Mercurial > hg > xemacs-beta
view etc/ONEWS @ 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 | 5aa1854ad537 |
| children | c84a9844ed2c |
line wrap: on
line source
-*- mode:outline -*- * Changes in XEmacs 20.4 ======================== ** XEmacs 20.4 is a bugfix release with no user-visible changes. * Changes in XEmacs 20.3 ======================== ** Quail input method is now available. Quail is a simple key-translation system that allows users to input any multilingual text from normal ASCII keyboard. This means that XEmacs with Mule now supports a number of European languages. ** More Windows NT support. Thanks to efforts of many people, coordinated by David Hobley <davidh@wr.com.au> and Marc Paquette <marcpa@cam.org>, beta versions of XEmacs now run on 32-bit Windows platforms (Windows NT and Windows 95). The current betas require having an X server to run XEmacs; however, a native NT/95 port is in alpha, thanks to Jonathan Harris <jhar@tardis.ed.ac.uk>. The NT development is now coordinated by a mailing list at <xemacs-nt@xemacs.org>. Mail to <xemacs-nt-request@xemacs.org> to subscribe. ** Multiple TTY frames are now available. On consoles that display only one frame at a time (e.g. TTY consoles), creating a new frame with `C-x 5 2' also raises and selects that frame. The behavior of window system frames is unchanged. ** Package starting changes. State of Emacs should never be changed with loading a package. The following XEmacs packages that used to break this have been changed. *** Loading `paren' no longer enables paren-blinking. Use `paren-set-mode' explicitly, or customize `paren-mode'. *** Loading `uniquify' no longer enables uniquify. Set `uniquify-buffer-name-style' to a legal value. *** Loading `time' no longer enables display time. Invoke `display-time' explicitly. *** Loading `jka-compr' no longer enables on-the-fly compression. Use `toggle-auto-compression' instead. *** Loading `id-select' no longer enables its behaviour. Use `id-select-install' instead. ** Zmacs region is not deactivated when an error is signaled. The behavior of the zmacs region can now be controlled in the event of a signaled error. The new variable `errors-deactivate-region' may be set to nil to revert to the old behaviour. As before, typing C-g deactivates the region. ** Multiple Info `dir' functionality has been merged with GNU Emacs 19.34. XEmacs will now correctly merge all the `dir' files in `Info-directory-list' (initialized from either the `INFOPATH' env. variable or `Info-default-directory-list'.) These files may be full-fledged info files containing subnodes or menus. Previously supported `localdir' files are looked for also, secondary to `dir's. See the manual for details. ** Abbreviations can now contain non-word characters. This means that it is finally possible to do such simple things as define `#in' to expand to `#include' in C mode, `s-c-b' to `save-current-buffer' in Lisp mode, `call/cc' to `call-with-current-continuation' in Scheme mode, etc. ** `C-x n d' now runs the new command `narrow-to-defun', which narrows the accessible parts of the buffer to just the current defun. ** The new command `C-x 4 0' (kill-buffer-and-window) kills the current buffer and deletes the selected window. It asks for confirmation first. ** `ESC ESC ESC' (keyboard-escape-quit) will now correctly abort recursive edits (as documented.) ** arc-mode has a new function called `archive-quit' bound to q, which quits archive mode in the same fashion dired-quit works. ** A `tetris' clone is now available within XEmacs, written by Glynn Clements. Try it out with `M-x tetris'. ** The feature to teach the key bindings of extended commands now prints the message after the command finishes. After some time, the previous echo area contents are restored (in case the command prints something useful). ** If you set scroll-conservatively to a small number, then when you move point a short distance off the screen, XEmacs will scroll the screen just far enough to bring point back on screen, provided that does not exceed `scroll-conservatively' lines. ** Face background colors now take precedence over the default face background pixmap, which means that background pixmaps no longer clash with zmacs-regions, or clickable buttons. ** Regexps can now contain additional Perl-like constructs. ** Modifiers can be added to a keystroke by preceding it with a `C-x @ <x>' sequence where <x> is one of letters `S', `c', `m', `a', `h', `s' corresponding to shift, control, meta, alt, hyper, and super modifiers, respectively. It is possible to add several modifiers by repeating this sequence. This feature is especially useful on text terminals where it allows one to enter keystrokes like, e.g., `M-home'. ** An arbitrary keystroke can be generated by entering `C-x @ k <keysym-name> RET'. For example a sequence: C-x @ c C-x @ k b a c k s p a c e RET will result in a `C-backspace' keystroke even on text terminals. ** Customize changes. *** Customize has undergone a massive speedup, and should now operate acceptably fast. Slowness of the interface used to be the biggest gripe. *** Many more packages have been modified to use the facility, so almost all of XEmacs options can now be examined through the Customize groups. *** There is a new `browser' mode of traversing customizations, in many ways easier to follow than the standard one. Try it out with `M-x customize-browse'. ** Pending-delete changes. *** Pending-delete is now a minor mode, with the normal minor-mode semantics and toggle functions. Old functions are left for compatibility. *** Loading pending-del no longer turns on pending-delete mode. In fact, it is no longer necessary to explicitly load pending-del. All you need to do to turn on pending-delete is run the pending-delete function: Within XEmacs: Type M-x pending-delete <ret> not M-x load-library <ret> pending-delete <ret> In .emacs: Use (turn-on-pending-delete) not (load "pending-del") ** XEmacs can now save the minibuffer histories from various minibuffers. To use this feature, add the line: (savehist-load) to your .emacs. This will load the minibuffer histories (if any) at startup, as well as instruct XEmacs to save them before exiting. You can use Customize to add or remove the histories being saved. ** The default format for ChangeLog entries (as created by `C-x 4 a') is now the international ISO 8601 format. To revert to the old behaviour, use: (setq add-log-time-format 'current-time-string) Or `M-x customize RET add-log RET'. ** In ChangeLog mode, you can now press `C-c C-c' to save the file and restore old window configuration, or `C-c C-k' to abandon the changes. ** The key `C-x m' no longer runs the `mail' command directly. Instead, it runs the command `compose-mail', which invokes the mail composition mechanism you have selected with the variable `mail-user-agent'. The default choice of user agent is `sendmail-user-agent', which gives behavior compatible with the old behavior. C-x 4 m now runs compose-mail-other-window, and C-x 5 m runs compose-mail-other-frame. ** When you kill a buffer that visits a file, if there are any registers that save positions in the file, these register values no longer become completely useless. If you try to go to such a register with `C-x j', then you are asked whether to visit the file again. If you say yes, it visits the file and then goes to the same position. ** When you visit a file that changes frequently outside Emacs--for example, a log of output from a process that continues to run--it may be useful for Emacs to revert the file without querying you whenever you visit the file afresh with `C-x C-f'. You can request this behavior for certain files by setting the variable revert-without-query to a list of regular expressions. If a file's name matches any of these regular expressions, find-file and revert-buffer revert the buffer without asking for permission--but only if you have not edited the buffer text yourself. ** Gnuserv changes *** The Lisp part of gnuserv has been rewritten to allow for more flexibility and features. *** Many new options and variables are now customizable. Try `M-x customize RET gnuserv RET'. *** The functionality of `gnuattach' and `gnudoit' programs is provided by `gnuclient', which now accepts the standard `-nw', `-display', `-eval' and `-f' options. ** Etags changes. *** In C, C++, Objective C and Java, Etags tags global variables by default. The resulting tags files are inflated by 30% on average. Use --no-globals to turn this feature off. Etags can also tag variables that are members of structure-like constructs, but it does not by default. Use --members to turn this feature on. *** C++ member functions are now recognized as tags. *** Java is tagged like C++. In addition, "extends" and "implements" constructs are tagged. Files are recognised by the extension .java. *** Etags can now handle programs written in Postscript. Files are recognised by the extensions .ps and .pdb (Postscript with C syntax). In Postscript, tags are lines that start with a slash. *** Etags now handles Objective C and Objective C++ code. The usual C and C++ tags are recognized in these languages; in addition, etags recognizes special Objective C syntax for classes, class categories, methods and protocols. *** Etags also handles Cobol. Files are recognised by the extension .cobol. The tagged lines are those containing a word that begins in column 8 and ends in a full stop, i.e. anything that could be a paragraph name. *** Regexps in Etags now support intervals, as in ed or grep. The syntax of an interval is \{M,N\}, and it means to match the preceding expression at least M times and as many as N times. ** Ada mode changes. *** There is now better support for using find-file.el with Ada mode. If you switch between spec and body, the cursor stays in the same procedure (modulo overloading). If a spec has no body file yet, but you try to switch to its body file, Ada mode now generates procedure stubs. *** There are two new commands: - `ada-make-local' : invokes gnatmake on the current buffer - `ada-check-syntax' : check syntax of current buffer. The user options `ada-compiler-make', `ada-make-options', `ada-language-version', `ada-compiler-syntax-check', and `ada-compile-options' are used within these commands. *** Ada mode can now work with Outline minor mode. The outline level is calculated from the indenting, not from syntactic constructs. Outlining does not work if your code is not correctly indented. *** The new function `ada-gnat-style' converts the buffer to the style of formatting used in GNAT. It places two blanks after a comment start, places one blank between a word end and an opening '(', and puts one space between a comma and the beginning of a word. ** New demand based locking implementation A faster, but experimental replacement for lazy-lock (called lazy-shot) is provided. Like lazy-lock it provides demand based and idle time font-lock-ing. However the lazy-lock versions that came with previous versions slowed down XEmacs (possibly quite a lot). Lazy-shot solves this problem by relying on new support from the C code part of XEmacs. The support however is experimental and will cause some flashing as parts of the buffer are colored. This likely to change in the future as the C support is completed. The current lazy-shot implementation is mostly interface compatible with lazy-lock v2.06 (the version shipped with XEmacs is v1.x). *** To enable: 1. Despite the flashing, lazy-shot was deemed such an improvement by the majority of beta testers that it is now the standard method provided by the options menu. Alternatively add (add-hook 'font-lock-mode-hook 'turn-on-lazy-shot) to '.emacs'. 2. If you were using lazy-lock before, just replace all occurrences of "lazy-lock" by "lazy-shot" in your '.emacs' file. *** To disable: If prefer to use lazy-lock in stead of lazy-shot, put (remove-hook 'font-lock-mode-hook 'turn-on-lazy-shot) (add-hook 'font-lock-mode-hook 'turn-on-lazy-lock) at the END of `.emacs'. ** RefTeX mode RefTeX mode is a new minor mode with special support for \label{}, \ref{} and \cite{} macros in LaTeX documents. RefTeX distinguishes labels of different environments (equation, figure, ...) and has full support for multifile documents. To use it, select a buffer with a LaTeX document and turn the mode on with M-x reftex-mode. Here are the main user commands: C-c ( reftex-label Creates a label semi-automatically. RefTeX is context sensitive and knows which kind of label is needed. C-c ) reftex-reference Offers in a menu all labels in the document, along with context of the label definition. The selected label is referenced as \ref{LABEL}. C-c [ reftex-citation Prompts for a regular expression and displays a list of matching BibTeX database entries. The selected entry is cited with a \cite{KEY} macro. C-c & reftex-view-crossref Views the cross reference of a \ref{} or \cite{} command near point. C-c = reftex-toc Shows a table of contents of the (multifile) document. From there you can quickly jump to every section. Under X, RefTeX installs a "Ref" menu in the menu bar, with additional commands. Full documentation and customization examples are in the file reftex.el. You can use the finder to view this information: C-h p --> tex --> reftex.el * Lisp and internal changes in XEmacs 20.3 ========================================== ** Autoconf 2 is supported, making XEmacs more conforming to conventions used by other free software. ** `tty-erase-char' is a new variable that reports which character was set up as the terminal's erase character at the time Emacs was started. ** It is now possible to attach the menubar accelerator keys to menu entries. Look at the Lispref under Menus->Menu Accelerators for details. ** `insert-file-contents' can now read from a special file, as long as the arguments VISIT and REPLACE are nil. ** `string-to-number' now accepts an optional BASE argument that specifies which base to use. The default base is 10. ** The TIME argument to `format-time-string' is now optional and defaults to the current time. ** The PATTERN argument to `split-string' is now optional and defaults to whitespace ("[ \f\t\n\r\v]+"). ** `set-extent-properties' is a new function that can be used to change properties of an extent at once, and is analogous to `set-frame-properties'. ** If a format field width is specified as `*', the field width is now assumed to have been specified as an argument (as in C.) (format "%*s" 10 "abc") => " abc" ** The new macro `with-current-buffer' lets you evaluate an expression conveniently with a different current buffer. It looks like this: (with-current-buffer BUFFER BODY-FORMS...) BUFFER is the expression that says which buffer to use. BODY-FORMS say what to do in that buffer. The old `eval-in-buffer' macro is obsoleted by `with-current-buffer'. ** The new primitive `save-current-buffer' saves and restores the choice of current buffer, like `save-excursion', but without saving or restoring the value of point or the mark. `with-current-buffer' works using `save-current-buffer'. ** The new macro `with-temp-file' lets you do some work in a new buffer and write the output to a specified file. Like `progn', it returns the value of the last form. ** The variable `debug-ignored-errors' now works in XEmacs. It allows one to ignore the debugger for some common errors, even when `debug-on-error' is t. It has no effect when `debug-on-signal' is non-nil. ** The new function `current-message' returns the message currently displayed in the echo area, or nil if there is none. ** File-access primitive functions no longer discard an extra redundant directory name from the beginning of the file name. In other words, they no longer do anything special with // or /~. The same goes for `expand-file-name'. That conversion is now done only in `substitute-in-file-name'. This makes it possible for a Lisp program to open a file whose name begins with ~. ** The regexp matcher has been extended to recognize the following constructs, borrowed from Perl: *** Additional quantifiers. In addition to `*', `+' and `?', XEmacs now recognizes the following quantifiers: \{n\} Match exactly n times \{n,\} Match at least n times \{n,m\} Match at least n but not more than m times *** Non-greedy quantifiers. Any of the standard quantifiers (`*', `+' and others) can now be followed by an optional `?', which will make them become "non-greedy", i.e. they will match as little text as possible. Note that the meanings don't change, just the "gravity." *** Shy groups. The \(?: ... \) groups things like \( ... \), but doesn't record the context for backreferences or future use. This is useful when you need a lot of groups for the sake of priorities, but actually want to record only one or two. ** The new function `regexp-opt' returns an efficient regexp to match a string. The arguments are STRINGS and (optionally) PAREN. This function can be used where regexp matching or searching is intensively used and speed is important, e.g., in Font Lock mode. ** The featurep syntax has been extended to resemble the Common Lisp one, as suggested by Erik Naggum. *** The `xemacs' feature is defined in XEmacs by default. *** The expression `#+fexp form' is equivalent to (when (featurep fexp) form), only it is evaluated at read-time. Also, `#-fexp form' is equivalent to (unless (featurep fexp) form). *** In addition to symbols, a FEXP can also be a number, or a logical operator. Here are some examples: ;; evaluates to non-nil on XEmacs: (featurep 'xemacs) ;; evaluates to non-nil on XEmacs 20.3 or later: (featurep '(and xemacs 20.03)) ;; evaluates to non-nil either on Emacs, or on XEmacs built without ;; X support: (featurep '(or emacs (and xemacs (not x)))) * Changes in XEmacs 20.2 ======================== ** Why XEmacs 20.1 is called 20.2 Testing of XEmacs 20.1 revealed a number of showstopping bugs at the very final moment. Instead of confusing the version numbers further, the `20.1' designation was abandoned, and the release was renamed to `20.2'. ** Delete/backspace keysyms have been separated The Delete and Backspace keysyms are now no longer identical. A better version of delbackspace.el has been added called delbs.el. ** XEmacs 20.0 MULE API supported for backwards compatibility XEmacs 20.2 primarily supports the MULE 3 API. It now also supports the XEmacs 20.0 MULE API. ** The logo has been changed, and the default background color is now a shade of gray instead of the eye-burning white. The sample .Xdefaults and .emacs files contain examples of how to revert to the old background color. ** Default modeline colors are now less of a color-salad. ** The `C-z' key now iconifies only the current X frame. You can use `C-x C-z' to get the old behavior. On the tty frames `C-z' behaves as before. ** The command `display-time' now draws a pretty image in the modeline when new mail arrives. It also supports balloon-help messages. ** Various commands that were previously disabled are now enabled, like eval-expression (`M-:') and upcase-region (`C-x C-u')/downcase-region (`C-x C-l'). ** It is now possible to customize the functions called by XEmacs toolbar. Type `M-x customize RET toolbar RET' to customize it. Customizations include the choice of functions for the buttons to invoke, as well as a wide choice of mailers and newsreaders to invoked by the respective functions. ** `temp-buffer-shrink-to-fit' now defaults to nil. There are unresolved issues regarding this feature, which is why the XEmacs developers decided to disable it by default. ** `ps-print-color-p' now defaults to nil. This is because the new default background color is non-white. The `Printing Options' in the `Options' menu now include an item that enables color printing, and sets the white background. ** `line-number-mode' should be used to get line numbers in the modeline, and `column-number-mode' to get column numbers. Line numbers now number from 1 by default. ** font-lock-mode will now correctly fontify `int a, b, c;' expressions in C mode. ** The blinking cursor is always "on" during movement. ** The XEmacs build process has been changed to make site administration easier. See lisp/site-load.el for details. ** Numerous causes of crashes have been fixed. XEmacs should now be even more stable than before. ** configure no longer defaults to using --with-xim=motif if Motif libraries are linked. There are many bugs in the Xlib XIM support in X11R6.3. ** A number of new packages are added, and many packages were updated. ** Gnus-5.4.52, courtesy of Lars Magne Ingebrigtsen *** nntp.el has been totally rewritten in an asynchronous fashion. *** Article prefetching functionality has been moved up into Gnus. *** Scoring can now be performed with logical operators like `and', `or', `not', and parent redirection. *** Article washing status can be displayed in the article mode line. *** gnus.el has been split into many smaller files. *** Suppression of duplicate articles based on Message-ID. (setq gnus-suppress-duplicates t) *** New variables for specifying what score and adapt files are to be considered home score and adapt files. See `gnus-home-score-file' and `gnus-home-adapt-files'. *** Groups can inherit group parameters from parent topics. *** Article editing has been revamped and is now usable. *** Signatures can be recognized in more intelligent fashions. See `gnus-signature-separator' and `gnus-signature-limit'. *** Summary pick mode has been made to look more nn-like. Line numbers are displayed and the `.' command can be used to pick articles. *** Commands for moving the .newsrc.eld from one server to another have been added. `M-x gnus-change-server' *** A way to specify that "uninteresting" fields be suppressed when generating lines in buffers. *** Several commands in the group buffer can be undone with `M-C-_'. *** Scoring can be done on words using the new score type `w'. *** Adaptive scoring can be done on a Subject word-by-word basis: (setq gnus-use-adaptive-scoring '(word)) *** Scores can be decayed. (setq gnus-decay-scores t) *** Scoring can be performed using a regexp on the Date header. The Date is normalized to compact ISO 8601 format first. *** A new command has been added to remove all data on articles from the native server. `M-x gnus-group-clear-data-on-native-groups' *** A new command for reading collections of documents (nndoc with nnvirtual on top) has been added -- `M-C-d'. *** Process mark sets can be pushed and popped. *** A new mail-to-news backend makes it possible to post even when the NNTP server doesn't allow posting. *** A new backend for reading searches from Web search engines (DejaNews, Alta Vista, InReference) has been added. Use the `G w' command in the group buffer to create such a group. *** Groups inside topics can now be sorted using the standard sorting functions, and each topic can be sorted independently. See the commands under the `T S' submap. *** Subsets of the groups can be sorted independently. See the commands under the `G P' submap. *** Cached articles can be pulled into the groups. Use the `Y c' command. *** Score files are now applied in a more reliable order. *** Reports on where mail messages end up can be generated. `M-x nnmail-split-history' *** More hooks and functions have been added to remove junk from incoming mail before saving the mail. See `nnmail-prepare-incoming-header-hook'. *** The nnml mail backend now understands compressed article files. ** Custom 1.86, courtesy of Per Abrahamsen The Customize library enables Emacs Lisp programmers to specify types of their variables, so that the users can customize them. Invoke the customizations buffer using the menus (Customize is at the top of the Options menu), or using commands `M-x customize', `M-x customize-variable' and `M-x customize-face'. Customize can save the changed settings to your `.emacs' file. Customize is now the preferred way to change XEmacs settings. Tens of packages have been converted to take advantage of the Customize features, including Gnus, Message, Supercite, Psgml, Comint, W3, cc-mode (and many other programming language modes), ispell.el, ps-print.el, id-select.el, most of the programming language modes, and many many more. See the "Lisp Changes" section later for a short description of why and how to add custom support to your Lisp packages. Custom is also documented in the XEmacs info manuals. ** W3-3.0.86, courtesy of William Perry Version 3 of Emacs/W3, the Emacs World Wide Web browser, has been included. It is significantly faster than any of the previous versions, and contains numerous new features. ** AUCTeX-9.7k, courtesy of Per Abrahamsen AUC TeX is a comprehensive customizable integrated environment for writing input files for LaTeX using Emacs. AUC TeX lets you run TeX/LaTeX and other LaTeX-related tools, such as a output filters or post processor from inside Emacs. Especially `running LaTeX' is interesting, as AUC TeX lets you browse through the errors TeX reported, while it moves the cursor directly to the reported error, and displays some documentation for that particular error. This will even work when the document is spread over several files. AUC TeX automatically indents your `LaTeX-source', not only as you write it -- you can also let it indent and format an entire document. It has a special outline feature, which can greatly help you `getting an overview' of a document. Apart from these special features, AUC TeX provides an large range of handy Emacs macros, which in several different ways can help you write your LaTeX documents fast and painless. ** redo.el-1.01, courtesy of Kyle Jones redo.el is a package that implements true redo mechanism in XEmacs buffers. Once you load it from your `.emacs', you can bind the `redo' command to a convenient key to use it. Emacs' normal undo system allows you to undo an arbitrary number of buffer changes. These undos are recorded as ordinary buffer changes themselves. So when you break the chain of undos by issuing some other command, you can then undo all the undos. The chain of recorded buffer modifications therefore grows without bound, truncated only at garbage collection time. The redo/undo system is different in two ways: *** The undo/redo command chain is only broken by a buffer modification. You can move around the buffer or switch buffers and still come back and do more undos or redos. *** The `redo' command rescinds the most recent undo without recording the change as a _new_ buffer change. It completely reverses the effect of the undo, which includes making the chain of buffer modification records shorter by one, to counteract the effect of the undo command making the record list longer by one. ** edmacro.el-3.10, courtesy of Dave Gillespie, ported to XEmacs by Hrvoje Niksic. Edmacro is a utility that provides easy editing of keyboard macros. Originally written by Dave Gillespie, it has been mostly rewritten by Hrvoje Niksic, in order to make it distinguish characters and integer, as well as to adapt it to XEmacs keysyms. Press `C-x C-k' to invoke the `edit-kbd-macro' command that lets you edit old as well as define new keyboard macros. You can also edit the last 100 keystrokes and insert them into a macro to be bound to a key or named as a command. The recorded/edited macros can be dumped to `.emacs' file. ** xmine.el-1.8, courtesy of Jens Lautenbacher XEmacs now includes a minesweeper game with a full-featured graphics and mouse interface. Invoke with `M-x xmine'. ** efs-1.15-x5 courtesy of Andy Norman and Michael Sperber EFS is now integrated with XEmacs, and replaces the old ange-ftp. It has many more features, including info documentation, support for many different FTP servers, and integration with dired. ** mic-paren.el-1.3.1, courtesy of Mikael Sjödin ** hyperbole-4.022, courtesy of Bob Weiner ** hm--html-menus-5.3, courtesy of Heiko Muenkel ** python-mode.el-2.90, courtesy of Barry Warsaw ** balloon-help-1.06, courtesy of Kyle Jones ** xrdb-mode.el-1.21, courtesy of Barry Warsaw ** igrep.el-2.56, courtesy of Kevin Rodgers ** frame-icon.el, courtesy of Michael Lamoureux and Bob Weiner ** itimer.el-1.05, courtesy of Kyle Jones ** VM-6.30, courtesy of Kyle Jones ** OO-Browser-2.10, courtesy of Bob Weiner ** viper-2.93, courtesy of Michael Kifer ** ediff-2.65, courtesy of Michael Kifer ** detached-minibuf-1.1, courtesy of Alvin Shelton ** whitespace-mode.el, courtesy of Heiko Muenkel ** winmgr-mode.el, courtesy of David Konerding, Stefan Strobel & Barry Warsaw ** fast-lock.el-3.11.01, courtesy of Simon Marshall ** lazy-lock.el-1.16, courtesy of Simon Marshall ** browse-cltl2.el-1.1, courtesy of Holger Schauer ** eldoc.el-1.10, courtesy of Noah Friedman ** tm-7.105, courtesy of MORIOKA Tomohiko ** verilog-mode.el-2.25, courtesy of Michael McNamara & Adrian Aichner ** overlay.el, courtesy of Joseph Nuspl ** live-icon.el-1.3, fixes courtesy of Karl Hegbloom ** tpu-edt.el, fixes courtesy of R. Kevin Oberman ** etags.c-11.86 Courtesy of F. Potortì * Lisp and internal changes in XEmacs 20.2 ========================================== ** `defcustom' and `defgroup' can now be used to specify types and placement of the user-settable variables. You can now specify the types of user-settable variables in your Lisp packages to be customized by users. To do so, use `defcustom' as a replacement for `defvar'. For example, the old declaration: (defvar foo-blurgoze nil "*non-nil means that foo will act very blurgozely.") can be rewritten as: (defcustom foo-blurgoze nil "*non-nil means that foo will act very blurgozely." :type 'boolean :group 'foo) From a package writer's point of view, nothing has been changed However, the user can now type `M-x customize RET foo-blurgoze RET' to customize the variable. Other, more complex data structures can be described with `defcustom' too, for instance: (defcustom foo-hairy-alist '((somekey . "somestring") (otherkey . (foo-doit)) (thirdkey . [1 2 3])) "*Alist describing the hairy options of the foo package. The CAR of each element is a symbol, whereas the CDR can be either a string, a form to evaluate, or a vector of integers. New Emacs users simply adore alists like this one." :type '(repeat (cons (symbol :tag "Key") (choice string (vector (repeat :inline t integer)) sexp))) :group 'foo) The user will be able to add and remove the entries to the list in a visually appealing way, as well as save the settings to his/her `.emacs'. Note that `defcustom' will also be included in GNU Emacs 19.35, and that both XEmacs and GNU Emacs will be using it in the future. Although the user-interface of customize may change, the Lisp interface will remain the same. This is why we recommend that you use `defcustom' for user-settable variables in your new Lisp packages. ** The `read-kbd-macro' function is now available. The `read-kbd-macro' function (as well as the read-time evaluated `kbd' macro) from the edmacro package is now available in XEmacs. For example: (define-key foo-mode-map (kbd "C-c <up>") 'foo-up) is completely equivalent to (define-key foo-mode-map [(control ?c) up] 'foo-up) The `kbd' macro is preferred over `read-kbd-macro' function , as it evaluates before compiling, thus having no loading overhead. Using `kbd' is not necessary for GNU Emacs compatibility (GNU Emacs supports the XEmacs-style keysyms), but adds to clarity. For example, (kbd "C-?") is usually easier to read than [(control ??)]. The full description of the syntax of keybindings accepted by `read-kbd-macro' is documented in the docstring of `edmacro-mode'. ** Overlay compatibility is implemented. The overlay support in XEmacs is now functional. Written by Joe Nuspl, the overlay compatibility library overlay.el is implemented on top of the native XEmacs extents, and can be used as a GNU Emacs-compatible way of changing display properties. ** You should use keysyms kp-* (kp-1, kp-2, ..., kp-enter etc.) rather than the old form kp_*. The new form is also compatible with GNU Emacs. ** The keysyms mouse-1, mouse-2, mouse-3 and down-mouse-1, down-mouse-2, and down-mouse-3 have been added for GNU Emacs compatibility. ** A new user variable `signal-error-on-buffer-boundary' has been added. Set this to variable to nil to avoid XEmacs usual lossage of zmacs region when moving up against a buffer boundary. ** lib-complete.el was MULE-ized. The commands `find-library', `find-library-other-window' and `find-library-other-frame' now take an optional coding system argument. ** Experimental support for Lisp reader macros #-, #+. The Common Lisp reader macros for feature test are now supported. This feature is present for evaluation purposes and is subject to change. ** `values' now has a setf method ** The `eval-after-load' and `eval-next-after-load' functions are now available. ** A bug that prevented `current-display-table' to be correctly set with `set-specifier' has been fixed. ** The bug in easymenu which prevented multiple menus from being accessible through button3 has been fixed. You can now safely use easymenu to define multiple menu entries in a compatible way, with the added menus accessible via button3 as local submenus. ** Many bugs in the scrollbar code have been fixed. ** First alpha level support of MS Windows NT is available, courtesy of David Hobley and Marc Paquette. ** Wnn/egg now has initial support Courtesy of Jareth Hein. ** Some old non-working code has been removed until someone chooses to work on it. This includes much of the NeXTStep stuff. The VMS support is also likely to be removed in the future. ** Many files have been purged out of the etc/ directory. If you still need the purged files, look for them in the GNU Emacs distribution. * Major Differences Between 19.14 and 20.0 =========================================== XEmacs 20.0 is the first public release to have support for MULE (Multi-Lingual Emacs). The --with-mule configuration flag must be used to enable Mule support. Many bugs have been fixed. An effort has been made to eradicate all XEmacs crashes, although we are not quite done yet. The overall quality of XEmacs should be higher than any previous release. XEmacs now compiles with nary a warning with some compilers. -- Multiple character sets can be displayed in a buffer. The file mule-doc/demo in the distribution contains a greeting in many different languages. -- Although the Mule work is for all languages, particular effort has been invested in Japanese, with particular focus on Japanese users of Sun WorkShop. Many menubar labels have been translated into Japanese. Martin Buchholz, the maintainer of MULE features within XEmacs normally runs XEmacs in a Japanese language environment. Some of the other contributors are Japanese, most importantly Morioka Tomohiko, author of the TM package, providing MIME support for Mail and News. -- Input for complex Asian languages is supported via XIM, a mechanism introduced in X11R5 to allow applications to get localized input without knowledge of the language. The way XIM works is that when the locale has a complex character set, such as Japanese, and extra minibuffer-like status window appears attached to various application windows, and indicates the status of the input method. Composed input in XEmacs should work the same as with other applications. If Motif and Mule support is configured into XEmacs, then XIM support is automatically configured in as well. -- TM (Tools for Mime) now comes with XEmacs. This provides MIME (Multipurpose Internet Mail Extensions) support for Mail and News. The primary author is Morioka Tomohiko. -- Japanese input can also be input using the `canna' input method. This support was contributed by Morioka Tomohiko. Setting up canna usually requires more user effort (and better knowledge of Japanese!) than XIM, but provides a better-integrated input method. -- A mini-tutorial on using Mule: -- Every time data passes between XEmacs and the rest of the environment, via file or process input or output, XEmacs must convert between its internal multi-character representation and the external representation (`coding system'). Many difficulties with Mule are related to controlling these coding system conversions. -- file-coding-system, file-coding-system-for-read, overriding-file-coding-system, and file-coding-system-alist are used to determine the coding systems used on file input and output. -- For each process, (set-process-input-coding-system) and (set-process-output-coding-system) determine the coding system used for I/O from the process. -- Many other things are encoded using pathname-coding-system: -- file and directory names -- window manager properties: window title, icon name -- process names and process arguments -- XIM input. -- In many cases, you will want to have the same values for all the above variables in many cases. For example, in a Japanese environment, you will want to use the 'euc-japan coding system consistently, except when running certain processes that do byte-oriented, rather than character-oriented I/O, such as gzip, or when processing Mail or News, where ISO2022-based coding systems are the norm, since they support multiple character sets. -- To add support for a new language or character set, start by trying to copy code in japanese-hooks.el. -- The traditional pre-Mule data conversion is equivalent to the 'binary coding system under Mule. In this case all characters are treated as iso8859-1 (i.e. characters for English + Western European languages). -- many fileio-related commands such as find-file and write-file take an extra argument, coding-system, which specifies the encoding to be used with the file on disk. For example, here is a command that converts from the Japanese EUC to ISO2022 format: xemacs -batch -eval '(progn (find-file "locale-start.el.euc" (quote euc-japan)) (write-file "locale-start.el" nil (quote iso-2022-8-unix)))' Interactively, you can be prompted for a coding system by providing a prefix argument to the fileio command. In particular, C-u C-x C-f is a useful sequence to edit a file using a particular coding system. -- In an Asian locale (i.e. if $LANG is set to ja, ko, or zh), XEmacs automatically sets up a language environment assuming that the operating system encodes information in the national version of EUC, which supports English and the national language, but typically no other character sets. -- Command line processing should work much better now - no more order dependencies. -- Many many package upgraded (thanks go to countless maintainers): -- ediff 2.64 (Michael Kifer) -- Gnus 5.2.40 (Lars Magne Ingebrigtsen) -- w3 3.0.51 (Bill Perry) -- ilisp 5.8 (Chris McConnell, Ivan Vasquez, Marco Antoniotti, Rick Campbell) -- VM 5.97 (Kyle Jones) -- etags 11.78 (Francesco Potorti`) -- ksh-mode.el 2.9 -- vhdl-mode.el 2.73 (Rod Whitby) -- id-select.el (Bob Weiner) -- EDT/TPU emulation modes should work now for the first time. -- viper 2.92 (Michael Kifer) is now the `official' vi emulator for XEmacs. -- big-menubar should work much better now. -- mode-motion+.el 3.16 -- backup-dir 2.0 (Greg Klanderman) -- ps-print.el-3.05 (Jacques Duthen Prestataire) -- lazy-lock-1.15 (Simon Marshall) -- reporter 3.3 (Barry Warsaw) -- hm--html-menus 5.0 (Heiko Muenkel) -- cc-mode 4.322 (Barry Warsaw) -- elp 2.37 (Barry Warsaw) -- Many new packages have been added: -- m4-mode 1.8 (Andrew Csillag) -- crisp.el - crisp/brief emulation (Gary D. Foster) -- Johan Vroman's iso-acc.el has been ported to XEmacs by Alexandre Oliva -- psgml-1.01 (Lennart Staflin, James Clark) -- python-mode.el 2.83 (Barry Warsaw) -- vrml-mode.el (Ben Wing) -- enriched.el, face-menu.el (Boris Goldowsky, Michael Sperber) -- sh-script.el (Daniel Pfeiffer) -- decipher.el (Christopher J. Madsen) -- New function x-keysym-on-keyboard-p helps determine keyboard characteristics for key rebinding: x-keysym-on-keyboard-p: (KEYSYM &optional DEVICE) -- a built-in function. Return true if KEYSYM names a key on the keyboard of DEVICE. More precisely, return true if pressing a physical key on the keyboard of DEVICE without any modifier keys generates KEYSYM. Valid keysyms are listed in the files /usr/include/X11/keysymdef.h and in /usr/lib/X11/XKeysymDB, or whatever the equivalents are on your system. -- Installed info files are now compressed (support courtesy of Joseph J Nuspl) -- (load-average) works on Solaris, even if you're not root. Thanks to Hrvoje Niksic. -- OffiX drag-and-drop support added -- lots of syncing with 19.34 elisp files, most by Steven Baur * For older news and for alternate news (the ones dealing with XEmacs 19.15 and 19.16), see the file ONEWS.