September 1998 - XEmacs-Beta - XEmacs Mailing List Archives

Re: FreeBSD 3.0, ELF and XEmacs 27 years

Josh Howard

While there's no actual reference to changes in unexelf, that I can see, in the ChangeLog, it seems that synching that up with Emacs 20.3 lets XEmacs compile and run peachy keen. I suppose I could simply diff out what's changed and pick out the bits which helped this particular problem, but I think that there were enough changes that maybe somebody should just look it over and see what other platforms might benefit from it. SL Baur <steve(a)xemacs.org> writes: > Josh Howard <jrh(a)zeppelin.net> writes in xemacs-beta(a)xemacs.org: > > > FreeBSD 3.0-CURRENT (soon to be -RELEASE) has switched from a.out to > > ELF and XEmacs won't build or run properly. If s/freebsd.h is synched > > up with Emacs 20.3, to use unexelf.c for instance, it will build > > properly, but dumps core when it attempts to run. Moreover, Emacs-20.3 > > builds and runs perfectly fine in the new ELF world of FreeBSD. > > This is with 21.2, though I imagine all versions would have this same > > problem. Here's a backtrace: > > In addition to changing s/freebsd.h, you might also grep the FSF > ChangeLogs for other related changes. Was unexelf.c changed, for > example? > > -- Josh Howard (jrh(a)zeppelin.net) -- http://www.zeppelin.net "One man's religion is another man's belly laugh." -- Robert A. Heinlein

1 participants
0 comments

Re: FreeBSD 3.0, ELF and XEmacs 27 years

SL Baur

Josh Howard <jrh(a)zeppelin.net> writes in xemacs-beta(a)xemacs.org: > FreeBSD 3.0-CURRENT (soon to be -RELEASE) has switched from a.out to > ELF and XEmacs won't build or run properly. If s/freebsd.h is synched > up with Emacs 20.3, to use unexelf.c for instance, it will build > properly, but dumps core when it attempts to run. Moreover, Emacs-20.3 > builds and runs perfectly fine in the new ELF world of FreeBSD. > This is with 21.2, though I imagine all versions would have this same > problem. Here's a backtrace: In addition to changing s/freebsd.h, you might also grep the FSF ChangeLogs for other related changes. Was unexelf.c changed, for example?

1 participants
0 comments

[Success] 21.0 "Irish Goat" XEmacs Lucid on i686-pc-linux 27 years

Hayashi Tanaka

Patched up from beta44. $ rpm -q glibc glibc-2.0.7-19 $ gcc -v Reading specs from /usr/local/lib/gcc-lib/i686-pc-linux-gnu/egcs-2.91.57/specs gcc version egcs-2.91.57 19980901 (egcs-1.1 release) > XEmacs Build Report as generated > by build-report-version 1.35 follows: > Contents of /home/tanaka/src/xemacs-21.0-b55/Installation: > (Output from most recent run of ./configure)

1 participants
0 comments

FreeBSD 3.0, ELF and XEmacs 27 years

Josh Howard

FreeBSD 3.0-CURRENT (soon to be -RELEASE) has switched from a.out to ELF and XEmacs won't build or run properly. If s/freebsd.h is synched up with Emacs 20.3, to use unexelf.c for instance, it will build properly, but dumps core when it attempts to run. Moreover, Emacs-20.3 builds and runs perfectly fine in the new ELF world of FreeBSD. This is with 21.2, though I imagine all versions would have this same problem. Here's a backtrace: (gdb) bt #0 0x8051ccd in allocate_string_chars_struct (string_it_goes_with=0x8398634, fullsize=56) at /jaz/xemacs/xemacs-21.2/src/alloc.c:2123 #1 0x8051e96 in make_uninit_string (length=50) at /jaz/xemacs/xemacs-21.2/src/alloc.c:2156 #2 0x8052412 in make_string ( contents=0xefbfd394 "/jaz/xemacs/xemacs-21.2/objdir/lisp/auto-autoloads", length=50) at /jaz/xemacs/xemacs-21.2/src/alloc.c:2391 #3 0x80d2803 in Fexpand_file_name (name={gu = {type = Lisp_Type_Record, val = 34496901}, s = {bits = 0, val = 68993802}, u = {bits = 0, val = 68993802}, ui = 137987604, i = 137987604, v = 0x8398614, cv = 0x8398614}, default_directory={gu = {type = Lisp_Type_Record, val = 34520969}, s = {bits = 0, val = 69041938}, u = {bits = 0, val = 69041938}, ui = 138083876, i = 138083876, v = 0x83afe24, cv = 0x83afe24}) at /jaz/xemacs/xemacs-21.2/src/fileio.c:1267 #4 0x808b130 in funcall_recording_as (recorded_as={gu = { type = Lisp_Type_Record, val = 34370561}, s = {bits = 0, val = 68741122}, u = {bits = 0, val = 68741122}, ui = 137482244, i = 137482244, v = 0x831d004, cv = 0x831d004}, nargs=2, args=0xefbfd734) at /jaz/xemacs/xemacs-21.2/src/eval.c:3197 #5 0x808b365 in Ffuncall (nargs=3, args=0xefbfd734) at /jaz/xemacs/xemacs-21.2/src/eval.c:3238 #6 0x80612e6 in Fbyte_code (bytestr={gu = {type = Lisp_Type_Record, val = 34190489}, s = {bits = 0, val = 68380978}, u = {bits = 0, val = 68380978}, ui = 136761956, i = 136761956, v = 0x826d264, cv = 0x826d264}, vector={gu = {type = Lisp_Type_Record, val = 34190535}, s = {bits = 0, val = 68381070}, u = {bits = 0, val = 68381070}, ui = 136762140, i = 136762140, v = 0x826d31c, cv = 0x826d31c}, maxdepth={ gu = {type = Lisp_Type_Int_Even, val = 3}, s = {bits = 1, val = 6}, u = { bits = 1, val = 6}, ui = 13, i = 13, v = 0xd, cv = 0xd}) at /jaz/xemacs/xemacs-21.2/src/bytecode.c:416 #7 0x808c3c4 in funcall_lambda (fun={gu = {type = Lisp_Type_Record, val = 34190664}, s = {bits = 0, val = 68381328}, u = {bits = 0, val = 68381328}, ui = 136762656, i = 136762656, v = 0x826d520, cv = 0x826d520}, nargs=0, arg_vector=0xefbfd7e0) at /jaz/xemacs/xemacs-21.2/src/eval.c:3599 #8 0x808bc8b in apply_lambda (fun={gu = {type = Lisp_Type_Record, val = 34190664}, s = {bits = 0, val = 68381328}, u = {bits = 0, val = 68381328}, ui = 136762656, i = 136762656, v = 0x826d520, cv = 0x826d520}, numargs=0, unevalled_args={gu = { type = Lisp_Type_Record, val = 34347009}, s = {bits = 0, val = 68694018}, u = {bits = 0, val = 68694018}, ui = 137388036, i = 137388036, v = 0x8306004, cv = 0x8306004}) at /jaz/xemacs/xemacs-21.2/src/eval.c:3501 #9 0x808ac14 in Feval (form={gu = {type = Lisp_Type_Record, val = 34413513}, s = {bits = 0, val = 68827026}, u = {bits = 0, val = 68827026}, ui = 137654052, i = 137654052, v = 0x8346f24, cv = 0x8346f24}) at /jaz/xemacs/xemacs-21.2/src/eval.c:3085 #10 0x8087f60 in condition_case_1 (handlers={gu = {type = Lisp_Type_Record, val = 34347037}, s = {bits = 0, val = 68694074}, u = {bits = 0, val = 68694074}, ui = 137388148, i = 137388148, v = 0x8306074, cv = 0x8306074}, bfun=0x8089e14 <Feval>, barg={gu = { type = Lisp_Type_Record, val = 34413513}, s = {bits = 0, val = 68827026}, u = {bits = 0, val = 68827026}, ui = 137654052, i = 137654052, v = 0x8346f24, cv = 0x8346f24}, hfun=0x8069140 <cmd_error>, harg={gu = {type = Lisp_Type_Record, val = 34347009}, s = {bits = 0, val = 68694018}, u = {bits = 0, val = 68694018}, ui = 137388036, i = 137388036, v = 0x8306004, cv = 0x8306004}) at /jaz/xemacs/xemacs-21.2/src/eval.c:1653 #11 0x806a158 in top_level_1 (dummy={gu = {type = Lisp_Type_Record, val = 34347009}, s = {bits = 0, val = 68694018}, u = {bits = 0, val = 68694018}, ui = 137388036, i = 137388036, v = 0x8306004, cv = 0x8306004}) at /jaz/xemacs/xemacs-21.2/src/cmdloop.c:206 #12 0x8087b74 in internal_catch (tag={gu = {type = Lisp_Type_Record, val = 34366591}, s = {bits = 0, val = 68733182}, u = {bits = 0, ---Type <return> to continue, or q <return> to quit--- val = 68733182}, ui = 137466364, i = 137466364, v = 0x83191fc, cv = 0x83191fc}, func=0x806a12c <top_level_1>, arg={gu = { type = Lisp_Type_Record, val = 34347009}, s = {bits = 0, val = 68694018}, u = {bits = 0, val = 68694018}, ui = 137388036, i = 137388036, v = 0x8306004, cv = 0x8306004}, threw=0x0) at /jaz/xemacs/xemacs-21.2/src/eval.c:1328 #13 0x8069290 in initial_command_loop (load_me={gu = {type = Lisp_Type_Record, val = 34347009}, s = {bits = 0, val = 68694018}, u = {bits = 0, val = 68694018}, ui = 137388036, i = 137388036, v = 0x8306004, cv = 0x8306004}) at /jaz/xemacs/xemacs-21.2/src/cmdloop.c:285 #14 0x8083a65 in xemacs_21_2_b1_i386_unknown_freebsd3_0 (argc=1, argv=0xefbfdac4, envp=0xefbfdacc, restart=0) at /jaz/xemacs/xemacs-21.2/src/emacs.c:1702 #15 0x808412b in main (argc=1, argv=0xefbfdac4, envp=0xefbfdacc) at /jaz/xemacs/xemacs-21.2/src/emacs.c:2119 #16 0x804e51e in _start () -- Josh Howard (jrh(a)zeppelin.net) -- http://www.zeppelin.net/

1 participants
0 comments

XEmacs 21.0-pre12 "Irish Goat" is released 27 years

SL Baur

Available from the usual places: ftp://ftp.xemacs.org/pub/xemacs/beta/xemacs-21.0/ CVS sources are at http://cvs.xemacs.org/ Installable Lisp packages are in: ftp://ftp.xemacs.org/pub/xemacs/beta/xemacs-21.0/packages/binary-packages/ XEmacs 21.2-pre4 has been committed to the CVS tree. The notable addition is Hrvoje Niksic's implementation of indirect buffers. As many of you are aware of, I'm getting beaten up badly with respect to how much time I have available for XEmacs. Things aren't going to get any better in the short-term. The office that I'm currently utilizing will be going away some time this month, perhaps as early as the 15th worst case. What this basically means in XEmacs-terms, is that if 21.0 is going to get out soonish, it has to go out now. I think the following schedule is doable. The remaining pending XEmacs package changes (I haven't looked at anything recent that has gone through xemacs-patches yet) get folded in over the remainder of the weekend. A final release of 21.0 gets made Monday evening (US/Pacific time), along with all the packages. If we can't hit that, it'll be October at the soonest. I'd just as soon put it off, but it's been put off way too long as it is, and we might as well release what we have and fix all the problems in 21.1. Comments? to 21.0 pre12 "Irish Goat" -- visual package installer from Darryl Okahata -- miscellaneous bug fixes from various people -- miscellaneous fixes from Hrvoje Niksic -- regexp docs from Karl Hegbloom

1 participants
0 comments

Re: Visual bell too loud 27 years

Jason R Mastaler

Hrvoje Niksic <hniksic(a)srce.hr> writes: > I think I might be able to change the code in x_flash() to do what FSF > Emacs does, but I am loathe to do so unless other people on > xemacs.beta convince me that it looks better than the whole screen > flashing. Although the violent, whole screen flashing might be a good deterrent against typing mistakes, I find it overkill and would prefer less of a visible bell as well. Do tell.

1 participants
0 comments

Re: intangible text? 27 years

Hrvoje Niksic

Lars Magne Ingebrigtsen <larsi(a)ifi.uio.no> writes: > I seem to recall someone saying intangible text props now worked in > XEmacs, They don't. I wanted to implement them a year or so ago, but I encountered as many different visions of how `intangible' extent property should work as there were people in the discussion, so I quit. You can try using atomic extents, but I advise against it. -- Hrvoje Niksic <hniksic(a)srce.hr> | Student at FER Zagreb, Croatia --------------------------------+-------------------------------- Union break!

1 participants
0 comments

Visual bell too loud 27 years

Lars Magne Ingebrigtsen

One of the major, major improvements in Emacs 20 is that the visual bell doesn't flash so much of the screen. I have a tendency to `C-g' a lot, and I'm afraid I'm going to go into an epileptic seizure if I use XEmacs much more... (I, like, sit programming mostly in almost complete darkness, using a black background on the monitor. So the white flashing is really *bright*.) Could we have a less visual visual bell in XEmacs? Pretty please? -- (domestic pets only, the antidote for overdose, milk.) larsi(a)ifi.uio.no * Lars Magne Ingebrigtsen

1 participants
0 comments

[PATCH] [TRY#2] Visual package installer & docs 27 years

Darryl Okahata

[ OK, this is a revamped patch. I've tested the broken EFS code, and it definitely works now. ] [ I'm not sure if this should really go into 21.0, but I'll let others decide. ] Here is a patch for B54 that implements the visual package installer (sorry about gzip'ing it, but it's big), with documentation updates. Although it's almost, but not quite, complete, I think it's better to send out what I have now, to get any necessary feedback (and bug reports ;-). ChangeLogs are attached, but the important highlights are: * The function `display-completion-list' was modified to take another optional keyword, `:completion-string', which allows the programmer to change the "Possible completions are:" prompt. While this is a worthwhile change, the reason for making this change is suspect: with this change, it's now possible to use `display-completion-list' as a general-purpose function for cleanly displaying a neatly-formatted list of objects. In the case of the visual package browser/installer, it's being used to format and display a list of packages to install. * Added a new menu pick to access the browser/installer: Options->Customize->List Packages The elisp command is `pui-list-packages', which is not currently bound to any key sequence. * Added ability to install packages from a local disk/CDROM, in addition to using EFS. * In addition to .tar.gz package files, the code will now search for package files with a .tgz extension. * If an existing package is updated with a newer one, the old package lisp directory is deleted. This is done using an all-elisp function for portability, instead of system tools like "rm -rf" or "deltree" or "rmdir /s". * No longer uses "add-big-package.sh". Under Unix, it calls gunzip and tar directly. However, unlike add-big-package.sh, it won't create the top-level package directory. * As a questionable convenience, after installing a package, the code tries to load "auto-autoloads", and updates `load-path' to include the package lisp directory. This shouldn't be a problem, but it needs to be fixed, just in case. * Native MS windows users (not cygwin users) can now set `package-admin-default-install-function' to `package-admin-install-function-mswindows', to install packages. However, this function uses "djtar" to extract .tar.gz packages, and few people have djtar. Currently, the user has to manually change this variable; a future version will automatically do this for the user. Here are the "TODO" items remaining, in no particular order (and I don't know how much time I'll have to do these): 1. Error handling is still lacking in areas. 2. Need to further document package installation (in particular, MS Windows docs are missing). Also need to document some of the faces used by the browser/installer. 3. Need to add a function to download (via EFS) a new copy of package-get-base.el. 4. Need to provide a way to install packages into ~/.xemacs/packages; currently, all packages are installed in the system package directory. 5. Need to automatically select between Unix/cygwin and native MS Windows installation methods. 6. Need to document new keyword for `display-completion-list'. 7. Need to create top-level package directory, if it doesn't already exist. 8. Need to somehow provide an uninstall method. 9. Need to delete more than just the package lisp directory, when updating a package. xemacs-21.0/lisp ChangeLog: =============================================================================== 1998-09-03 Darryl Okahata <darrylo(a)sr.hp.com> * list-mode.el: `display-completion-list': added new/optional keyword `:completion-string', which allows the programmer to change the "Possible completions are:" prompt. * menubar-items.el: Added new pulldown menu-pick to start up the visual package browser/installer: Options->Customize->List Packages * package-admin.el: Added hooks for installing under both Unix and MS Windows. Does additional error checking. No longer calls "add-big-package.sh" to install packages under Unix; now calls gunzip & tar directly. * package-get.el: Added ability to install packages from files on a local disk/CDROM. Now deletes any existing package lisp directory. Does completion on available packages when querying for package names. Will also search for .tgz files in addition for .tar.gz files. Tries to reload auto-autoloads, as a convenience when loading new packages, and also tries to add any new package paths to `load-path'. Changed all occurences of `concat' to use `expand-file-name'. * package-ui.el: New file which implements the main visual package browser/installer, which is started via a menu pick or M-x pui-list packages. =============================================================================== xemacs-21.0/man/xemacs ChangeLog: =============================================================================== 1998-09-03 Darryl Okahata <darrylo(a)sr.hp.com> * packages.texi: Correct and update package documentation. Updated the package installation section to mention the visual package browser/installer. Added a section on manually installing binary packages. =============================================================================== -- Darryl Okahata darrylo(a)sr.hp.com DISCLAIMER: this message is the author's personal opinion and does not constitute the support, opinion, or policy of Hewlett-Packard, or of the little green men that have been following him all day.

2 participants
1 comments

intangible text? 27 years

Lars Magne Ingebrigtsen

I seem to recall someone saying intangible text props now worked in XEmacs, but I've just tried stuff like (put-text-property (1+ (point-min)) (+ (point-min) 6) 'intangible t) and I get no effect whatsoever... -- (domestic pets only, the antidote for overdose, milk.) larsi(a)ifi.uio.no * Lars Magne Ingebrigtsen

1 participants
0 comments

2025

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