Mercurial > hg > xemacs-beta
view src/ChangeLog.GTK @ 665:fdefd0186b75
[xemacs-hg @ 2001-09-20 06:28:42 by ben]
The great integral types renaming.
The purpose of this is to rationalize the names used for various
integral types, so that they match their intended uses and follow
consist conventions, and eliminate types that were not semantically
different from each other.
The conventions are:
-- All integral types that measure quantities of anything are
signed. Some people disagree vociferously with this, but their
arguments are mostly theoretical, and are vastly outweighed by
the practical headaches of mixing signed and unsigned values,
and more importantly by the far increased likelihood of
inadvertent bugs: Because of the broken "viral" nature of
unsigned quantities in C (operations involving mixed
signed/unsigned are done unsigned, when exactly the opposite is
nearly always wanted), even a single error in declaring a
quantity unsigned that should be signed, or even the even more
subtle error of comparing signed and unsigned values and
forgetting the necessary cast, can be catastrophic, as
comparisons will yield wrong results. -Wsign-compare is turned
on specifically to catch this, but this tends to result in a
great number of warnings when mixing signed and unsigned, and
the casts are annoying. More has been written on this
elsewhere.
-- All such quantity types just mentioned boil down to EMACS_INT,
which is 32 bits on 32-bit machines and 64 bits on 64-bit
machines. This is guaranteed to be the same size as Lisp
objects of type `int', and (as far as I can tell) of size_t
(unsigned!) and ssize_t. The only type below that is not an
EMACS_INT is Hashcode, which is an unsigned value of the same
size as EMACS_INT.
-- Type names should be relatively short (no more than 10
characters or so), with the first letter capitalized and no
underscores if they can at all be avoided.
-- "count" == a zero-based measurement of some quantity. Includes
sizes, offsets, and indexes.
-- "bpos" == a one-based measurement of a position in a buffer.
"Charbpos" and "Bytebpos" count text in the buffer, rather than
bytes in memory; thus Bytebpos does not directly correspond to
the memory representation. Use "Membpos" for this.
-- "Char" refers to internal-format characters, not to the C type
"char", which is really a byte.
-- For the actual name changes, see the script below.
I ran the following script to do the conversion. (NOTE: This script
is idempotent. You can safely run it multiple times and it will
not screw up previous results -- in fact, it will do nothing if
nothing has changed. Thus, it can be run repeatedly as necessary
to handle patches coming in from old workspaces, or old branches.)
There are two tags, just before and just after the change:
`pre-integral-type-rename' and `post-integral-type-rename'. When
merging code from the main trunk into a branch, the best thing to
do is first merge up to `pre-integral-type-rename', then apply the
script and associated changes, then merge from
`post-integral-type-change' to the present. (Alternatively, just do
the merging in one operation; but you may then have a lot of
conflicts needing to be resolved by hand.)
Script `fixtypes.sh' follows:
----------------------------------- cut ------------------------------------
files="*.[ch] s/*.h m/*.h config.h.in ../configure.in Makefile.in.in ../lib-src/*.[ch] ../lwlib/*.[ch]"
gr Memory_Count Bytecount $files
gr Lstream_Data_Count Bytecount $files
gr Element_Count Elemcount $files
gr Hash_Code Hashcode $files
gr extcount bytecount $files
gr bufpos charbpos $files
gr bytind bytebpos $files
gr memind membpos $files
gr bufbyte intbyte $files
gr Extcount Bytecount $files
gr Bufpos Charbpos $files
gr Bytind Bytebpos $files
gr Memind Membpos $files
gr Bufbyte Intbyte $files
gr EXTCOUNT BYTECOUNT $files
gr BUFPOS CHARBPOS $files
gr BYTIND BYTEBPOS $files
gr MEMIND MEMBPOS $files
gr BUFBYTE INTBYTE $files
gr MEMORY_COUNT BYTECOUNT $files
gr LSTREAM_DATA_COUNT BYTECOUNT $files
gr ELEMENT_COUNT ELEMCOUNT $files
gr HASH_CODE HASHCODE $files
----------------------------------- cut ------------------------------------
`fixtypes.sh' is a Bourne-shell script; it uses 'gr':
----------------------------------- cut ------------------------------------
#!/bin/sh
# Usage is like this:
# gr FROM TO FILES ...
# globally replace FROM with TO in FILES. FROM and TO are regular expressions.
# backup files are stored in the `backup' directory.
from="$1"
to="$2"
shift 2
echo ${1+"$@"} | xargs global-replace "s/$from/$to/g"
----------------------------------- cut ------------------------------------
`gr' in turn uses a Perl script to do its real work,
`global-replace', which follows:
----------------------------------- cut ------------------------------------
: #-*- Perl -*-
### global-modify --- modify the contents of a file by a Perl expression
## Copyright (C) 1999 Martin Buchholz.
## Copyright (C) 2001 Ben Wing.
## Authors: Martin Buchholz <martin@xemacs.org>, Ben Wing <ben@xemacs.org>
## Maintainer: Ben Wing <ben@xemacs.org>
## Current Version: 1.0, May 5, 2001
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2, or (at your option)
# any later version.
#
# This program is distributed in the hope that it will be useful, but
# WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
# General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with XEmacs; see the file COPYING. If not, write to the Free
# Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA
# 02111-1307, USA.
eval 'exec perl -w -S $0 ${1+"$@"}'
if 0;
use strict;
use FileHandle;
use Carp;
use Getopt::Long;
use File::Basename;
(my $myName = $0) =~ s@.*/@@; my $usage="
Usage: $myName [--help] [--backup-dir=DIR] [--line-mode] [--hunk-mode]
PERLEXPR FILE ...
Globally modify a file, either line by line or in one big hunk.
Typical usage is like this:
[with GNU print, GNU xargs: guaranteed to handle spaces, quotes, etc.
in file names]
find . -name '*.[ch]' -print0 | xargs -0 $0 's/\bCONST\b/const/g'\n
[with non-GNU print, xargs]
find . -name '*.[ch]' -print | xargs $0 's/\bCONST\b/const/g'\n
The file is read in, either line by line (with --line-mode specified)
or in one big hunk (with --hunk-mode specified; it's the default), and
the Perl expression is then evalled with \$_ set to the line or hunk of
text, including the terminating newline if there is one. It should
destructively modify the value there, storing the changed result in \$_.
Files in which any modifications are made are backed up to the directory
specified using --backup-dir, or to `backup' by default. To disable this,
use --backup-dir= with no argument.
Hunk mode is the default because it is MUCH MUCH faster than line-by-line.
Use line-by-line only when it matters, e.g. you want to do a replacement
only once per line (the default without the `g' argument). Conversely,
when using hunk mode, *ALWAYS* use `g'; otherwise, you will only make one
replacement in the entire file!
";
my %options = ();
$Getopt::Long::ignorecase = 0;
&GetOptions (
\%options,
'help', 'backup-dir=s', 'line-mode', 'hunk-mode',
);
die $usage if $options{"help"} or @ARGV <= 1;
my $code = shift;
die $usage if grep (-d || ! -w, @ARGV);
sub SafeOpen {
open ((my $fh = new FileHandle), $_[0]);
confess "Can't open $_[0]: $!" if ! defined $fh;
return $fh;
}
sub SafeClose {
close $_[0] or confess "Can't close $_[0]: $!";
}
sub FileContents {
my $fh = SafeOpen ("< $_[0]");
my $olddollarslash = $/;
local $/ = undef;
my $contents = <$fh>;
$/ = $olddollarslash;
return $contents;
}
sub WriteStringToFile {
my $fh = SafeOpen ("> $_[0]");
binmode $fh;
print $fh $_[1] or confess "$_[0]: $!\n";
SafeClose $fh;
}
foreach my $file (@ARGV) {
my $changed_p = 0;
my $new_contents = "";
if ($options{"line-mode"}) {
my $fh = SafeOpen $file;
while (<$fh>) {
my $save_line = $_;
eval $code;
$changed_p = 1 if $save_line ne $_;
$new_contents .= $_;
}
} else {
my $orig_contents = $_ = FileContents $file;
eval $code;
if ($_ ne $orig_contents) {
$changed_p = 1;
$new_contents = $_;
}
}
if ($changed_p) {
my $backdir = $options{"backup-dir"};
$backdir = "backup" if !defined ($backdir);
if ($backdir) {
my ($name, $path, $suffix) = fileparse ($file, "");
my $backfulldir = $path . $backdir;
my $backfile = "$backfulldir/$name";
mkdir $backfulldir, 0755 unless -d $backfulldir;
print "modifying $file (original saved in $backfile)\n";
rename $file, $backfile;
}
WriteStringToFile ($file, $new_contents);
}
}
----------------------------------- cut ------------------------------------
In addition to those programs, I needed to fix up a few other
things, particularly relating to the duplicate definitions of
types, now that some types merged with others. Specifically:
1. in lisp.h, removed duplicate declarations of Bytecount. The
changed code should now look like this: (In each code snippet
below, the first and last lines are the same as the original, as
are all lines outside of those lines. That allows you to locate
the section to be replaced, and replace the stuff in that
section, verifying that there isn't anything new added that
would need to be kept.)
--------------------------------- snip -------------------------------------
/* Counts of bytes or chars */
typedef EMACS_INT Bytecount;
typedef EMACS_INT Charcount;
/* Counts of elements */
typedef EMACS_INT Elemcount;
/* Hash codes */
typedef unsigned long Hashcode;
/* ------------------------ dynamic arrays ------------------- */
--------------------------------- snip -------------------------------------
2. in lstream.h, removed duplicate declaration of Bytecount.
Rewrote the comment about this type. The changed code should
now look like this:
--------------------------------- snip -------------------------------------
#endif
/* The have been some arguments over the what the type should be that
specifies a count of bytes in a data block to be written out or read in,
using Lstream_read(), Lstream_write(), and related functions.
Originally it was long, which worked fine; Martin "corrected" these to
size_t and ssize_t on the grounds that this is theoretically cleaner and
is in keeping with the C standards. Unfortunately, this practice is
horribly error-prone due to design flaws in the way that mixed
signed/unsigned arithmetic happens. In fact, by doing this change,
Martin introduced a subtle but fatal error that caused the operation of
sending large mail messages to the SMTP server under Windows to fail.
By putting all values back to be signed, avoiding any signed/unsigned
mixing, the bug immediately went away. The type then in use was
Lstream_Data_Count, so that it be reverted cleanly if a vote came to
that. Now it is Bytecount.
Some earlier comments about why the type must be signed: This MUST BE
SIGNED, since it also is used in functions that return the number of
bytes actually read to or written from in an operation, and these
functions can return -1 to signal error.
Note that the standard Unix read() and write() functions define the
count going in as a size_t, which is UNSIGNED, and the count going
out as an ssize_t, which is SIGNED. This is a horrible design
flaw. Not only is it highly likely to lead to logic errors when a
-1 gets interpreted as a large positive number, but operations are
bound to fail in all sorts of horrible ways when a number in the
upper-half of the size_t range is passed in -- this number is
unrepresentable as an ssize_t, so code that checks to see how many
bytes are actually written (which is mandatory if you are dealing
with certain types of devices) will get completely screwed up.
--ben
*/
typedef enum lstream_buffering
--------------------------------- snip -------------------------------------
3. in dumper.c, there are four places, all inside of switch()
statements, where XD_BYTECOUNT appears twice as a case tag. In
each case, the two case blocks contain identical code, and you
should *REMOVE THE SECOND* and leave the first.
| author | ben |
|---|---|
| date | Thu, 20 Sep 2001 06:31:11 +0000 |
| parents | 0784d089fdc9 |
| children | aa729daae5e2 |
line wrap: on
line source
2000-10-03 William M. Perry <wmperry@aventail.com> * objects-gtk.c (gtk_font_instance_truename): Make sure we get the fully expanded version of the font. * device-gtk.c (convert_font): Ditto. * gtk-xemacs.c (convert_font): Tell __get_gtk_font_truename to not expand wildcards. * objects-gtk.c (__get_gtk_font_truename): Use the internal name-list in a GdkFont structure to find the truename of the font. This protects us from crashing if we get a FontSet instead of a Font. (__get_gtk_font_truename): Accept new argument 'expandp' for whether to return the FULL font name or the wildcarded version. 2000-09-21 William M. Perry <wmperry@aventail.com> * device-gtk.c (Fgtk_init): Moved calls to gtk_init or gnome_init to separate function, and expose it to lisp. It is now possible to create GTK applications from batch mode. (gtk_init_device): Use the new function. 2000-09-12 William M. Perry <wmperry@aventail.com> * gtk-glue.c (gdk_event_to_emacs_event): Special case double and triple clicks when converting events to lisp objects. This allows something like GtkCList to treat double-clicks differently in the 'select_row' signal. 2000-09-11 William M. Perry <wmperry@aventail.com> * menubar-gtk.c (menu_create_menubar): Set a special name for GtkMenuItems directly in the menubar. 2000-09-10 William M. Perry <wmperry@aventail.com> * gtk-xemacs.c (gtk_xemacs_size_request): Deal with frame being NULL. * gtk-xemacs.c (gtk_xemacs_size_allocate): Ditto. 2000-09-09 William M. Perry <wmperry@aventail.com> * sound.c (init_native_sound): Enable sound for GTK devices. * device-gtk.c (gtk_init_device): Attempt to load a default gtkrc file from the data directory. This way we can enable the default face font handling in gtk-faces.el but not screw the majority of users with a proportional font by default. * device-gtk.c (gtk_init_device): Attempt to load ~/.xemacs/gtk-options.el when GTK devices are created. This allows for setting a persistent geometry without requiring GNOME. * gtk-xemacs.c (gtk_xemacs_style_set): Deal with NULL frame. * device-gtk.c (gtk_init_device): Make app_shell a GtkXEmacs widget so that style information is retrieved correctly. * menubar-gtk.c (gtk_xemacs_menubar_get_type): New subclass of GtkMenuBar that always requests the same width as the text widget. This fixes the spurious frame resizes when there were too many menu items to display in the desired width. (create_menubar_widget): Use the new subclass. 2000-09-08 William M. Perry <wmperry@aventail.com> * device-gtk.c (Fgtk_keysym_on_keyboard_p): Ported function from the X side of things. * device-gtk.c (gtk_mark_device): Make sure that we mark the keysym hashtable or things go boom. 2000-09-07 William M. Perry <wmperry@aventail.com> * menubar-gtk.c (gtk_update_frame_menubars): Don't actually update the menubars if the menu is up. This fixes the weird problem where if you had a menu up and 'message' was called (this happens a lot with the 'customize' menu), the menu would disappear. This is because XEmacs is fairly lame about when it updates the menus - the message logging code eventually does a (save-excursion (set-buffer " *Message Log*") ...). The set-buffer caused magic variable current-menubar to be evaluated, which calls gtk_update_frame_menubars, which would nuke the menus. Gack. 2000-09-06 William M. Perry <wmperry@aventail.com> * event-gtk.c (gtk_event_to_emacs_event): Reworked how we handle removing the shift modifier of normal keys (a & A, etc) to be more like MS-windows. This makes everything work pretty happily with query-replace and apropos now. 2000-09-05 William M. Perry <wmperry@aventail.com> * select-gtk.c (emacs_gtk_selection_received): Signal a fake event here so that the event loop will wake up. Should fix the strange pause seen when pasting. * select-gtk.c (Fgtk_get_clipboard): Signal an error if no selections are available. This is more meaningful than 'insert' throwing an error when it gets 'nil'. * select-gtk.c (emacs_gtk_selection_received): Don't bother checking whether the data returned as the selection is a string. If it is not, we convert it to binary data anyway. This fixes the bug where you could not paste between two separate XEmacs instances w/mule enabled (it sends selections as COMPOUND_TEXT). * device-gtk.c (Fgtk_style_info): Return the default font name as part of the style info. * menubar-gtk.c (__generic_button_callback): make sure that we provide a channel for our menu events. This fixes things like get-dialog-box-response that rely on event-channel !nilp 2000-09-05 William M. Perry <wmperry@aventail.com> * glyphs-gtk.c (__downcase): Actually return the downcased string! Thanks to Michael Altenhofen <Michael.Altenhofen@sap.com> for spotting this. 2000-09-05 William M. Perry <wmperry@aventail.com> * menubar-gtk.c (gtk_popup_menu): Make sure we call __activate_menu correctly if the menu is dynamic. This fixes popup menus with :filter effects. 2000-09-01 William M. Perry <wmperry@aventail.com> * gpmevent.c, gpmevent.h: Updated to the latest XEmacs 21.2 version of gpmevent.c. This means that GPM support works with GTK now. * console-tty.c, console-tty.h: Removed old GPM support. * device-tty.c, emacs.c, event-Xt.c: Ditto. * event-unixoid.c, frame-tty.c: Ditto. 2000-08-30 William M. Perry <wmperry@aventail.com> * gtk-xemacs.c (smash_scrollbar_specifiers): Don't bother looking for nextstep themed scrollbars - didn't work anyway. * glade.c (Fglade_xml_textdomain): deal with old versions of Glade that use 'textdomain' instead of 'txtdomain' in the structure. * menubar-gtk.c (gtk_popup_menu): use gtk_widget_show instead of gtk_widget_show_all so that the magic `space-saver' menu item does not get shown on popups. 2000-08-27 William M. Perry <wmperry@aventail.com> * scrollbar-gtk.c (gtk_update_scrollbar_instance_status): Fiddle with scrollbar_values -> GtkAdjustment conversion and scrolling up with the arrows works correctly now. * event-gtk.c (gtk_event_to_emacs_event): Fixed the shifted key lossage (pc-select works now). 2000-08-11 William M. Perry <wmperry@aventail.com> * scrollbar-gtk.c (scrollbar_cb): Need to make sure we look at the appropriate scrollbar instances from the mirror. Was looking only at the vertical scrollbar instance. Don't know if this has anything to do with the weird scrolling behaviour, but it is worth a shot. 2000-07-26 William M. Perry <wmperry@aventail.com> * menubar-gtk.c (run_menubar_hook): New function that runs activate-menubar-hook if the menu shell it was invoked by is currently inactive. (create_menubar_widget): Hook up the button-press-event to run_menubar_hook to cater to broken packages. 2000-07-22 William M. Perry <wmperry@aventail.com> * frame-gtk.c (gtk_popup_frame): When the window is supposed to be initially unmapped, we need to make sure we realize the text widget (to avoid lossage in redisplay assuming there is a valid window), but NOT show it. 2000-07-12 William M. Perry <wmperry@aventail.com> * menubar-gtk.c (__kill_stupid_gtk_timer): New function to kill the timer on a GtkMenuItem widget. This timer and its callback appear to be what was causing heavily filtered menus (like customize and other things under 'options') to crash. The GTK code installs a timer when the user is moving the mouse around to scan menus. Submenus are only popped up when this timer expires. But if the filters are constantly running and creating/destroying submenus, then you can blow up when they unconditionally check GTK_WIDGET_FLAGS(menu_item->submenu), when submenu is NULL. 2000-07-11 William M. Perry <wmperry@aventail.com> * device-gtk.c (gtk_init_device): Can now pass the entire argv array that is in gtk-initial-argv-list, since gtk-init filters out unknown options that may cause GTK to puke for us. This means that GNOME session management works now. * frame-gtk.c (gnome_parse_geometry): Ripped this out of the GNOME libraries to parse geometry settings, in case the user did not compile with GNOME. (gtk_initialize_frame_size): If gtk_initial_geometry is !NILP, then try to parse it and use that. If the geometry cannot be parsed, fall back to using 80x30. * device-gtk.c (Vgtk_initial_geometry): New variable exposed to lisp holding the desired geometry of initial frames. 2000-07-09 William M. Perry <wmperry@aventail.com> * ui-gtk.c (Fgtk_call_function): Outlined possible `enhancement' if someone calls gtk-call-function with too few arguments. After I implemented it I realized it was probably a bad idea, so I commented it out. Maybe for the future... * menubar-gtk.c (menu_convert): Can now pass in a GtkWidget to reuse. It detaches submenus, cleans up any GCPROs on it, and attaches a new submenu. All done in an effort to stop the menubar flickering. (menu_create_menubar): No longer willy-nilly deletes all of the items in the top-level menubar. We now check to see if the widget already existing at the desired position has the same label as this menu. If it does, we pass it in to menu_convert. This drastically reduces the flickering menu problem that most people have been seeing (try speedbar now). 2000-07-04 William M. Perry <wmperry@aventail.com> * event-gtk.c (gtk_event_to_emacs_event): If FRAME is NULL, then default to the selected frame on Vdefault_gtk_device. This will only happen when being called from gtk-glue.c * ui-gtk.c (gtk_type_to_lisp): Properly convert GTK_TYPE_GDK_EVENT objects to lisp events. * event-gtk.c (gtk_event_to_emacs_event): Made this non-static so that gtk-glue.c can use it. * gtk-glue.c (gdk_event_to_emacs_event): New function to convert a GDK event into something sensible for lisp. Just uses gtk_event_to_emacs_event() from event-gtk.c to avoid code duplication. Not perfect (the channel is set to the selected frame, not the GtkObject it actually happened on). * event-gtk.c (gtk_event_to_emacs_event): Finally fixed the weird selection problem where if you released the mouse button over the root window, XEmacs would get confused and still think the selection was going on. * ui-gtk.c (Fgtk_describe_type): New function to return the signals and magic properties of an object given its name or type identifier. 2000-07-03 William M. Perry <wmperry@aventail.com> * ui-byhand.c (Fgtk_ctree_recurse): New function gtk-ctree-recurse that encompasses gtk_ctree_post_recursive, gtk_ctree_pre_recursive, gtk_ctree_post_recursive_to_depth, and gtk_ctree_pre_recursive_to_depth. All hail lisp bindings with variable number of arguments. (Fgtk_ctree_recurse): Allow nil for the node. * ui-gtk.c (emacs_gtk_boxed_equality): New function for comparing two GtkBoxed objects. This is needed because there is no way to store a `user_data' or equivalent on them, so we cannot resurrect it like a GtkObject so we always get the same Lisp_Object. This allows callbacks to use `equal' on two GtkBoxed types and get sane results. (emacs_gtk_boxed_hash): New function for hashing GtkBoxed objects. 2000-07-01 William M. Perry <wmperry@aventail.com> * glade.c: New file to implement glue code for libglade. 2000-06-30 William M. Perry <wmperry@aventail.com> * ui-gtk.c (lisp_to_gtk_type): Know how to deal with GTK_TYPE_GDK_GC. * gtk-glue.c: Need to defien a GTK type for GdkGC so that we can import all the GDK drawing primitives. (face_to_gc): New function to convert a face object to a GC. 2000-06-27 William M. Perry <wmperry@aventail.com> * ui-gtk.c (Fgtk_import_variable_internal): Renamed to -internal. 2000-06-25 Vladimir Vukicevic <vladimir@helixcode.com> * frame-gtk.c (gtk_set_initial_frame_size): Added in a call to gtk_window_set_policy so that you can resize the window down below the minimum size of the menubar. 2000-06-23 William M. Perry <wmperry@aventail.com> * emacs.c (Fkill_emacs): Make sure we clean up the .saves* files on normal exit. 2000-06-13 William M. Perry <wmperry@aventail.com> * menubar-gtk.c (__activate_filtermenu): Put in some of the same protections for filter menus that the normal menus have for widget deletion and selected menu items, etc. 2000-06-12 William M. Perry <wmperry@aventail.com> * scrollbar-gtk.c (gtk_create_scrollbar_instance): hide a scrollbar widget until we are absolutely sure we need to see it. Fixes the problem we were seeing with mouse-avoidance-mode causing severe scrollbar breakage. (update_one_widget_scrollbar_pointer): Move the call to gtk_widget_realize() here instead of in the upper code. Isolates the dependency-on-having-a-window logic. * gtk-xemacs.c (smash_scrollbar_specifiers): When setting the scrollbar specifiers, need to take the x/y thickness of the border into account. Horizontal scrollbar placement is much much nicer now. Some themes would cause them to get positioned partially on top of the modeline. 2000-06-08 William M. Perry <wmperry@aventail.com> * console.c (select_console_1): Duh, forgot to put in an 'else' statement, so Vwindow_system was getting set to Qnil all the time. *sigh*. 2000-06-02 William M. Perry <wmperry@aventail.com> * glyphs-gtk.c (gtk_xpm_instantiate): Do not allow pixmaps to be instantiated as pointers under GTK. The pixmap and cursor routines under GDK do not expose enough information to let use do the same magic that glyphs-x.c does. *sigh* * ui-byhand.c (Fgtk_toolbar_insert_item): Hand-rolled function. (Fgtk_toolbar_prepend_item): Ditto. (generic_toolbar_insert_item): Utility function to take care of all the common code for the Fgtk_toolbar_*_item functions. 2000-06-01 William M. Perry <wmperry@aventail.com> * gtk-glue.c (face_to_style): DOH! You should only load it if IMAGE_INSTANCEP, not !IMAGE_INSTANCEP you doofus. * ui-byhand.c (Fgtk_toolbar_append_item): Hand-rolled function. * ui-gtk.c (Fgtk_import_function_internal): No longer need to use dll_function() to get the marshallers. They are now stored in a hashtable. Lookups should be a bit quicker, and it will work on platforms where you cannot do a dll_open (NULL) to look at your own symbol table. 2000-05-31 William M. Perry <wmperry@aventail.com> * select-gtk.c (emacs_gtk_selection_handle): Better MULE string handling. * gtk-xemacs.c (gtk_xemacs_realize): Make sure we set the style on the widget from the realize function. Otherwise for some themes the color slots are not allocated yet, and we end up with icky looking colors for things like the modeline/toolbar. * select-gtk.c (Fgtk_get_clipboard): If we cannot get the selection from anyone, return the last selection we received. This make us work more like the X selection behaviour with cutbuffers enabled. 2000-05-30 William M. Perry <wmperry@aventail.com> * ui-byhand.c: Removed definitions of gtk-clist-prepend, gtk-clist-append, and gtk-clist-insert. * ui-gtk.c (lisp_to_gtk_type): Use it. * gtk-glue.c (face_to_style): Routine to convert from faces to GtkStyle * menubar-gtk.c (gtk_popup_menu): Honor popup_up_p here. (popdown_menu_cb): and here. 2000-05-29 William M. Perry <wmperry@aventail.com> * frame-gtk.c (gtk_popup_frame): Do not show the widgets if we were told to be initially unmapped. (gtk_init_frame_1): Remember whether we were told to be initially unmapped. Balloon help looks a little better now. 2000-05-28 William M. Perry <wmperry@aventail.com> * redisplay-gtk.c (gtk_output_string): Fixed multi-dimensional text run drawing. gdk_draw_text does too much, by dividing the length by 2. So we fake them out my multiplying the length by the dimension of the text. This will do the right thing for single-dimension runs as well of course. 2000-05-26 William M. Perry <wmperry@aventail.com> * ui-gtk.c (get_enumeration): Utility function that does its best to import unknown enumeration types. * glyphs-gtk.c (resource_symbol_to_type): Fixed pointer instantiation. * gtk-xemacs.c (FROB_FACE): Make sure to pass the device to Fmake_image_instance or the initial background pixmaps would not show up. 2000-05-25 William M. Perry <wmperry@aventail.com> * device-gtk.c (gtk_init_device): Call gnome_init if available. * menubar-gtk.c (create_menubar_widget): Use gnome_app_set_menus instead of dealing with all the handlebox/menu crap ourselves. * frame-gtk.c (gtk_create_widgets): Use GnomeApp if it is available. Looks much sexier. :) * glyphs-gtk.c (gtk_resource_instantiate): New image instantiator gtk-resource, similar to the mswindows-resource stuff. This allows you to get to the stock cursors defined in GDK. May eventually allow you access to things like stock gnome pixmaps, not sure. * frame-gtk.c (gtk_set_frame_pointer): Actually handle setting the frame cursor. * redisplay-gtk.c (gdk_draw_bitmap): New function to output a bitmap using XCopyPlane instead of XCopyArea. (gtk_output_gdk_pixmap): Use it when PIXMAP_DEPTH == 0. This means bitmaps look correct now. 2000-05-24 William M. Perry <wmperry@aventail.com> * ui-gtk.c (flags_to_list): New function to convert a GtkFlags argument to a list of symbols. (gtk_type_to_lisp): Use it for converting from a flag. 2000-05-23 William M. Perry <wmperry@aventail.com> * frame-gtk.c (gtk_set_frame_position): Do not use gtk_window_reposition - this apparently does not exist in some versions of GTK 1.2 * gtk-glue.c (xemacs_gtklist_to_list): Don't call g_list_foreach on a NULL list, just in case. * redisplay-gtk.c (gtk_text_width_single_run): Use gdk_text_width instead of gdk_text_measure. Proportional fonts appear to work now. * objects-gtk.c (gtk_initialize_font_instance): Use X specific crap lifted from objects-x.c to figure out proportionality of a font, and better location of the default character. 2000-05-22 William M. Perry <wmperry@aventail.com> * ui-gtk.c (enum_to_symbol): Routine to convert an enum to a lisp symbol. We now return this instead of an integer for GtkFlags or GtkEnum types. (lisp_to_flag): Routine to convert from a symbol or list of symbols to a GtkEnum or GtkFlags type. (lisp_to_gtk_type): Use it exclusively. No more sending integers to functions. (import_gtk_enumeration_internal): Do not import the GTK_FOO_BAR flags/enums into the lisp namespace. We now store mappings from symbol names (both the 'real' and nickname fields) to the actual internal values. Much more lispy way of doing things. * menubar-gtk.c (__maybe_destroy): If we delete the menu item that was 'active' when the menu was cancelled, GTK gets upset because it tries to remove the focus rectangle from a (now) dead widget. This widget will eventually get killed because it will not be `precious' the next time the window is shown, because it is unselectable. * frame-gtk.c (delete_event_cb): Signal a fake event to make the event loop wake up and notice we've been destroyed. 2000-05-20 William M. Perry <wmperry@aventail.com> * ui-gtk.c (object_putprop): Allow `put'ing of arbitrary slots on gtk objects. This will be useful for tacking lisp shit onto composite widgets, etc. (object_getprop): Ditto for `get'ing. * frame-gtk.c (gtk_set_initial_frame_size): Don't delay when changing the frame size, or creating initially unmapped frames is screwed up. This showed up when trying to use the cheesy XEmacs file selector dialog implemented using frames. * ui-byhand.c: Removed a lot of functions that can now be imported directly by gtk-import-function using the Gtk(Array|List)Of(String|Object) types. * ui-gtk.c (type_to_marshaller_type): Deal with new array/list types. (Fgtk_call_function): Need to free array and list storage after calling functions. (lisp_to_gtk_type): Deal with the new list and array types when converting to GTK types. * gtk-glue.c: New file to deal with lists of strings or GtkObjects. Defines new types GtkArrayOf, GtkListOf, GtkArrayOfString, GtkListOfString, and GtkListOfObject. (xemacs_list_to_array): Convert from a lisp list to a GtkArrayOf of items. (xemacs_list_to_list): Convert from a lisp list to a GtkListOf of items. * dialog.c (Fpopup_dialog_box): Don't crap out if the car of dbox_desc is not a string... this allows us to follow ben's system/native dialog box specification. 2000-05-19 William M. Perry <wmperry@aventail.com> * ui-gtk.c (lisp_to_gtk_type): Can now convert to GDK colors. Can deal with color specifiers or instances. (lisp_to_gtk_type): Can now convert to GDK fonts. Can deal with face names, actual faces, font specifiers, or font instances. 2000-05-18 William M. Perry <wmperry@aventail.com> * gtk-xemacs.c (smash_scrollbar_specifiers): Function that attempts to set the scrollbar width/height correctly, but does not do a very good job. Commented out for now. * redisplay-gtk.c (gtk_output_vertical_divider): Got rid of lots of useless code, since we ended up just calling gtk_output_shadows anyway. (gtk_output_vertical_divider): Make sure we fill the rectangle completely, otherwise the transparent background shows thru. * menubar-gtk.c: Don't nuke menus all the time... should speed up submenu traversal a bit, and save on the GtkMenuItem creation. * device-gtk.c (Fgtk_style_info): Return a list of all the pixmaps, not just GTK_STATE_NORMAL. * menubar-gtk.c (menu_descriptor_to_widget_1): Better menu labels with keybindings. * frame-gtk.c (gtk_set_frame_size): This function actually works now. (gtk_set_initial_frame_size): Better default sizing method. * event-gtk.c (init_event_gtk_late): Push an error trap handler so that XEmacs won't abort at the drop of a hat on X errors. We could get X specific here and override the default GDK XError and XIOError handlers to be more like those in device-x.c. Not sure if that is worth it or not - you would at least get information that an error occurred. * scrollbar-gtk.c (gtk_update_scrollbar_instance_status): Don't always call gtk_widget_set_usize because that causes the widget to redraw itself, which can cause some annoying flicker. (gtk_update_scrollbar_instance_status): Don't always move it either, because that can cause the GtkFixed container to get a resize event. (update_one_widget_scrollbar_pointer): Try to set the cursor 2000-05-17 William M. Perry <wmperry@aventail.com> * device-gtk.c (Fgtk_style_info): Back to taking only 1 argument - the device. (Fgtk_style_info): Now returns ALL of the information about colors. Returns a list of colors instead of just the one associated with GTK_STATE_NORMAL. 2000-05-16 William M. Perry <wmperry@aventail.com> * gtk-xemacs.c (smash_face_fallbacks): New function to reset the gtk-specific fallbacks on various face specifiers. This means that if the user has not changed the face, when a theme or style is changed, the faces will automatically change as well. (gtk_xemacs_style_set): Call it. * toolbar-gtk.c (get_toolbar_gc): Swap the fg/bg of the toolbar face when getting the GC. It looks better this way. * gtk-xemacs.c (gtk_xemacs_style_set): Override the style-set method so that we can nuke the pixmaps again. (__nuke_background_items): Moved the voodoo out into its own function so that it can be called from both style_set and realize. * console-gtk.h (struct gtk_frame): Removed hardcoded GCs ala X. * toolbar-gtk.c (get_toolbar_gc): New function that dynamically creates a GC from the `toolbar' face. (gtk_draw_blank_toolbar_button): Use it instead of the hardcoded GC. (gtk_output_toolbar_button): Ditto. (gtk_output_toolbar): Ditto. * event-gtk.c (gtk_check_for_quit_char): Quit handling actually works now. Feh! * device-gtk.c (gtk_device_init_x_specific_cruft): New function that gets the socket we are listening to the X server on so that the SIGIO lossage works correctly for GTK devices. (gtk_init_device): Call it at device creation time. 2000-05-15 William M. Perry <wmperry@aventail.com> * ui-gtk.c (__internal_callback_marshal): We now correctly handle the 'data' argument. This is an arbitrary lisp object passed to the callback routine as its last argument. 2000-05-14 William M. Perry <wmperry@aventail.com> * event-gtk.c (gtk_event_to_emacs_event): Needed to reinstate the magic to NOT differentiate betwen ! and shift-!. *sigh* * ui-gtk.c (lisp_to_gtk_type): Allow 'nil' for string types so that we can pass NULL to gtk-frame-new. 2000-05-13 William M. Perry <wmperry@aventail.com> * gtk-xemacs.c (gtk_xemacs_size_request): Needed to override the size_request method to get frame sizing correct. Whoo hoo! (gtk_xemacs_realize): Don't set the background on the GtkXEmacs window - this reduces the flicker even more. * device-gtk.c (gtk_init_device): Don't use shared memory under FreeBSD - it is apparently flaky as hell and causes lots of themes to crash and burn quite prettily. * gtk-xemacs.c (gtk_xemacs_realize): Added new widget method that makes sure to nuke the background pixmap of the XEmacs text area and its parent (the GtkWindow it is contained in). This fixes the flashing you would see with themes/styles that define a background pixmap. 2000-05-12 William M. Perry <wmperry@aventail.com> * menubar-gtk.c (gtk_update_frame_menubar_internal): Duh, should actually pay attention to menubar_will_be_visible instead of just calling gtk_widget_show_all in both branches. :) 2000-05-11 William M. Perry <wmperry@aventail.com> * menubar-gtk.c (vars_of_menubar_gtk): New variable menubar-dockable-p that controls whether to use the GtkHandleBox or not. * select-gtk.c: Implemented all of the selection callbacks for GTK * frame-gtk.c (resize_event_cb): Force a redisplay when the frame is resized. * event-gtk.c (gtk_event_to_emacs_event): When we are doing our hackery to make sure we don't see button events inside the scrollbar, make sure we correctly forward events that did not happen over ANY widget, otherwise selection gets royally screwed and thinks you still have the mouse button down. * redisplay-gtk.c (gtk_output_string): Don't bother calling gdk_draw_text_wc - I misunderstood what XDrawString16 did - gdk_draw_text encapsulates this nicely for us. 2000-05-10 William M. Perry <wmperry@aventail.com> * menubar-gtk.c: Changed how the menubar is created and managed. We no longer create and destroy it at will. Only one GtkMenuBar is ever created, and the children are just added/removed from it. Much less flickering when switching buffers - cleaner in general. (create_menubar_widget): Wheee - menubars are now detachable. * ui-gtk.c (Fgtk_import_function_internal): Don't drop everything down to its fundamental type just yet - we need to know specifics about boxed types. (object_putprop): Duh, actually SET the property. (Fgtk_fundamental_type): New function that does the obvious. (Fgtk_object_type): New function that does the obvious. (lisp_to_gtk_type): Implement glyph handling! GtkPixmap works! * ui-byhand.c (Fgtk_pixmap_get): Implemented by hand. *sigh* * dialog-gtk.c: Call into lisp to do dialog boxes. 2000-05-08 William M. Perry <wmperry@aventail.com> * make-src-depend (PrintPatternDeps): Make sure we generate the xx-gtk.* dependencies. * depend: Regenerated * make-src-depend: Handle inclusion of .c files. * Makefile.in.in (extra_doc_files): Need to include ui-byhand.c here as well. * ui-gtk.c (type_to_marshaller_type): Don't abort if we don't know about the marshaller. Just return Qnil and let the caller figure it out. (Fgtk_import_function_internal): Deal with new return value of Qnil for error signalling from type_to_marshaller_type(). (lisp_to_gtk_type): Half-assed implementation of dealing with callbacks. (gtk_type_to_lisp): Ditto. * ui-byhand.c: New file that hand-codes some annoying functions in Gtk. *sigh* * emacs-widget-accessors.c (Fgtk_adjustment_lower): New auto-generated file that spits out all the widget slot accessor functions. * ui-gtk.c (Fgtk_signal_connect_internal): Holy shit, signals work! (lisp_to_gtk_type): Allow ints to be passed in as arguments to functions that expect floats, otherwise you have to write stupid code like 9.0 instead of just passing in '9'. 2000-05-07 William M. Perry <wmperry@aventail.com> * ui-gtk.c (gtk_type_to_lisp): Return Qnil if we get a NULL pointer/object/boxed. (lisp_to_gtk_type): Allow Qnil to be passed in to mean a NULL pointer/object/boxed. * event-gtk.c (gtk_event_to_emacs_event): Make sure a button press focuses on the text widget, otherwise if someone packs a widget into the container from lisp, we end up not getting keyboard focus quite a bit. Eek. * frame-gtk.c (gtk_create_widgets): Set the name of the xemacs shell so we can control sylte issues from .gtkrc (gtk_set_initial_frame_size): Set the default size of the frame correctly. (gtk_create_widgets): Expose the frame shell and container widgets to lisp. * emacs-marshals.c: Added a whole shitload of marshallers - don't edit this file by hand if you need to add one. Check out ../lisp/ui/gtk-marshal.el for a script that auto-generates them for you. 2000-05-05 William M. Perry <wmperry@aventail.com> * ui-gtk.c (describe_gtk_arg): Debugging function to spit out human-readable version a GtkArg. (lisp_to_gtk_type): Made this function much more sane. (Fgtk_call_function): New function to allow calling generic functions from DLLs in lisp. (CONVERT_RETVAL): New macro (ugh) to convert a pointer that the return value filled in back into a valid GtkArg. (Fgtk_call_function): This actually works now! Scary stuff. 2000-05-04 William M. Perry <wmperry@aventail.com> * ui-gtk.c (Fgtk_import_type): New function to import a GtkType into the XEmacs namespace. For starters this only fully imports enumerations and flags. Importing a GtkObject or subclass results in getting a list back of possible signals and argument lists. * scrollbar-gtk.c (gtk_create_scrollbar_instance): Fixed some compiler warnings. (scrollbar_cb): Aha! Thanks to the magic of reading the gtkrange.c source code, we now have better behaving scrollbars. Clicking in the trough/end arrows now works correctly instead of the semi-hackish way it did earlier today. * ui-gtk.c (Fgtk_define_enumeration): New function that can * scrollbar-gtk.c (scrollbar_cb): Combined the horizontal and vertical scrolling callbacks. Differentiate between them by user_data. (scrollbar_cb): Don't bother trying to figure out whether this is a line-by-line or page-wide scrolling operation. Just go by what is in the value of the adjustment and issue a scrollbar_*_drag event. This is not 100% correct, but it at least gets it working. Doing it 'right' might not be possible with standard Gtk scrollbars. (scrollbar_drag_hack_cb): New function to set vertical_drag_in_progress and inhibit_slider_change when we might be starting a drag in the scrollbar. (gtk_update_scrollbar_instance_status): Better setting of the adjustment variables. Scrolling by pages (clicking in the trough) works now. Line-by-line seems to work too. 2000-05-03 William M. Perry <wmperry@aventail.com> * scrollbar-gtk.c (gtk_update_scrollbar_instance_status): Got the thumb sizing correct finally. Thanks to scrollbar-msw.c for showing how to do it sanely - scrollbar-x.c is a mess. * window.c (specifier_vars_of_window): Added GTK specific fallback for Vvertical_divider_line_width. * toolbar.c (specifier_vars_of_toolbar): Handle specifier fallbacks for GTK toolbars. * gui-gtk.c (button_item_to_widget): Compiles under Gtk at least - more than likely completely broken. * glyphs-gtk.c (write_lisp_string_to_temp_file): Resurrected this blasphemous function to deal with XPMs. (gtk_xpm_instantiate): Now writes the XPM to a temp file, since GTK cannot deal with XPMs from memory. Lame, but not fixable for gtk 1.2 - maybe for 1.4. (gtk_xpm_instantiate): Transparency works now. * gccache-gtk.c (gc_cache_lookup): Made non-hash based code compile. It is not used, but what the hell. * faces.c (complex_vars_of_faces): Do GTK specific mucking about with face property fallbacks (fonts and colors) * events.c (event_equal): Added special case for GTK. (event_hash): Added special case for GTK. (format_event_object): Added special case for GTK events. * event-gtk.c (event_name): Use gtk_event_name helper function here. (handle_focus_event_1): Got rid of gtk-redisplay-hack variable and usage. * device.c (delete_device_internal): Delete 'popup' frames under Gtk as well as X. Should this happen for HAVE_MSWINDOWS as well? * console.c (select_console_1): Make sure we set Vwindow_system like all the others. * frame-gtk.c (gtk_update_frame_external_traits): Added comments about why I didn't implement some portions of this function. * redisplay-gtk.c (gtk_output_string): Fixed the bizarre redisplay problem where all the default text would be drawn twice - once with the normal background colour, and once with the text-cursor background. This was caused by a stupid typo and using the wrong GdkGC in the second call to gdk_draw_text_image... basically no clipping was being done on the image, so the whole thing was redrawn. (gtk_output_string): Call gdk_draw_text if we have a background pixmap, otherwise things look REALLY weird. (gtk_clear_region): Had a misplaced brace that caused the non-text area of a frame not to get the correct background. 2000-05-02 William M. Perry <wmperry@aventail.com> * glyphs-gtk.c (gtk_xpm_instantiate): Need to write pixmaps to a temp file and then read with gdk_pixmap_create_from_xpm () - there is no way to read from a memory buffer (yet - I might write one for Gtk 1.4) * glyphs.c: Don't include xpm.h when building with Gtk. (pixmap_to_lisp_data): Alternate implementation for Gtk. * device-gtk.c (gtk_init_device): Call gdk_imlib_init if available, otherwise the 'Pixmap' theme engine doesn't work. Losers. * glyphs-gtk.c (gtk_xpm_instantiate): Now at least tries to instantiate XPM images. (init_image_instance_from_gdk_pixmap): Utility function to create a glyph from an existing GdkPixmap. * device-gtk.c (Fgtk_style_info): Attempt to expose the background pixmaps from a Gtk style.