) size <integer-bytes> md5sum "<checksum" type single ) ... )) For version information, it is assumed things are listed in most recent to least recent -- in other words, the version names don't have to be lexically ordered. It is debatable if it makes sense to have more than one version of a package available. Jari> I suppose that can be somehow created from current Jari> packaging system, but I only have access to 20.3. I Jari> could grep the provide commands from the source tree, Jari> but is this accurate? Jari> Could you put this list available somewhere public and Jari> update as Jari> needed? Jari> jari -- Adrian Aichner European Design Center Teradyne GmbH Semiconductor Test Group Telephone +49/89/41861(0)-208 Dingolfinger Strasse 2 Fax +49/89/41861-217 (What is a Fax?) D-81673 MUENCHEN E-mail adrian.aichner(a)teradyne.com

1 participants
0 comments

Re: [PATCH] x-symbol should have a _real_ binary package 25 years, 11 months

Jan Vroonhof

Christoph Wedler <wedler(a)fmi.uni-passau.de> writes: > Hm, if you want to remove everything, why not use > rm {etc,lisp,...}/PACKAGE/* ? That is what we used to do but no longer. All the files belonging to the package should be in the MANIFEST file. That's why it there. This is mainly to make sure the automated package maintenance tools can do the right thing. Hopefully the MANIFEST files will even contain md5sums in the future. Now if you really want to be a bad boy you can put the generated files in the MANIFEST file but NOT include them in the distribution. > JR> The pcf font files should come bundled with the binary release. > > Hm, I've though of that. But this would add 200k (tarred) to the binary > package (versus 1x "make" taking 5s)... I think this is a minor concern. I binary package should be exactly that, just some stuff to put somewhere. Maybe in the future XEmacs will allow scripts to be run before and after installing/removing like real packagers. If you want to trade off download for compile time, get the source. > much about this font stuff. Thus, if more people think that putting the > pcf font files with the binary package, this would be fine with me. I think with the current state of the package system this is the best solution. > JR> It might be better to have a separate font package > JR> x-symbol-fonts-pkg.tar.gz to help save down-loading time during an > JR> upgrade. > > I have though of that, too. (Jan, was it you who said it is better not > to split the distribution?) I am not sure. It does create some hassle (mainly because the package system does not support dependencies now). However the hassle is not that BIG. Maybe you can make the x-symbol package signal a warning if there is no fonts package installed. > But I think down-loading a second package > is more work than to type "make" once. How would this work with > pui? When xsymbol is part of the package-index database downloading a second package is a matter of typing 'RET' and 'g'. 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 December 1998