August 1999 - XEmacs-Beta - XEmacs Mailing List Archives

trouble getting started with binary distribution 25 years, 4 months

Steve Grout

The FAQ said to post this to xemacs-beta-request, but that alias does not seem to exist. I apologize for posting the list directly. ... Can you please direct me to the appropriate INSTALL or README.INSTALL? I downloaded Xemacs 20 from the ftp site and WinZip7.0 extracted it into c:\bin\Xemacs. I can't get it to start. I've been thru the FAQ. The backgraound of this is that I am a long time emacs and gnuemacs user and am wanting to try out Xemacs to see which I like best. I currently have NtEmacs installed and running (I have clipped in the 'set' output following the command line output. --Steve Grout Sematech ------------------------------ C:\bin\Xemacs\bin\i686-pc-cygwin32>xemacs -q -no-site-file > startup-ab WARNING: Couldn't find an obvious default for the root of the XEmacs hierarchy. WARNING: Couldn't find obvious defaults for: doc-directory data-directory exec-directory Perhaps some directories don't exist, or the XEmacs executable, /bin/Xemacs/bin/i686-pc-cygwin32/xemacs.exe is in a strange place? C:\bin\Xemacs\bin\i686-pc-cygwin32>dir Volume in drive C has no label. Volume Serial Number is C047-32FA Directory of C:\bin\Xemacs\bin\i686-pc-cygwin32 08/23/99 11:19a <DIR> . 08/23/99 11:19a <DIR> .. 07/02/99 06:59a 300,068 b2m.exe 07/02/99 06:59a 384,936 ctags.exe 12/03/98 08:44p 456,192 cygwin1.dll 07/02/99 06:59a 379,700 etags.exe 07/02/99 06:59a 872 gnuattach 07/02/99 06:59a 311,056 gnuclient.exe 07/02/99 06:59a 1,075 gnudoit 07/02/99 06:59a 381,305 ootags.exe 07/02/99 06:59a 5,402 pstogif 07/02/99 06:59a 3,243 rcs-checkin 07/02/99 06:59a 542,403 runemacs.exe 08/23/99 11:19a 68 startup-abort.log 07/02/99 06:59a 4,516,578 xemacs-21.1-p3.exe 07/02/99 06:59a 4,516,578 xemacs.exe 16 File(s) 11,799,476 bytes 1,072,521,728 bytes free C:\bin\Xemacs\bin\i686-pc-cygwin32>type startup-abort.log Cannot open load file: /build/xemacs/xemacs-20/lisp/auto-autoloads C:\bin\Xemacs\bin\i686-pc-cygwin32> --------------------------------- ClassPath=C:\Program Files\Exceed.nt\hcljrcsv.zip;C:\Program Files\Exceed.nt;c\bin;c:\bin\unix;c:\bin\unixnt;n:\orawin\bin COMPUTERNAME=GROUTS3 COMSPEC=C:\WINNT\system32\cmd.exe EM_PARENT_PROCESS_ID=309 emacs_dir=C:/bin/emacs-20.4 EMACSDATA=C:/bin/emacs-20.4/etc EMACSDIR=c:\bin\emacs-20.4 EMACSDOC=C:/bin/emacs-20.4/etc EMACSLOADPATH=C:/bin/emacs-20.4/site-lisp;C:/bin/emacs-20.4/lisp EMACSLOCKDIR=C:/bin/emacs-20.4/lock EMACSPATH=C:/bin/emacs-20.4/bin HOME=c:\bin\emacs-20.4 HOMEDRIVE=H: HOMEPATH=\ HOMESHARE=\\share1\myname$ LOGONSERVER=\\whoever MWPATH=C:\MWW32\MANAGER;C:\MWW32\PCMWAVE;C:\MWW32\DSP; NUMBER_OF_PROCESSORS=1 OS=Windows_NT Os2LibPath=C:\WINNT\system32\os2\dll; PATH=c:\bin\perl\bin;C:\WINNT\SYSTEM32;C:\WINNT;C:\MWW32\BIN;C:\MWW32\MODEM;c:\bin;c:\bin\unix;c:\bin\unixnt;c:\Program Files\Exceed.nt;C:\PROGRA~1\MICROS~1\Office;C:\WINNT\System32\WBEM;c:\Program Files\exceed.nt;c:\bin;c:\bin\unix;c:\bin\unixnt;c:\bin\emacs-20.4\bin;c:\bin\Xemacs\bin\i686-pc-cygwin32;N:\ORAWIN\BIN;C:\ORANT\BIN PATHEXT=.COM;.EXE;.BAT;.CMD PROCESSOR_ARCHITECTURE=x86 PROCESSOR_IDENTIFIER=x86 Family 6 Model 5 Stepping 0, GenuineIntel PROCESSOR_LEVEL=6 PROCESSOR_REVISION=0500 PROMPT=$P$G SHELL=C:/bin/emacs-20.4/bin/cmdproxy.exe SMS=YES SMS_LOCAL_DIR=C:\WINNT SystemDrive=C: SystemRoot=C:\WINNT TEMP=C:\TEMP TERM=dumb TMP=C:\TEMP TMPDIR=C:\TEMP USERDOMAIN=LOGON USERNAME=myname USERPROFILE=C:\WINNT\Profiles\myname WIDGETDIR=c:\bin\emacs-20.4\site-lisp\custom windir=C:\WINNT

1 participants
0 comments

Re: [Jan Vroonhof <vroonhof@math.ethz.ch>] Re: constantly crashing in Gnus 25 years, 4 months

Ulrich Drepper

Jan Vroonhof <vroonhof(a)math.ethz.ch> writes: > Ok, if you want you can remove the debug message. Not finding the > input context is not going to hurt. OK, but aren't there resource problems? Do these windows still exist though they are not used? -- ---------------. drepper at gnu.org ,-. 1325 Chesapeake Terrace Ulrich Drepper \ ,-------------------' \ Sunnyvale, CA 94089 USA Cygnus Solutions `--' drepper at cygnus.com `------------------------

1 participants
0 comments

Layout support + widget fixes 25 years, 4 months

Andy Piper

This allows you to layout glyphs according to some predefined strategy within a predefined area, e.g. right justified. You can optionally draw a border round the layout area which can be etched in/out or bevelled in/out. Layouts can be organised within layouts allowing for arbitrarily complex structures. My calculations tell me that if you put a border round sub-layouts you can get a nice table effect (which might interest BillP). This is based on the stuff Ben wrote up in architecting XEmacs, the bugs are all my own. This also fixes a whole bunch of bugs in the subwindow support. I intend applying this tomorrow morning GMT unless 21.2.20 is imminent. andy 1999-08-23 Andy Piper <andy(a)xemacs.org> * event-Xt.c (emacs_Xt_handle_magic_event): use find_matching_subwindow to check whether the exposure is totally inside a subwindow, if it is then ignore it. * glyphs.c (find_matching_subwindow): new function. * gutter-items.el (update-tab-in-gutter): call add-tab-to-gutter here if we don't have one. * toolbar-x.c (x_draw_blank_toolbar_button): use new x_output_shadows signature. (x_output_toolbar_button): use new x_output_x_pixmap signature. * redisplay.h (struct display_box): new structure. (struct display_glyph_area): ditto. declare new display_box functions. * redisplay-x.c (x_output_pixmap): update for new display box stuff. (x_bevel_area): update to allow multiple edge styles and border segments. (x_output_display_block): update for new display_box stuff and signatures. (x_clear_region): ditto. (x_output_x_pixmap): ditto. adjust offsets rather than clip. (console_type_create_redisplay_x): declare new console methods. (x_output_vertical_divider): update for new bevel_area signature. (x_output_shadows): update to allow selective drawing of edges. * redisplay-tty.c (tty_output_display_block): do nothing for layouts. * redisplay-output.c (redisplay_output_subwindow): convert to use display_box structure, this allows us to put subwindows anywhere with any offset, useful for layout glyphs. (redisplay_output_layout): new function. output a layout and its sub-glyphs. (redisplay_output_pixmap): new modular function combining generic parts of the X and mswindows versions. convert to use display_box structures. (redisplay_clear_clipped_region): new function. clear the area a glyph is going into. (redisplay_normalize_glyph_area): new function. calculate the bounds of a display_glyph_area given a display_box. (redisplay_normalize_display_box): new function. shrink a display_box to enclose a display_glyph_area. (redisplay_display_boxes_in_window_p): check whether the input display_box and display_glyph_area are actually in a window. (redisplay_calculate_display_boxes): calculate display boxes based on conventional display_line metrics. (bevel_modeline): update for new bevel_area signature. * redisplay-msw.c (mswindows_output_blank): update for new display_box stuff. (mswindows_output_string): ditto. (mswindows_output_dibitmap): ditto. (mswindows_output_dibitmap_region): ditto. (mswindows_output_pixmap): ditto. (mswindows_bevel_area): update to allow multiple edge styles and border segments. (mswindows_output_display_block): update for new display_box stuff and signatures. (mswindows_clear_region): ditto. (console_type_create_redisplay_mswindows): declare new console methods. * lisp.h (edge_style): new enum. declare display_box and display_glyph_area. * gutter.c (output_gutter): update for new bevel_area signature. * glyphs.h (image_instance_type): add layout. (IMAGE_LAYOUT_MASK): ditto. (LAYOUT_IMAGE_INSTANCEP): new macro. (CHECK_LAYOUT_IMAGE_INSTANCE): ditto. (struct Lisp_Image_Instance): add offsets for layout and the layout type itself. move mask here also. (IMAGE_INSTANCE_LAYOUT_CHILDREN): new macro. (IMAGE_INSTANCE_LAYOUT_BORDER): ditto. (XIMAGE_INSTANCE_LAYOUT_CHILDREN): ditto. (XIMAGE_INSTANCE_LAYOUT_BORDER): ditto. * glyphs.c (mark_image_instance): update for layouts. (print_image_instance): ditto. (image_instance_equal): ditto. (image_instance_hash): ditto. (allocate_image_instance): initialise offsets for layout. (decode_image_instance_type): update for layouts. (encode_image_instance_type): ditto. (Fimage_instance_height): ditto. (Fimage_instance_width): ditto. (allocate_glyph): ditto. (glyph_width): allow image instances as an argument. update for layouts. (glyph_height_internal): ditto. (syms_of_glyphs): add layout symbols. * glyphs-x.h (struct x_image_instance_data): remove mask entry. (IMAGE_INSTANCE_X_MASK): update. * glyphs-x.c (x_label_instantiate): new function. instantiate a label. (image_instantiator_format_create_glyphs_x): initialise new layout glyph type. * glyphs-widget.c: new layout type. (check_valid_orientation): new keyword checker for layouts. (check_valid_justification): ditto. (check_valid_border): ditto. (check_valid_glyph_or_image_list): ditto. (layout_possible_dest_types): new function for layout glyph type. (layout_normalize): ditto. (layout_instantiate): ditto. (syms_of_glyphs_widget): new keywords for layout. (image_instantiator_format_create_glyphs_widget): initialise the layout glyph type. * glyphs-msw.h (struct mswindows_image_instance_data): remove mask argument. (IMAGE_INSTANCE_MSWINDOWS_MASK): update. * glyphs-msw.c: declare layout format. (image_instantiator_format_create_glyphs_mswindows): initialise it. * general.c: new symbols for layouts. * console.h (struct console_methods): new console methods for outputting pixmaps and strings. * lwlib-Xm.c (xm_create_label_field): new function for creating labels. (xm_creation_table): use it. * lwlib-Xaw.c (xaw_create_label_field): new function for creating labels. (xaw_creation_table): use it. -------------------------------------------------------------- Dr Andy Piper Senior Consultant Architect, BEA Systems Ltd

1 participants
0 comments

Re: [SITE] make over finished(?) 25 years, 4 months

Jan Vroonhof

john s jacobs anderson <jacobs(a)azstarnet.com> writes: > I'll be adding more links, etc. over the next week or two, as I work > thru my bookmarks file. At some point, the Elisp packages section is > going to get expanded, with packages categorized by function and > blurb'd as to exactly what they do. I think it might be good to make a clear distinction between the XEmacs packages (i.e. the things you can install using M-x package-admin-add-binary-package) and the more loosely defined "Elisp addons". > If anybody's got a bit of free time this afternoon, they could (a) > pound on all the links and make sure I didn't miss something and (b) > check for obvious mistakes and typos and stuff. What would be nice is a link to the bug tracking system http://cvs.xemacs.org/XEmacs-bugs Well done! Jan

2 participants
1 comments

Re: XEmacs has just (more than) doubled in size 25 years, 4 months

Jan Vroonhof

Andy Piper <andy(a)xemacs.org> writes: > Hmmn, that's irritating. The widget should not be doing this. I'll see if I > can force it off. It seems like something would like. Maybe there should be yet another specifier? gutter-tabs-max-lines? Jan

1 participants
0 comments

Re: [Jan Vroonhof <vroonhof@math.ethz.ch>] Re: constantly crashing in Gnus 25 years, 4 months

Jan Vroonhof

Ulrich Drepper <drepper(a)cygnus.com> writes: > This is on my laptop which has a 1024x768 display with 16 bits. The > Emacs window is at the right side of the screen. Therefore I'd say > these windows are indeed scrollbars. Yes. > Now that the problems are simply ignored Emacs still runs nicely. Ok, if you want you can remove the debug message. Not finding the input context is not going to hurt. > If you have more to test just send it. Thanks, I will when I have a more ore less final solution. I need some time to decide what the optimal implementation is. Jan

1 participants
0 comments

Re: package-get.el : package-get-user-package-location 25 years, 4 months

Jan Vroonhof

sperber(a)informatik.uni-tuebingen.de (Michael Sperber [Mr. Preprocessor]) writes: > >> This looks wrong. Packages go into a package hierarchy, and ~/.xemacs > >> isn't one. It is/was in 21.1. > >> The general concept of having a single package hierarchy > >> into which user packages go is broken by now. Huh? Why? > >> Actually, looking at > >> package-get.el (for the first time), the entire logic associated with > >> `package-get-user-package-location' is complete bogus. Apart from the obvious bug with the missing ~, could you please define bogus? > Well, if you want to really fix the problem, you need to coordinate > the workings of `package-get-user-package-location', > `package-get-maybe-save-index', and `package-get-locate-index-file'. They are coordinated. If you fix the stupid ~ problem and it works. > Especially the latter is totally sloppy and should be rewritten. Please define 'sloppy'. > There's clearly a blatant bug here, but I'd rather leave this eyesore > in to remind people things are really broken here, rather than going > for the quick fix. Please define broken? It al does what it is supposed to (modulo "~") 1. Get the package index from the best location possible. 2. Get the file locally using the standard XEmacs data file algorithms. 3. If the two don't match offer to update the local file if we can write to it. 4. If we cannot write to it, offer to write it to a location where the user is likely to be able to write and where XEmacs will find it again (i.e. the etc directory of the users package hierarchy). Jan

1 participants
0 comments

test, please ignore (sorry) 25 years, 4 months

Didier Verna

1 participants
0 comments

Re: kde-xemacs or gnome-xemacs? 25 years, 4 months

Neal D. Becker

>>>>> "sb" == SL Baur <steve(a)xemacs.org> writes: sb> William Perry <wmperry(a)aventail.com> writes in xemacs-beta(a)xemacs.org: >> SL Baur wrote: >>> Qt uses C++ constructors and >>> destructors and thus is impossible to use under the current XEmacs >>> architecture. >> Why do you say that? sb> Based on past experiences with egcs/the new gcc. Constructor stuff sb> can get called before main() and before we have a chance to take any sb> of our usual precautions. I don't fully understand the technical sb> details, but it was giving me corrupted memory and when I asked one of sb> the now-gcc maintainers about how to turn off global constructors, sb> explaining briefly XEmacs startup & dumping, I was told that dumping sb> was stupid and deserved to lose. This issue only applies to global objects. Global objects are rarely used (and never needed). The design pattern "singleton" is made to address this. There's no big magic to the gcc implementation of global constructors on any architecture I am familiar with. I'm not sure what the details of your issue were, but I'll bet it's not difficult to fix. I have so far kept my $0.02 out of this, but personally if I had to write any GUI I would certainly not want to use a C API. I would much rather use a C++ API, particularly a professionally implemented one like QT.

1 participants
0 comments

Re: package-get.el : package-get-user-package-location 25 years, 4 months

Michael Sperber

>>>>> "Stef" == Stef Epardaud <stef(a)lunatech.com> writes: Stef> On Mon, Aug 23, 1999 at 02:23:00PM +0200, Michael Sperber [Mr. Preprocessor] wrote: >> >> This looks wrong. Packages go into a package hierarchy, and ~/.xemacs >> isn't one. The general concept of having a single package hierarchy >> into which user packages go is broken by now. Actually, looking at >> package-get.el (for the first time), the entire logic associated with >> `package-get-user-package-location' is complete bogus. >> Stef> i agree, it might be that the problem is elsewhere, but the fact is that Stef> package-get.el tries to store the package index into your .xemacs/ Stef> directory. i have nothing against that, except it cannot be and will Stef> never be a directory whose absolute path is "/.xemacs/" because noone Stef> will have a ".xemacs/" at the root of the filesystem. Stef> i agree it should be updating package-index.LATEST.pgp somewhere else Stef> than in ~/.xemacs/etc/package-index.LATEST.pgp but when that someone Stef> wrote the bit where it wanted to store it there, he didn't mean in Stef> /.xemacs/etc/package-index.LATEST.pgp but probably something like Stef> /usr/local/lib/xemacs/xemacs-packages/etc/package-index.LATEST.pgp Stef> it just seemed to me that it aimed for the user's package dir, and Stef> didn't get it correctly, so i helped it. Stef> but i realize that the concept is wrong, and it should aim at the site Stef> directory for packages. it was just probably faster to do what i did, Stef> but wrong i agree. Well, if you want to really fix the problem, you need to coordinate the workings of `package-get-user-package-location', `package-get-maybe-save-index', and `package-get-locate-index-file'. Especially the latter is totally sloppy and should be rewritten. I may get to this next weekend, not before, however. I suspect that the user-local package index really should go into ~/.xemacs/package-index.LATEST.pgp (since it transcends package hierarchies). There's clearly a blatant bug here, but I'd rather leave this eyesore in to remind people things are really broken here, rather than going for the quick fix. -- Cheers =8-} Mike Friede, Völkerverständigung und überhaupt blabla

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 August 1999