Mercurial > hg > xemacs-beta
view lisp/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
2000-09-12 William M. Perry <wmperry@aventail.com> * dialog-gtk.el (popup-builtin-open-dialog): Went back to using our lisp implementation of the file dialog. Much more featureful. * ui/gtk-file-dialog.el: Reworked to use CList instead of Tree elements (more like the `real' GTK file selector. 2000-09-10 William M. Perry <wmperry@aventail.com> * gtk-iso8859-1.el (gtk-iso8859-1): Need to actually provide the feature 2000-09-09 William M. Perry <wmperry@aventail.com> * dialog-gtk.el (popup-builtin-open-dialog): Guard against calling gtk-main-quit too many times when destroying the file-selection dialog. 2000-09-08 William M. Perry <wmperry@aventail.com> * gtk-init.el (gtk-initialize-compose): Initialize the compose map like X does. 2000-09-03 William M. Perry <wmperry@aventail.com> * menubar-items.el (default-menubar): Include the font & size menus when running under GTK. * gtk-faces.el (x-font-regexp-*): Added variable aliases for the x-font-regexp-* variables. x-font-menu works now. * x-font-menu.el (font-menu-set-font): When setting the font, make sure we don't set the type to 'x' blithely. This code is shared with GTK now. 2000-08-30 William M. Perry <wmperry@aventail.com> * dialog-gtk.el (popup-builtin-open-dialog): Signal 'quit' if the user hits the cancel button. This gets rid of the 'wrong type argument: stringp, nil' error. 2000-08-28 William M. Perry <wmperry@aventail.com> * menubar-items.el (default-menubar): Disable the GTK font menu item. * dialog-gtk.el (popup-builtin-open-dialog): Reimplemented the file-open dialog to use the normal GTK selector. 2000-07-26 William M. Perry <wmperry@aventail.com> * gtk-init.el (init-post-gtk-win): Define the mule-fonts specifier tag and default fonts for it when mule is provided. This will make x-symbol.el work. 2000-07-24 William M. Perry <wmperry@aventail.com> * ui/gdk.el: Ditto. * ui/gnome-widgets.el: Ditto. * ui/gtk-widgets.el: Updated all gtk-import-function calls to pass a symbol instead of a quoted string. * ui/gtk-ffi.el (gtk-import-variable): Make gtk-import-variable able to take symbols instead of just strings. More consistent with gtk-import-function this way. 2000-07-22 William M. Perry <wmperry@aventail.com> * gtk-init.el (init-pre-gtk-win): Did not realize I had to do lisp hackery to get '-unmapped' to work. 2000-07-12 William M. Perry <wmperry@aventail.com> * ui/gtk-widgets.el (gtk-ctree-post-recursive): (gtk-ctree-post-recursive-to-depth): (gtk-ctree-pre-recursive): (gtk-ctree-pre-recursive-to-depth): Added wrappers around the combined gtk-ctree-recurse to make things easier on GTK authors porting other code. 2000-07-11 William M. Perry <wmperry@aventail.com> * gtk-init.el (init-gtk-win): Set gtk-initial-geometry out of command-line-args-left. (gtk-filter-arguments): New function to filter out only GTK/GNOME approved command line arguments. (init-gtk-win): Set gtk-initial-argv-list by filtering it. This way we get session management/etc from GNOME. 2000-07-07 William M. Perry <wmperry@aventail.com> * gtk-faces.el (gtk-init-global-faces): Make sure to pass in a GTK device (any GTK device) try-font-name when initializing the global faces or it gets confused and cannot find a font, so the code in faces.el ends up setting device-specific faces, which are hard (or at least non-obvious) for users to work around. 2000-07-01 William M. Perry <wmperry@aventail.com> * ui/gtk-widgets.el (GtkType): Added GtkCTree finally. * ui/gtk-ffi.el (gtk-ffi-check-function): New function that will try to call a newly defined FFI function and report on whether we need to define any new marshallers for it. (gtk-ffi-debug): If non-nil, then we will check ALL functions that come through gtk-import-function. All existing imported functions have been checked, and a few missings ones were added. * ui/glade.el: New file to import libglade functions. 2000-06-30 William M. Perry <wmperry@aventail.com> * ui/gdk.el: Added most of the GDK drawing primitives. 2000-06-27 William M. Perry <wmperry@aventail.com> * ui/gtk-widgets.el: Import gtk-major-version, gtk-minor-version, gtk-micro-version, gtk-interface-age, and gtk-binary-age. * ui/gtk-ffi.el (gtk-import-variable): New macro to import a variable. Needed to do it as a function, otherwise you could not byte compile / dump the file in a non-windowed XEmacs. 2000-06-23 William M. Perry <wmperry@aventail.com> * menubar-items.el (default-menubar): Disable make-frame-on-display if the function is not available. 2000-06-02 William M. Perry <wmperry@aventail.com> * gtk-mouse.el (gtk-init-pointers): Make sure we set a toolbar pointer. Looked kind of silly to have the 'xterm' cursor in there by default. 2000-06-01 William M. Perry <wmperry@aventail.com> * ui/gtk-widgets.el: Imported GtkStatusbar * ui/gtk-marshal.el (define-marshaller): All marshalling functions are now static. * ui/gtk-marshal.el: Now outputs a function to populate a hashtable with mappings from function name -> function pointer. Also emits the find_marshaller () function that looks at this hashtable. 2000-05-29 William M. Perry <wmperry@aventail.com> * ui/gtk-ffi.el (gtk-import-function): Allow passing in of a symbol for the function name, as well as a string. * ui/gtk-widgets.el: Import the GtkSpinButton widget. 2000-05-26 William M. Perry <wmperry@aventail.com> * ui/gtk-ffi.el (gtk-import-function): Rewrote as a macro so that teh file can be safely byte-compiled. Argument values no longer need to be quoted, and the function is not actually imported until the function is called. Should save even more on loadup time. (gtk-import-function): Do not defvar `lisp-name' - put the FFI object on the symbols plist. Makes the lisp variable namespace that much cleaner. * ui/gtk-widgets.el: New uber-file containing all the GTK imported functions. Load time is significantly faster than requiring ~90 different (usually 5 line) .el files, and polluting the 'features' variable. 2000-05-23 William M. Perry <wmperry@aventail.com> * ui/gtk-widget-accessors.el (define-widget-accessors): Fixed bogus handling of GTkListOfString and GtkListOfObject slots. These need to have the FULL type, not the fundamental type. And _POINTER was just plain wrong for them - cause beautiful crashes on linux. :) * gtk-faces.el (gtk-choose-font): New function to change fonts based on a GTK font selection dialog. * menubar-items.el (default-menubar): Only show the font/size submenus when we are in an X frame. (default-menubar): Show a gtk-specific item when on GTK frames. * ui/gtk-marshal.el (define-marshaller): We need to special case anything with FLOAT in the argument list or the parameters get screwed up royally. 2000-05-21 William M. Perry <wmperry@aventail.com> * ui/gtk-file-dialog.el (gtk-file-dialog-fill-file-list): New file dialog that is actually useful and much prettier than the default GTK one. 2000-05-20 William M. Perry <wmperry@aventail.com> * ui/gtk-ffi.el (gtk-import-function): Auto-run gnome-*-get-type routines as well as the gtk-*-get-type ones. * minibuf.el (mouse-read-file-name-1): Now tries to use the new builtin dialog spec ben wrote about. Uses the GTK file selection dialog. We should be able to come up with something much sexier though - the default dialog box for GTK sucks hard. 2000-05-17 William M. Perry <wmperry@aventail.com> * gtk-faces.el (gtk-init-device-faces): Make use of the extended return values of gtk-style-info. Set the 'highlight' face to look like GTK_STATE_PRELIGHT and 'zmacs-region' to be GTK_STATE_SELECTED. Unfortunately these two faces will not automatically be updated because they are not exposed to lisp like Vdefault_face and friends. 2000-05-16 William M. Perry <wmperry@aventail.com> * gtk-faces.el (gtk-init-device-faces): Removed a whole bunch of face munging that is now done down in the guts of the GtkXEmacs widget. * gnuserv.el (gnuserv-edit-files): Handle GTK devices. * ui/gtk-ffi.el (gtk-import-function): Make this a noop if noninteractive. This allows us to compile the files during the make process. 2000-05-10 William M. Perry <wmperry@aventail.com> * dialog-gtk.el: New file implementing popup dialogs in Lisp using GTK primitives. Called from dialog-gtk.c * dumped-lisp.el (preloaded-file-list): Make sure we load up menubar-items under GTK. (preloaded-file-list): Load up dialog-gtk when using GTK. 2000-05-08 William M. Perry <wmperry@aventail.com> * ui/gtk-widget-accessors.el (define-widget-accessors): New file to define C functions that go into ../../src/emacs-widget-accessors.c. This is a hack to get around the lack of accessor/settor functions in GTK for a LOT of things that are required for full functionality (like dialogs) 2000-05-07 William M. Perry <wmperry@aventail.com> * gtk-faces.el (gtk-init-face-from-resources): Set the highlight face as well.