2:34 p.m.
* XEmacs 21.4.5 "Civil Service" is released.
"Civil Service" is the sixth in the OXYMORON series.
Relative to XEmacs 21.4.4 "Artificial Intelligence", it contains
critical fixes for the "GC in windows procedure" and "setitimer"
crashes on the MS Windows platform. There are also a few typo
fixes in the documentation, and some minor bugfixes in the user
interface.
Due to the forced move from our CVS server and other infrastructure
issues, this release has been prepared in some haste. It does appear
that the stability on Windows platforms is acceptable, or better.
There are very few changes that affect stability of other platforms,
so they are expected to give excellent stability as usual.
A binary release for Windows with improvements in the installation
process should follow shortly.
More detailed information about the changes is presented below.
The first release in this series, Xemacs 21.4.0 "Solid Vapor",
contained a large number of improvements and extensions to the current
stable version, XEmacs 21.1.14. For more information about the
OXYMORON series, see etc/NEWS, the initial release announcement
http://www.xemacs.org/Releases/21.4.0.html
and the release planning page,
http://www.xemacs.org/Releases/Public-21.2/.
For general information about XEmacs, the developers, and the user
community, see our home page,
http://www.xemacs.org/.
* XEmacs 21.4.5 is "gamma" software.
Besides the usual "no warranty" disclaimer (see etc/WARRANTY, sections
10 and 11), we are now experimenting with a level of stability
intermediate between "beta" and "stable", dubbed "gamma".
At this
point all the developers and most of our beta testers trust the 21.4
code base with all their editing needs.
After several months of testing, XEmacs 21.4.5 now is sufficiently
stable to be recommended to all users. However, there still remain a
few outstanding problems with fontlock that must be fixed before we
retire 21.1.
* Availability
Anonymous ftp:
ftp://ftp.xemacs.org/pub/xemacs/xemacs-21.4
Windows netinstaller:
Andy Piper will make an announcement when it is ready.
Anonymous (pserver) CVS:
The pserver is presently unavailable. The root is likely to
change in the move. We will announce when it is available again.
* Brief summary of Changes to 21.4.5 "Civil Service"
Possibly annoying UI changes:
- (**) configure.ac added to auto-mode-alist.
- (**) Recognize .9 files as man pages (shadowed by ChangeLog and .scm).
- (**) Derive buffers-tab face from the modeline face by default.
- (**) Enable win32 paths under cygwin.
Other UI and API changes:
- Add GTK defaults in several places.
- Printer improvements and fixes for Windows.
- Speed up saving of custom file.
- Improve fontlock handling of unterminated comments.
- Use new politically correct 'report-xemacs-bug' nomenclature in menus.
- DDE improvements for Windows.
Bug fixes:
- Fix those dreaded "GC during window procedure" and setitimer crashes.
- Fix several wrong-type-argument bugs in face handling.
- Fix Lisp syntax in several files.
- Fix pty allocation on SCO Unix.
- call-process-internal reads files as 'binary, now writes 'binary too.
- Mouse-track fixes.
- Fix black borders in tab control widgets.
- Fix progress gauge flashing under OpenMotif.
- Case sensitivity and shell quoting fixes for Windows.
- Multicast data conversion fix.
- Correct handling of dedicated windows when killing buffer.
- Fix progress gauge flashing under OpenMotif.
Build improvements:
- Correct case for XEmacs file names under msw (Cygwin is pedantic).
- Install arch-specific files relative to $execdir, not $prefix.
- New configure.ac guards against autoconf 2.5x, not yet supported.
- Now detects xpm-nox on Cygwin.
- Conditionalize *-msw.c against Cygwin and MSVC5 deficiencies.
- Eliminate a few warnings.
- Detect Cygwin xpm-nox.
- Remove references to XMU from #ifndef HAVE_XMU code.
Other:
- FAQ corrections, typo fixes, other documentation improvements.
- Texinfo syntax corrections in Info sources.
* Full ChangeLogs:
I do not have time to fix up the mailing list archive references for
the ChangeLogs this time. I'm sorry about that, as some of the
changes are quite important.
-------- ChangeLog entries from ChangeLog --------
2001-10-23 Stephen J. Turnbull <stephen(a)xemacs.org>
* XEmacs 21.4.5 "Civil Service" is released.
2001-10-05 Stephen J. Turnbull <stephen(a)xemacs.org>
* configure.in: Document that autoconf 2.5x is unsupported.
2001-10-02 Stephen J. Turnbull <stephen(a)xemacs.org>
* PROBLEMS: configure doesn't support --with-{ld,as}, use environment
variables. Suggested by Goran Koruga <goran.koruga(a)hermes.si>.
2001-09-17 Ben Wing <ben(a)xemacs.org>
* configure.in (TAB):
* configure.in (XE_COMPUTE_RUNPATH):
Don't use -Wshadow when compiling with g++ or you get buried in
silly warnings. This patch was already applied but somehow got
unapplied. Stephen?
2001-08-19 Charles Wilson <cwilson(a)ece.gatech.edu>
* configure.in (for graphics libraries): Detect Cygwin xpm-nox.
2001-08-02 Peter Brown <rendhalver(a)users.sourceforge.net>
Synch options documentation with those in configure.in:
* configure.usage (--with-dialogs): added lucid to list of options
* configure.usage (--with-widgets): added lucid to list of options
* configure.usage (--with-dragndrop): added GTK to list of protocols
* configure.usage (--mail-locking): added `locking' or `mmdf'. to list
of options
2001-09-25 Didier Verna <didier(a)xemacs.org>
* configure.ac: new. Autoconf 2.5x guard.
2001-08-29 Jered Floyd <jered(a)MIT.EDU>
* configure.in (moduledir,sitemoduledir,archlibdir):
Place arch-specific files under ${libdir}, not under ${datadir}.
-------- ChangeLog entries from lib-src/ChangeLog --------
2001-10-23 Stephen J. Turnbull <stephen(a)xemacs.org>
* XEmacs 21.4.5 "Civil Service" is released.
-------- ChangeLog entries from lisp/ChangeLog --------
2001-10-23 Stephen J. Turnbull <stephen(a)xemacs.org>
* XEmacs 21.4.5 "Civil Service" is released.
2001-04-23 Didier Verna <didier(a)xemacs.org>
* cus-edit.el (custom-variable-pre-save): New.
* cus-edit.el (custom-variable-post-save): New.
* cus-edit.el (custom-variable-save): use them.
* cus-edit.el (custom-face-pre-save): New.
* cus-edit.el (custom-face-post-save): New.
* cus-edit.el (custom-face-save): use them.
* cus-edit.el (custom-group-pre-save): New.
* cus-edit.el (custom-group-post-save): New.
* cus-edit.el (custom-group-save): use them.
* cus-edit.el (Custom-save): use the pre/post functions above,
call `custom-save-all' only once.
* cus-edit.el (custom-variable-pre-reset-standard): New.
* cus-edit.el (custom-variable-post-reset-standard): New.
* cus-edit.el (custom-variable-reset-standard): use them.
* cus-edit.el (custom-face-pre-reset-standard): New.
* cus-edit.el (custom-face-post-reset-standard): New.
* cus-edit.el (custom-face-reset-standard): use them.
* cus-edit.el (custom-group-pre-reset-standard): New.
* cus-edit.el (custom-group-post-reset-standard): New.
* cus-edit.el (Custom-reset-standard): use them.
* cus-edit.el (custom-face-reset-saved): use the pre/post
functions above, call `custom-save-all' only once.
2001-04-08 Danny Colascione <qtmstr(a)optonline.net>
* gutter-items.el (buffers-tab face): Derive from the modeline
face by default. Many people use white-on-black for the default
font, and the tabs look hideous with that color combination. The
modeline usually has the correct colors.
2001-09-28 Robert Pluim <rpluim(a)bigfoot.com>
* etags.el (find-tag): Doc fix, point users at pop-tag-mark.
(tag-mark-stack-max): Typo fix.
2001-09-28 Ben Wing <ben(a)xemacs.org>
* font-lock.el:
* font-lock.el (font-lock-after-change-function):
fix problem with last patch, reported by Katsumi Yamaoka.
2001-09-17 Ben Wing <ben(a)xemacs.org>
* font-lock.el (font-lock-after-change-function):
fix problem when you insert a comment on the line before a line of
code: if we use the following char, then when you hit backspace,
the following line of code turns the comment color.
2001-06-08 Ben Wing <ben(a)xemacs.org>
* process.el: comment about shell-command-switch.
* process.el (shell-quote-argument): Need to quote a null
argument, too. From Dan Holmsand.
* startup.el (normal-top-level): Delete some unused FSF junk.
* startup.el (command-line-early):
* startup.el (command-line): Call MS Windows init function.
* win32-native.el:
* win32-native.el (mswindows-system-shells): New.
* win32-native.el (mswindows-system-shell-p): New.
* win32-native.el (init-mswindows-at-startup): New.
* win32-native.el (mswindows-quote-one-vc-runtime-arg):
* win32-native.el (mswindows-quote-one-simple-arg):
* win32-native.el (mswindows-quote-one-command-arg):
* win32-native.el (mswindows-construct-process-command-line-alist):
Correct comments at top. Correctly handle passing arguments
to Cygwin programs and to bash. Fix quoting of zero-length
arguments (from Dan Holmsand). Set shell-command-switch based
on shell-file-name, which in turn comes from env var SHELL.
2001-07-18 Alexey Mahotkin <alexm(a)hsys.msk.ru>
* toolbar.el (toolbar-visible-p, toolbar-captioned-p,
default-toolbar-position): Fill docstrings for balloon-help.
2001-03-15 Stephen J. Turnbull <stephen(a)xemacs.org>
* faces.el: Reenable specifications over 'gtk domain.
2001-08-08 Didier Verna <didier(a)xemacs.org>
* autoload.el (update-autoload-files): make defdir a file name,
not directory name.
2001-08-06 Steve Youngs <youngs(a)xemacs.org>
* menubar-items.el (default-menubar): Use 'report-xemacs-bug'.
* lisp-mnt.el (lm-report-bug): Use 'report-xemacs-bug-beta-address'.
2001-08-01 Edward M. Lee <tailbert(a)yahoo.com>
* font-menu.el (font-menu-set-font): Default to current font-family.
(font-menu-preferred-resolution):
(font-menu-size-scaling):
Provide gtk defaults, too.
2001-07-29 Mike Alexander <mta(a)arbortext.com>
* cus-face.el (custom-set-face-font-size): Last arg to apply is a
list
(custom-set-face-font-family): ditto
2001-07-26 Mike Sperber <mike(a)xemacs.org>
* files.el (auto-mode-alist): .9 files are man pages, too.
2001-07-23 Edwin Steiner <esteiner(a)net4you.at>
* process.el (call-process-internal): bind
coding-system-for-write to 'binary when calling
start-process-internal, so the process receives
the literal contents of INFILE.
2001-09-15 Edward M. Lee <tailbert(a)yahoo.com>
* files.el (auto-mode-alist): Add configure.ac support.
2001-07-19 Adrian Aichner <adrian(a)xemacs.org>
* buff-menu.el:
mouse.el:
Adrian's mouse-track patch. See <y9plq0nc.fsf(a)ecf.teradyne.com>.
-------- ChangeLog entries from lwlib/ChangeLog --------
2001-10-23 Stephen J. Turnbull <stephen(a)xemacs.org>
* XEmacs 21.4.5 "Civil Service" is released.
2001-10-02 Darryl Okahata <darrylo(a)sonic.net>
* If HAVE_XMU was not defined (xmu was not being used), xlwgcs.c
was still referencing xmu functions.
2001-04-08 Danny Colascione <qtmstr(a)optonline.net>
* lwlib-Xlw.c:
xlwtabs.c:
xlwtabs.h:
Force buffer-tab highlight update when moving the buffer, avoiding
an ugly black border, used for keyboard focus indication, on a
widget that can never have keyboard focus, the buffer tab.
2001-07-26 Andy Piper <andy(a)xemacs.org>
Fix progress gauge flashing under OpenMotif:
* lwlib-Xm.c (xm_update_progress): new function. Set Scale height
and width from normal height and width.
* lwlib-Xm.c (xm_create_progress): ditto.
* lwlib-Xm.c (xm_update_one_widget): call xm_update_progress.
* lwlib.c (lw_get_value_arg): new function. Return an argument
based on its name.
* lwlib.h: declare it.
-------- ChangeLog entries from man/ChangeLog --------
2001-10-23 Stephen J. Turnbull <stephen(a)xemacs.org>
* XEmacs 21.4.5 "Civil Service" is released.
2001-10-17 Stephen J. Turnbull <stephen(a)xemacs.org>
* xemacs/text.texi (Outline Visibility): hide-subtree is bound to
C-c C-d, not C-c C-h. Thanks: Keith Browne <kbrowne(a)legato.com>.
2001-07-25 Jim Horning <jim.horning(a)lmco.com>
* xemacs/packages.texi (Package Terminology):
* xemacs/packages.texi (Automatically):
In itemize and enumerate lists ï¼ item should be on its own line.
2001-09-16 Adrian Aichner <adrian(a)xemacs.org>
* xemacs-faq.texi (Q4.0.7): Fix link to VM FAQ thanks to word I
got from Gregory Neil Shapiro.
* xemacs-faq.texi (Q1.3.7): Update moved link.
* xemacs-faq.texi (Q7.0.2): Comment out empty list of bullets to
avoid HTML error in texi2html translation.
2001-09-15 Adrian Aichner <adrian(a)xemacs.org>
* xemacs-faq.texi (Q4.6.1): Infodock is now hosted on SourceForge,
infodock.com is no more.
* xemacs-faq.texi (Q4.7.1): Take auc out of http://sunsite.auc.dk.
2001-09-09 Adrian Aichner <adrian(a)xemacs.org>
* xemacs-faq.texi (Q1.0.9): Correct link to snapshots as suggested
by Robin S. Socha.
-------- ChangeLog entries from netinstall/ChangeLog --------
2001-10-23 Stephen J. Turnbull <stephen(a)xemacs.org>
* XEmacs 21.4.5 "Civil Service" is released.
-------- ChangeLog entries from nt/ChangeLog --------
2001-10-23 Stephen J. Turnbull <stephen(a)xemacs.org>
* XEmacs 21.4.5 "Civil Service" is released.
2001-08-02 Dr. Volker Zell <Dr.Volker.Zell(a)oracle.com>
* xemacs.rc:
Under cygwin, when environment variable CYGWIN is set
to check_case:strict, filenames are considered case sensitive.
-------- ChangeLog entries from src/ChangeLog --------
2001-10-23 Stephen J. Turnbull <stephen(a)xemacs.org>
* XEmacs 21.4.5 "Civil Service" is released.
2001-04-08 Danny Colascione <qtmstr(a)optonline.net>
* glyphs-x.c: Force buffer-tab highlight update when moving the
buffer, avoiding an ugly black border, used for keyboard focus
indication, on a widget that can never have keyboard focus, the
buffer tab.
2001-08-30 Daiki Ueno <ueno(a)unixuser.org>
* process-unix.c (unix_open_multicast_group): Remove too much
conversion between host and network byte order.
2001-09-10 Nix <nix(a)esperi.demon.co.uk>
* src/buffer.c (kill_buffer): Undedicate windows showing BUF
before replacing them.
* src/window.c (window_loop): Implement new UNDEDICATE_BUFFER
operation.
* src/window.c (undedicate_windows): Use it.
* src/window.h: Add prototype.
2001-09-17 Ben Wing <ben(a)xemacs.org>
* fileio.c (normalize_filename):
* fileio.c (Fexpand_file_name):
Fix various C++ compile errors in Andy's recent code.
* callint.c (Fcall_interactively):
* editfns.c (Ftemp_directory):
* editfns.c (Fuser_full_name):
* emacs.c (argmatch):
* lread.c (locate_file_map_suffixes):
* redisplay-x.c (x_ring_bell):
Fix sign-compare warnings.
2001-09-08 Andy Piper <andy(a)xemacs.org>
* fileio.c (normalize_filename): copied from nt.c
* (Ffile_name_directory): enable win32 paths under cygwin.
* (Ffile_name_nondirectory): ditto.
* (directory_file_name): ditto.
* (Fexpand_file_name): ditto.
* (Ffile_truename): ditto.
* (Fsubstitute_in_file_name): ditto.
* (Ffile_name_absolute_p): ditto.
* (Ffile_readable_p): ditto.
2001-09-08 Andy Piper <andy(a)xemacs.org>
* device-msw.c (mswindows_finish_init_device): remove dde
initialization.
* device-msw.c (mswindows_init_dde): factor out from
mswindows_finish_init_device()
* device-msw.c (init_mswindows_very_early): new function
initialize but don't enable dde.
* emacs.c (main_1): call init_mswindows_very_early.
* event-msw.c:
* event-msw.c (mswindows_dde_callback): only execute when we are
ready.
* symsinit.h: declare init_mswindows_very_early.
2001-05-02 Kirill 'Big K' Katsnelson <kkm(a)dtmx.com>
[sjt: This has been in 21.5 for months with no complaints.]
* console.h (struct console_methods): Added flags member.
(CONSOLE_IMPLEMENTATION_FLAGS): Defined.
(CONMETH_IMPL_FLAG):
(CONSOLE_IMPL_FLAG): Macro to check implememntation flags.
Defined XDEVIMPF_DONT_PREEMPT_REDISPLAY.
* device.c (window_system_pixelated_geometry): Use the above macros.
* device.h (DEVICE_IMPL_FLAG): Macro to check a device
implememntation flag.
* device.h (DEVICE_DISPLAY_P): Use it.
* frame.c (delete_frame_internal): Use the above macro.
* redisplay.c (redisplay_device): Use it.
(redisplay_device): Obey XDEVIMPF_DONT_PREEMPT_REDISPLAY.
(redisplay_frame): Ditto.
* device-msw.c (mswindows_device_implementation_flags): Removed.
(msprinter_device_implementation_flags): Removed.
(console_type_create_device_mswindows): Removed references to
implementation_flags methods, set implementation flags here.
(console_type_create_device_mswindows): Added XDEVIMPF_DONT_PREEMPT.
* device-gtk.c (gtk_device_implementation_flags): Removed method.
(console_type_create_device_gtk): Removed method declaration.
Added commented out statement which semantically matches the
commented out statement in the above removed method.
2001-04-14 Gordon Sadler <gbsadler1(a)lcisp.com>
The attached patch fixes a few warnings.
* src/emacs.c: prototype console_type_create_select_gtk for GTK only
* src/frame-gtk.c: guard against GNOME calls in GTK only
* src/redisplay-gtk.c: include <sys/poll.h> #ifdef HAVE_POLL
* src/select-gtk.c: prototype lisp_to_time
2001-06-08 Ben Wing <ben(a)xemacs.org>
------ gc-in-window-procedure fixes ------
* alloc.c:
* alloc.c (struct post_gc_action):
* alloc.c (register_post_gc_action):
* alloc.c (run_post_gc_actions):
* alloc.c (garbage_collect_1):
Create "post-gc actions", to avoid those dreaded "GC during window
procedure" problems.
* event-msw.c:
Abort, clean and simple, when GC in window procedure. We want
to flush these puppies out.
* glyphs-msw.c:
* glyphs-msw.c (finalize_destroy_window):
* glyphs-msw.c (mswindows_finalize_image_instance):
Use a post-gc action when destroying subwindows.
* lisp.h:
Declare register_post_gc_action().
* scrollbar-msw.c:
* scrollbar-msw.c (unshow_that_mofo):
Use a post-gc action when unshowing scrollbar windows, if in gc.
* redisplay.c (mark_redisplay):
Add comment about the utter evilness of what's going down here.
------ cygwin setitimer fixes ------
* Makefile.in.in (sheap_objs):
* Makefile.in.in (profile_objs):
* Makefile.in.in (objs):
Compile profile.c only when HAVE_SETITIMER.
* nt.c (mswindows_sigset):
* nt.c (mswindows_sighold):
* nt.c (mswindows_sigrelse):
* nt.c (mswindows_sigpause):
* nt.c (mswindows_raise):
* nt.c (close_file_data):
Style fixes.
* nt.c:
Move setitimer() emulation to win32.c, because Cygwin needs it too.
* profile.c:
* profile.c (Fstart_profiling):
* profile.c (Fstop_profiling):
Make sure we don't compile if no setitimer(). Use qxe_setitimer()
instead of just plain setitimer().
* signal.c:
* signal.c (set_one_shot_timer):
* signal.c (alarm):
Define qxe_setitimer() as an encapsulation around setitimer() --
call setitimer() directly unless Cygwin or MS Win, in which case
we use our simulated version in win32.c.
* systime.h:
* systime.h (struct itimerval):
* systime.h (ITIMER_REAL):
Prototype mswindows_setitimer() and qxe_setitimer(). Long
comment about "qxe" and the policy regarding encapsulation.
* win32.c:
* win32.c (setitimer_helper_proc):
* win32.c (setitimer_helper_period):
* win32.c (setitimer_helper):
* win32.c (mswindows_setitimer):
Move setitimer() emulation here, so Cygwin can use it.
Rename a couple of functions and variables to be longer and more
descriptive. In setitimer_helper_proc(), send the signal
using either mswindows_raise() or (on Cygwin) kill(). If for
some reason we are still getting lockups, we'll change the kill()
to directly invoke the signal handlers.
------ windows shell fixes ------
* callproc.c:
* ntproc.c:
Comments about how these two files must die.
* callproc.c (init_callproc):
On MS Windows, init shell-file-name from SHELL, then COMSPEC,
not just COMSPEC. (more correct and closer to FSF.) Don't
force a value for SHELL into the environment. (Comments added
to explain why not.)
* nt.c (init_user_info):
Don't shove a fabricated SHELL into the environment. See above.
2001-06-01 Ben Wing <ben(a)xemacs.org>
* Makefile.in.in (ldflags):
eliminate that pesky "defaulting to 00401000" warning.
* nt.h:
eliminate warnings.
2001-06-15 Golubev I. N. <gin(a)mo.msk.ru>
* s/sco5.h (FORCE_ALLOCATE_PTY_THE_OLD_FASHIONED_WAY): #define.
* process-unix.c (allocate_pty): #ifndef "modern" pty allocation.
2001-04-24 Jerry James <james(a)xemacs.org>
* faces.h: pass parameters to Fadd_spec_to_specifier in the
correct order.
2001-06-01 Ben Wing <ben(a)xemacs.org>
* event-msw.c (debug_mswin_messages):
conditionalize the messages that were causing compile errors.
(if only macros could generate #ifdef statements ...)
-------- ChangeLog entries from tests/ChangeLog --------
2001-10-23 Stephen J. Turnbull <stephen(a)xemacs.org>
* XEmacs 21.4.5 "Civil Service" is released.
--
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
6:15 p.m.
After upgrading from 21.4.3 to 21.4.5, it seems some font-locking in
Gnus is now broken. Compare the two Summary buffer snapshots below.
This only seems to happen after I increase the XEmacs font size
(Options --> Font Size --> 14).
Any ideas how I can get the proper highlighting back?
7:25 p.m.
>>>> "Jason" == Jason R Mastaler
<jason(a)mastaler.com> writes:
Jason> This only seems to happen after I increase the XEmacs font
Jason> size (Options --> Font Size --> 14).
Customize problem, I guess.
The possibly relevant changes would seem to be
2001-09-28 Ben Wing &lt;ben(a)xemacs.org&gt;
* font-lock.el:
* font-lock.el (font-lock-after-change-function):
fix problem with last patch, reported by Katsumi Yamaoka.
2001-09-17 Ben Wing &lt;ben(a)xemacs.org&gt;
* font-lock.el (font-lock-after-change-function):
fix problem when you insert a comment on the line before a line of
code: if we use the following char, then when you hit backspace,
the following line of code turns the comment color.
2001-08-01 Edward M. Lee &lt;tailbert(a)yahoo.com&gt;
* font-menu.el (font-menu-set-font): Default to current font-family.
(font-menu-preferred-resolution):
(font-menu-size-scaling):
Provide gtk defaults, too.
2001-07-29 Mike Alexander &lt;mta(a)arbortext.com&gt;
* cus-face.el (custom-set-face-font-size): Last arg to apply is a
list
(custom-set-face-font-family): ditto
Especially the last one. All from 21.4.5, I don't see anything
relevant in 21.4.4. I'll try to replicate the bug tomorrow.
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
11:58 p.m.
Jason R Mastaler <jason(a)mastaler.com> writes:
> This only seems to happen after I increase the XEmacs font
> size (Options --> Font Size --> 14).
"Stephen J. Turnbull" <stephen(a)xemacs.org> writes:
Customize problem, I guess.
I observed problems with font size changes 10+ weeks ago:
| Subject: 21.4.4 doesn't like to change font size?
| Message-ID: <vxkhevbbm94.fsf(a)cinnamon.vanillaknot.com>
| Date: Mon, 13 Aug 2001 18:58:31 -0400
| X-Sent: 10 weeks, 23 hours, 53 minutes, 31 seconds ago
I don't believe anyone paid attention to it at the time.
A quick (and probably wrong, but workable) patch was sent, ~5 weeks later.
| From: Jeff Putsch <putsch(a)fiber.mxim.com>
| To: Karl Kleinpaste <karl(a)charcoal.com>
| Cc: xemacs-beta(a)xemacs.org
| Subject: Re: 21.4.4 doesn't like to change font size?
| Message-ID: <20010920111617.A3266(a)blue.mxim.com>
| Date: Thu, 20 Sep 2001 11:16:17 -0700
| X-Sent: 4 weeks, 5 days, 4 hours, 36 minutes, 56 seconds ago
2:03 a.m.
>>>> "Karl" == Karl Kleinpaste
<karl(a)charcoal.com> writes:
Karl> I observed problems with font size changes 10+ weeks ago:
Karl> I don't believe anyone paid attention to it at the time.
Oh, I paid attention. Messing with the customize hairball is not
something I have time for at present, unless it's actually throwing
errors. And our customize experts all went and got day jobs.
So the whole face customization issue is very much on my mind.
Volunteers to take a close look at would be very welcome.
Karl> A quick (and probably wrong, but workable) patch was sent,
Karl> ~5 weeks later.
Thanks for mentioning it. I'll take a look at it.
Please note: patches that are not sent to <xemacs-patches(a)xemacs.org>
are quite likely to fall on the floor under current circumstances.
N.B. The criterion for sending a patch to xemacs-patches is not
perfection. It's workability. If you think that a patch is
dangerous, poorly understood, and really needs a couple of people to
try it out before even being considered, _don't_ send it to
xemacs-patches. If it works for you, you see no particular hazard in
it, and you think there's positive chance that it can go in as is or
with slight revision, _do_ send it to xemacs-patches.
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
3:43 a.m.
From: "Stephen J. Turnbull" <stephen(a)xemacs.org>
Date: 24 Oct 2001 10:03:25 +0900
>>>> "Karl" == Karl Kleinpaste
<karl(a)charcoal.com> writes:
Karl> I observed problems with font size changes 10+ weeks ago:
Karl> I don't believe anyone paid attention to it at the time.
Oh, I paid attention. Messing with the customize hairball is not
something I have time for at present, unless it's actually throwing
errors. And our customize experts all went and got day jobs.
Which font issue is this? In 21.4.4, there seems to be something
really funky going in with the 'bold&italic faces when you do
something to customize the faces. I had to do reset-face to get it to
accept my customizations.
Anyway, the symptom that I was seeing, was that after calling
custom-set-faces on default, bold, bold-italic, and italic (to set
:size, :family, and :background properties), only default face would
be set properly. Any face that subsequently used the :bold t property
would not be customized properly (eg at all). The size and I think
family were completely ignored. Nothing I did would make the bold
font look right, until I did the reset-face.
So the whole face customization issue is very much on my mind.
Volunteers to take a close look at would be very welcome.
-jeff
4:08 a.m.
>>>> "Jeff" == Jeff Mincy
<jeff(a)delphioutpost.com> writes:
Jeff> Which font issue is this? In 21.4.4, there seems to be
Jeff> something really funky going in with the 'bold&italic faces
Jeff> when you do something to customize the faces. I had to do
Jeff> reset-face to get it to accept my customizations.
There's only one font issue as far as I know. Customize. Customize
has a "user-friendly" model of fonts which does not correspond exactly
to any of the platform models, and a similarly "user-friendly" model
of faces which is quite far removed from the XEmacs model based on
specifiers. Furthermore, it knows nothing of Mule.
If you use the direct interfaces to faces and specifiers, you get what
you ask for (subject to the braindeadness of X11's font handling)---
until Customize decides to reset them. Then, Customize does a lot of
"intelligent" frobbing of faces (to get bold and italic versions of a
given default, for example). But this means that if those functions
aren't working right (and it is hard), you get stuff like:
Jeff> Any face that subsequently used the :bold t property would
Jeff> not be customized properly (eg at all). The size and I think
Jeff> family were completely ignored.
This may be fixed now, as Customize was calling apply on things not
guaranteed to be lists, which also means that it probably was splicing
a list into argument lists where the list itself was intended as a
single argument. Unfortunately, the patch to that is likely to be the
cause of the breakage in Jason's Gnus faces AFAICT.
Customize is very complicated, and simple fixes generally turn out to
be neither simple nor harmless.
Let me repeat:
So the whole face customization issue is very much on my mind.
Volunteers to take a close look at it would be very welcome.
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
9:18 a.m.
Stephen J. Turnbull writes:
>>>>> "Karl" == Karl Kleinpaste
<karl(a)charcoal.com> writes:
Karl> I observed problems with font size changes 10+ weeks ago:
Karl> I don't believe anyone paid attention to it at the time.
Oh, I paid attention. Messing with the customize hairball is not
something I have time for at present, unless it's actually throwing
errors. And our customize experts all went and got day jobs.
So the whole face customization issue is very much on my mind.
Volunteers to take a close look at would be very welcome.
hey steven :)
what needs to be looked at in customisation ??
the whole thing or just the fonts stuff for now ??
ive got some spare time (ie no job right now)
--
-------------------------------------------------------------------------
Peter (Rendhalver)
XEmacs Advocate
Perl Hacker
Apache God
FreeBSD Devote
rendhalver(a)users.sourceforge.net
do not try and bend the spoon, thats impossible.
but try to realise the truth.
and that is?
there is no spoon.
- Neo and child Prodigy
- The Matrix
-------------------------------------------------------------------------
2:45 a.m.
>>>> "Peter" == Peter Brown
<rendhalver(a)users.sourceforge.net> writes:
Peter> the whole thing or just the fonts stuff for now ??
The faces code. IIRC, primarily faces.el (Customize has infested the
lower-level code), cus-face.el, and cust-edit.el.
The primary sticking points right now are
1. Customize is completely unaware of character set registries, for
sure under X and maybe Windows too.
2. Customizations that involve frobbing font specifications (deriving
bold and italic versions in particular) for some reason often fail.
3. Customize generally assumes that it owns the faces, lock, stock
and barrel, and therefore may reset faces any time it wants to. This
caused immense pain to users of non-iso-8859 fonts. The "fix" was to
make Customize more careful about when it does such a reset.
4. My theory is that all of the above are due to the fact that
Customize duplicates many aspects of XEmacs specifiers. Only with the
recent changes mentioned in 3. did Customize become at all aware of
specifier tags, for example.
So for the moment an audit of the correspondence of Customize data
structures and transformations from Customize form to XEmacs native
form seems the way to go.
My personal opinion (which is as yet backed up by zero lines of code,
so take it with a grain of salt) is that Customize should be
reimplemented from the ground up to take advantage of specifiers,
something like
a. Customize should store its specifications in specifiers. It
is claimed that some of the things that Customize can handle are
not supported by specifiers; if so the specifiers themselves may
need to be revamped. That would be for 21.5 and later. But
specifiers are really flexible, and certainly can (very
inefficiently) emulate ordinary variables. So I suspect that may
not be true.
b. Customize's specifications would be marked by a special
specifier tag, say 'custom. (See define-specifier-tag.) This
would be redundand for ordinary variables, but it would allow
Customize to operate directly on variables that are already
specifiers, such as most face properties.
c. This might allow a nearly trivial implementation of Custom
themes by simply adding a theme-specific specifier tag.
With the exception of changing the C-level specifier implementation,
all of this can be done at the Lisp level. So I would suggest that it
be implemented as a new-custom (or pb-custom, if you like :-) package
that can be used separately by 21.4 (and even 21.1), and later be
integrated into the core.
Note that this implies the coexistence of the older custom and
new-custom stuff, because custom.el at least is dumped into the XEmacs
binary by default.
Also, there's no reason to change the Customize UI and high-level
APIs, just the "back end".
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
7:58 a.m.
--On Wednesday, October 24, 2001 3:25 AM +0900 "Stephen J. Turnbull"
<stephen(a)xemacs.org> wrote:
>>>>> "Jason" == Jason R Mastaler
<jason(a)mastaler.com> writes:
Jason> This only seems to happen after I increase the XEmacs font
Jason> size (Options --> Font Size --> 14).
Customize problem, I guess.
The possibly relevant changes would seem to be
2001-09-28 Ben Wing &lt;ben(a)xemacs.org&gt;
* font-lock.el:
* font-lock.el (font-lock-after-change-function):
fix problem with last patch, reported by Katsumi Yamaoka.
2001-09-17 Ben Wing &lt;ben(a)xemacs.org&gt;
* font-lock.el (font-lock-after-change-function):
fix problem when you insert a comment on the line before a line of
code: if we use the following char, then when you hit backspace,
the following line of code turns the comment color.
2001-08-01 Edward M. Lee &lt;tailbert(a)yahoo.com&gt;
* font-menu.el (font-menu-set-font): Default to current font-family.
(font-menu-preferred-resolution):
(font-menu-size-scaling):
Provide gtk defaults, too.
2001-07-29 Mike Alexander &lt;mta(a)arbortext.com&gt;
* cus-face.el (custom-set-face-font-size): Last arg to apply is a
list
(custom-set-face-font-family): ditto
Especially the last one. All from 21.4.5, I don't see anything
relevant in 21.4.4. I'll try to replicate the bug tomorrow.
All that patch did was make the code avoid throwing an error. The error
was caught by something further up the stack so it was never being reported
to the user, but it was definitely not working. if the patch caused
problems it's because some code that was being skipped because the error
was thrown has a bug in it. This is quite possible, of course, but it's
really a downstream bug and not a bug in this patch itself.
By the way, this patch has never been applied to 21.5 since I never got
approval to commit it. I've been running with it for weeks and haven't
seen any problems. I do use font lock a lot and I do change font sizes,
but I don't use the Unix version of XEmacs much.
Mike Alexander <mailto:mtaï¼ arbortext.com>
Arbortext, Inc. +1-734-997-0200
9:07 a.m.
>>>> "Mike" == Mike Alexander
<mta(a)arbortext.com> writes:
Mike> was thrown has a bug in it. This is quite possible, of
Mike> course, but it's really a downstream bug and not a bug in
Mike> this patch itself.
I didn't mean to imply otherwise, just that reverting that patch
and/or examining callers of the patched function would be a start at
localizing the problem.
Mike> By the way, this patch has never been applied to 21.5 since
Yeah, I was hoping Hrvoje or Jan would look at it.
Once this is resolved, I'll find, apply, and commit it, along with any
relevant fixes. Thanks for mentioning that, with Yoshiki out of
service at the moment patch tracking is non-existent.
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
7:59 p.m.
Stephen Turnbull <stephen(a)xemacs.org> writes:
Jason> This only seems to happen after I increase the XEmacs
Jason> font size (Options --> Font Size --> 14).
Customize problem, I guess.
[...]
Especially the last one. All from 21.4.5, I don't see anything
relevant in 21.4.4. I'll try to replicate the bug tomorrow.
The problem doesn't exist in 21.4.4, so it must have crept in between
that and 21.4.5. I'm dropping back to 21.4.4 because this problem is
just too annoying to me, but let me know if you want any patches,
etc. tested.
Cheers.
11:17 a.m.
Hi all
Xemacs compilation is broken with my cygwin install. I'm using cygwin-1.3.2-1.
Is anybody using cygwin-1.3.3.-x and is CBufbyte declared in the new headers ?
I get the following compile error:
gcc -c -g -O3 -Wall -Wno-switch -Winline -Wmissing-prototypes -Wsign-compare -Wshadow
-Wpointer-arith -Demacs -I. -DHAVE_CONFIG_H -Wno-sign-compare -fno-caller-saves
-I/usr/X11R6/include fileio.c
In file included from /usr/include/dirent.h:4,
from sysdir.h:31,
from fileio.c:47:
/usr/include/sys/dirent.h:58: warning: declaration of `select' shadows global
declaration
fileio.c: In function `Fexpand_file_name':
fileio.c:955: `CBufbyte' undeclared (first use in this function)
fileio.c:955: (Each undeclared identifier is reported only once
fileio.c:955: for each function it appears in.)
fileio.c:955: parse error before `)'
What is it good for to enable win32 paths under a cygwin-xemacs ? From the Changelog I can
see this
new feature broke my compilation.
Ciao
Volker
5:21 p.m.
At 12:17 PM 10/24/2001 +0200, Dr. Volker Zell wrote:
Xemacs compilation is broken with my cygwin install. I'm using
cygwin-1.3.2-1.
Is anybody using cygwin-1.3.3.-x and is CBufbyte declared in the new
headers ?
I get the following compile error:
gcc -c -g -O3 -Wall -Wno-switch -Winline -Wmissing-prototypes
-Wsign-compare -Wshadow -Wpointer-arith -Demacs -I. -DHAVE_CONFIG_H
-Wno-sign-compare -fno-caller-saves -I/usr/X11R6/include fileio.c
In file included from /usr/include/dirent.h:4,
from sysdir.h:31,
from fileio.c:47:
/usr/include/sys/dirent.h:58: warning: declaration of `select' shadows
global declaration
fileio.c: In function `Fexpand_file_name':
fileio.c:955: `CBufbyte' undeclared (first use in this function)
fileio.c:955: (Each undeclared identifier is reported only once
fileio.c:955: for each function it appears in.)
fileio.c:955: parse error before `)'
What is it good for to enable win32 paths under a cygwin-xemacs ? From the
Changelog I can see this
new feature broke my compilation.
This is Ben's lossage. I dunno why it compiles for him. In my windows
workspace I just removed the C and that works just fine. The cast is
supposedly only necessary for a C++ compiler anyway.
I am pretty much done with the windows branch of 21.4.5, so look for the
announcement soon.
andy
2:14 a.m.
>>>> "Andy" == Andy Piper <andyp(a)bea.com>
writes:
Andy> This is Ben's lossage.
Wrong. It's mine. Ben probably never builds 21.4 at all these days.
I simply applied your patches, as we agreed. I was not aware that
your 21.4 workspace varied from canonical in that way. How could I
be?
And I don't build under Windows at the moment. The box that used to
be my Windows test bed now serves the CVS repository. I'll be
reinstalling Cygwin and VC++ elsewhere, but it's no longer critical
that release-21-4 build on Windows as windows-21-4 is now canonical
for that platform.
Andy> my windows workspace I just removed the C and that works
Andy> just fine. The cast is supposedly only necessary for a C++
Andy> compiler anyway.
I disagree. The cast is primarily for Mule-aware people reading the
code.
This difference of philosophy and goals is why I rather doubt there
will be a merge of release-21-4 and windows-21-4.
Once again, I don't disagree with your decision to vary from the main
line code that way---Windows is your domain, after all, I'm no expert
---but I certainly resist that kind of change in the Unix line.
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
4:20 a.m.
>>>> "Andy" == Andy Piper <andyp(a)bea.com>
writes:
Andy> This is Ben's lossage.
>>>> "Stephen" == Stephen J Turnbull
<stephen(a)xemacs.org> writes:
Stephen> Wrong. It's mine. Ben probably never builds 21.4 at all
Stephen> these days. I simply applied your patches, as we agreed.
Stephen> I was not aware that your 21.4 workspace varied from
Stephen> canonical in that way. How could I be?
Ah, OK, I see what it is, an incomplete application of one of those
Martin-style lexical-level global-replacements.
Sorry for being so testy.
It's still not something I want to spend much time on tracking. It
would be really helpful if you'd explain what's gone wrong instead of
just saying "Ben done it."
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
4:23 p.m.
At 12:20 PM 10/25/01 +0900, Stephen J. Turnbull wrote:
It's still not something I want to spend much time on tracking.
It
would be really helpful if you'd explain what's gone wrong instead of
just saying "Ben done it."
C'mon Steve. I supplied a working patch and then you broke it in the real
release without *anyone* doing any testing on the changed version. Seems
like a recipe for disaster to me - and I don't really understand why you
did it in the first place, who really cares about C++ compilation?
andy
7:01 p.m.
>>>> "Andy" == Andy Piper <andyp(a)bea.com>
writes:
Andy> At 12:20 PM 10/25/01 +0900, Stephen J. Turnbull wrote:
> It's still not something I want to spend much time on
tracking.
> It would be really helpful if you'd explain what's gone wrong
> instead of just saying "Ben done it."
Andy> C'mon Steve. I supplied a working patch and then you broke
Andy> it in the real release without *anyone* doing any testing on
Andy> the changed version. Seems like a recipe for disaster to me
Andy> - and I don't really understand why you did it in the first
Andy> place, who really cares about C++ compilation?
So? You still knew a lot more than I did and could have saved me a
fair amount of time with a more precise description. There are 550
lines of ChangeLog, I don't remember all the patches in detail.
Instead, I had to wallow through the code to figure out what was going
on. Please?
"Why?" Well, IMO there's no excuse for an _enhancement_ that
introduces warnings in a stable branch. I should have just let it
wait for 21.4.6, but you were in a hurry to get it in.
As for the "disaster," yes, I've screwed up. I wanted a Windows code
freeze, I didn't get it, I tried to manage the code base anyway. I've
made too many mistakes. Fortunately, most resulted in build problems,
so people wasted some time (my apologies) but hopefully no data.
So now we're going to fix that. I just wanted to get this release
out, the branch created, and
"Tag, you're it."
Not an edifying motivation, but the beneficial effect on the Windows
port should be obvious very shortly. Why postpone that? That's the
way I felt about it, unfortunately for 21.4.5.
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
7:09 p.m.
At 03:01 AM 10/26/01 +0900, Stephen J. Turnbull wrote:
So? You still knew a lot more than I did and could have saved me a
fair amount of time with a more precise description. There are 550
lines of ChangeLog, I don't remember all the patches in detail.
Instead, I had to wallow through the code to figure out what was going
on. Please?
I don't understand. What is the problem that you are referring to? Why
didn't the patch cut it?
"Why?" Well, IMO there's no excuse for an _enhancement_
that
introduces warnings in a stable branch. I should have just let it
wait for 21.4.6, but you were in a hurry to get it in.
Actually I was assuming that you weren't going to put it in :)
As for the "disaster," yes, I've screwed up. I wanted a
Windows code
I don't think its a disaster. Kudos to you for actually getting something
out there.
freeze, I didn't get it, I tried to manage the code base anyway.
I've
made too many mistakes. Fortunately, most resulted in build problems,
so people wasted some time (my apologies) but hopefully no data.
I think you did ok.
So now we're going to fix that. I just wanted to get this
release
out, the branch created, and
"Tag, you're it."
:)
Not an edifying motivation, but the beneficial effect on the Windows
port should be obvious very shortly. Why postpone that? That's the
way I felt about it, unfortunately for 21.4.5.
Hey, I'm glad that you did it. It remains to be seen how I will get on.
andy
8:54 p.m.
>>>> "Andy" == Andy Piper <andyp(a)bea.com>
writes:
Andy> I don't understand. What is the problem that you are
Andy> referring to? Why didn't the patch cut it?
["patch"? You mean your instructions to "remove the `C'"? Or did
my
mailer lose something you sent to xemacs-patches?]
All I'm saying is that, especially with respect to the Windows code,
I'm very much a "manager", not a "maintainer." I'm juggling a
lot of
balls at once, and I get confused. It helps a lot if you really spell
things out for me: my "short-term memory" about the context of any
given patch is pretty minimal.
Andy> Hey, I'm glad that you did it. It remains to be seen how I
Andy> will get on.
I'm not ashamed of getting something out, and even in hindsight it
seems like it was the only way to get it done.
But I do regret the number of broken Windows releases.
I'm quite confident you'll do a lot better, because you build daily on
that platform and you know it thoroughly. And the basic decisions
about what goes in will be a lot better, because you don't have the
same biases I have toward stability (ie, "stasis"). The Windows port
will be a lot stronger for it.
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
9:03 p.m.
At 04:54 AM 10/26/01 +0900, Stephen J. Turnbull wrote:
["patch"? You mean your instructions to "remove the
`C'"? Or did my
mailer lose something you sent to xemacs-patches?]
No, the 'C' was *introduced* by a patch on top of my original fileio.c
patch (which did not not have the 'C' obviously). I don't understand why
the patched patch - rather than the original - was necessary.
andy
4:21 p.m.
I don't understand, the patched fileio.c contained changes I didn't submit,
so I assume you got them from somewhere else. It seems risky to do this if
you are not even going to compile them.
andy
At 10:14 AM 10/25/01 +0900, Stephen J. Turnbull wrote:
>>>>> "Andy" == Andy Piper
<andyp(a)bea.com> writes:
Andy> This is Ben's lossage.
Wrong. It's mine. Ben probably never builds 21.4 at all these days.
I simply applied your patches, as we agreed. I was not aware that
your 21.4 workspace varied from canonical in that way. How could I
be?
And I don't build under Windows at the moment. The box that used to
be my Windows test bed now serves the CVS repository. I'll be
reinstalling Cygwin and VC++ elsewhere, but it's no longer critical
that release-21-4 build on Windows as windows-21-4 is now canonical
for that platform.
Andy> my windows workspace I just removed the C and that works
Andy> just fine. The cast is supposedly only necessary for a C++
Andy> compiler anyway.
I disagree. The cast is primarily for Mule-aware people reading the
code.
This difference of philosophy and goals is why I rather doubt there
will be a merge of release-21-4 and windows-21-4.
Once again, I don't disagree with your decision to vary from the main
line code that way---Windows is your domain, after all, I'm no expert
---but I certainly resist that kind of change in the Unix line.
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
9:18 p.m.
From: "Stephen J. Turnbull, XEmacs 21.4 Release Manager"
<stephen(a)xemacs.org>
Date: Tue, 23 Oct 2001 22:34:34 +0900
* Brief summary of Changes to 21.4.5 "Civil Service"
Possibly annoying UI changes:
- (**) Derive buffers-tab face from the modeline face by default.
Actually, I've been meaning on sending mail about this.
I think that the font for the buffers-tab face should be, by default,
a nice condensed variable width font (on windows I'm using "Clarendon
Condensed").
Using a condensed font makes the tabs a lot less wide and makes it
possible to have longer names and or more tabs without sacrificing
readability.
-jeff
1:50 a.m.
>>>> "Jeff" == Jeff Mincy
<jeff(a)delphioutpost.com> writes:
Jeff> I think that the font for the buffers-tab face should be, by
Jeff> default, a nice condensed variable width font (on windows
Jeff> I'm using "Clarendon Condensed").
This seems rather difficult, since "what is nice" and "what is
available" are user and system-dependent. We have a way to
characterize availability of color, but to do font choice at all well
we need an external database of "common" and "desirable" fonts.
Putting it into sample.init.el would be a good idea, though (how about
a patch?)
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
3:18 a.m.
From: Stephen J. Turnbull <stephen(a)xemacs.org>
Date: 24 Oct 2001 09:50:24 +0900
>>>> "Jeff" == Jeff Mincy
<jeff(a)delphioutpost.com> writes:
Jeff> I think that the font for the buffers-tab face should be, by
Jeff> default, a nice condensed variable width font (on windows
Jeff> I'm using "Clarendon Condensed").
This seems rather difficult, since "what is nice" and "what is
available" are user and system-dependent. We have a way to
characterize availability of color, but to do font choice at all well
we need an external database of "common" and "desirable" fonts.
Yes, fonts are non trivial.
Putting it into sample.init.el would be a good idea, though (how about
a patch?)
Do you by chance play tennis?
Yes, I agree (but, see next message).
-jeff
3:28 a.m.
From: Stephen J. Turnbull <stephen(a)xemacs.org>
Date: 24 Oct 2001 09:50:24 +0900
Putting it into sample.init.el would be a good idea, though (how about
a patch?)
You know, the sample.init.el is at the same time, both way too much
and not nearly enough. The sample.init.el file is way too complicated
for a novice type person who just wants to add some keybindings, or
load some packages. On the other hand, there are no examples of face
customizations. There are some good ideas is the sample file, but all
those defuns are scary.
I would suggest that we have a simple init file that goes into a
certain level of detail, does some requires and global-set-keys, maybe
a few useful options.
Then I would suggest that we split the sample.init.el file into
several sample files, with each file being more topic oriented.
And then have a not so simple init file that is split up into several files.
For example, our main init file is something like:
;; Standard Packages that we want all the time.
(require 'vc-hooks)
(require 'font-lock)
(require 'paren)
(if (null paren-mode) (paren-set-mode 'paren))
(require 'redo)
(setq-default uniquify-buffer-name-style 'post-forward-angle-brackets) ;; forward,
reverse,
(require 'uniquify)
(icomplete-mode)
;; preference files split up by topic.
(require 'new-features) ;; nifty new functions
(require 'xemacs-patches) ;; bug fix patches and improvements
(require 'font-preferences) ;; face customizations
(require 'general-preferences) ;; global settings and bindings
(require 'mode-preferences) ;; option settings for various modes
(require 'mailer-preferences) ;; settings for vm, gnus, mailcrypt,
I have the init files under cvs and shared on different machines
(windows & unix) with multiple xemacs releases. Up till now my .emacs
file has always been a complete mess. So far, the only real problems
in switching from one xemacs release to another has been in the face
customizations.
BTW, The sample.init.el has various grep commands - like
'grep-all-files-in-current-directory'. Is there a reason why the
igrep package is not being used? Maybe igrep doesn't do the recursive
grep command?
-jeff
4:15 a.m.
>>>> "Jeff" == Jeff Mincy
<jeff(a)delphioutpost.com> writes:
Jeff> Then I would suggest that we split the sample.init.el file
Jeff> into several sample files, with each file being more topic
Jeff> oriented.
In other words, make a sample.init package. You'd really have to do
that formally, as otherwise managing the sub-libraries would be a mess
for new users.
We also were working on custom themes. This would make it possible to
do a lot of the sample initializations as (more or less) invertible
themes, selectable from the menu.
But we're short on manpower for both of those right now; the guy who
was working on custom themes went and got a day job. Ben, who heavily
revised sample.init.el in the first place, doesn't consider it a
terribly high priority, and he's never made a package himself. So it
will probably remain a monolith until somebody else picks it up and
runs with it.
Jeff> The sample.init.el has various grep commands - like
Jeff> 'grep-all-files-in-current-directory'. Is there a reason
Jeff> why the igrep package is not being used? Maybe igrep
Jeff> doesn't do the recursive grep command?
Maybe it wasn't available when Ben wrote the functions originally. Or
maybe he considers them sufficiently simple and useful to make good
examples despite the existence of the package.
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
2:33 p.m.
From: Stephen J. Turnbull <stephen(a)xemacs.org>
Date: 24 Oct 2001 12:15:27 +0900
>>>> "Jeff" == Jeff Mincy
<jeff(a)delphioutpost.com> writes:
Jeff> Then I would suggest that we split the sample.init.el file
Jeff> into several sample files, with each file being more topic
Jeff> oriented.
In other words, make a sample.init package. You'd really have to do
that formally, as otherwise managing the sub-libraries would be a mess
for new users.
Yep, good point.
We also were working on custom themes. This would make it possible to
do a lot of the sample initializations as (more or less) invertible
themes, selectable from the menu.
But we're short on manpower for both of those right now; the guy who
was working on custom themes went and got a day job. ...
Can you describe custom themes (and current status) in more detail?
Sounds like it could be useful.
One idea that I had is that you almost want an 'interview' mode,
similar to the turbotax interview. Maybe this idea could be mixed in
with the theme idea?
-jeff
2:58 a.m.
>>>> "Jeff" == Jeff Mincy
<jeff(a)delphioutpost.com> writes:
Jeff> Can you describe custom themes (and current status) in more
Jeff> detail? Sounds like it could be useful.
Not very much detail, I'm afraid. Jan Vroonhof was working on it,
there were discussions on some other channel (maybe there is a
Customize mailing list or something), but they petered out a long time
ago I gather. There is some partly working code (but completely
undocumented) in the core Lisp (lisp/cus-edit.el would be the place to
start looking), and a very minimal example theme in etc/custom.
Jeff> One idea that I had is that you almost want an 'interview'
Jeff> mode, similar to the turbotax interview. Maybe this idea
Jeff> could be mixed in with the theme idea?
It might be good for newbies, but I doubt you'd get a lot of
enthusiasm from the old hands. So you'd probably be on your own in
creating such a thing.
A theme, on the other hand, could easily be derived from current
settings by querying all the customized variables. Naively done this
would contain a lot of redundant or even undesirable information, but
that could easily be improved. This would make it easy for people to
experiment with environments "just like" someone else's, while making
it possible to restore one's own easily (and perhaps temporarily)
without restarting XEmacs. So I would guess that this would be useful
to everybody, and probably you'd get more contribution to that.
All IMHO, of course.
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
12:32 p.m.
Stephen J. Turnbull, XEmacs 21.4 Release Manager writes:
* XEmacs 21.4.5 "Civil Service" is released.
"Civil Service" is the sixth in the OXYMORON series.
hey steven
installed fine on my FreeBSD 4.4 box from tarballs using adrians build package
stuffed up my first build report though
i forgot to save the output files
so i sent another one
--
-------------------------------------------------------------------------
Peter (Rendhalver)
XEmacs Advocate
Perl Hacker
Apache God
FreeBSD Devote
rendhalver(a)users.sourceforge.net
do not try and bend the spoon, thats impossible.
but try to realise the truth.
and that is?
there is no spoon.
- Neo and child Prodigy
- The Matrix
-------------------------------------------------------------------------
8220
days inactive
8227
days old
29 comments
9 participants
participants (9)
-
Andy Piper -
Dr. Volker Zell -
Jason R. Mastaler -
Jason R. Mastaler -
Jeff Mincy -
Karl Kleinpaste -
Mike Alexander -
Peter Brown -
Stephen J. Turnbull