Support new IGNORE-INVALID-SEQUENCESP argument, #'query-coding-region. lisp/ChangeLog addition: 2009-02-07 Aidan Kehoe <kehoea@parhasard.net> * coding.el (query-coding-clear-highlights): Rename the BUFFER argument to BUFFER-OR-STRING, describe it as possibly being a string in its documentation. (default-query-coding-region): Add a new IGNORE-INVALID-SEQUENCESP argument, document that this function does not support it. Bind case-fold-search to nil, we don't want this to influence what the function thinks is encodable or not. (query-coding-region): Add a new IGNORE-INVALID-SEQUENCESP argument, document what it does; reflect this new argument in the associated compiler macro. (query-coding-string): Add a new IGNORE-INVALID-SEQUENCESP argument, document what it does. Support the HIGHLIGHT argument correctly. * unicode.el (unicode-query-coding-region): Add a new IGNORE-INVALID-SEQUENCESP argument, document what it does, implement this. Document a potential problem. Use #'query-coding-clear-highlights instead of reimplementing it ourselves. Remove some debugging messages. * mule/arabic.el (iso-8859-6): * mule/cyrillic.el (iso-8859-5): * mule/greek.el (iso-8859-7): * mule/hebrew.el (iso-8859-8): * mule/latin.el (iso-8859-2): * mule/latin.el (iso-8859-3): * mule/latin.el (iso-8859-4): * mule/latin.el (iso-8859-14): * mule/latin.el (iso-8859-15): * mule/latin.el (iso-8859-16): * mule/latin.el (iso-8859-9): * mule/latin.el (windows-1252): * mule/mule-coding.el (iso-8859-1): Avoid the assumption that characters not given an explicit mapping in these coding systems map to the ISO 8859-1 characters corresponding to the octets on disk; this makes it much more reasonable to implement the IGNORE-INVALID-SEQUENCESP argument to query-coding-region. * mule/mule-cmds.el (set-language-info): Correct the docstring. * mule/mule-cmds.el (finish-set-language-environment): Treat invalid Unicode sequences produced from invalid-sequence-coding-system and corresponding to control characters the same as control characters in redisplay. * mule/mule-cmds.el: Document that encode-coding-char is available in coding.el * mule/mule-coding.el (make-8-bit-generate-helper): Change to return the both the encode-program generated and the relevant non-ASCII charset; update the docstring to reflect this. * mule/mule-coding.el (make-8-bit-generate-encode-program-and-skip-chars-strings): Rename this function; have it return skip-chars-strings as well as the encode program. Have these skip-chars-strings use ranges for charsets, where possible. * mule/mule-coding.el (make-8-bit-create-decode-encode-tables): Revise this to allow people to specify explicitly characters that should be undefined (= corresponding to keys in unicode-error-default-translation-table), and treating unspecified octets above #x7f as undefined by default. * mule/mule-coding.el (8-bit-fixed-query-coding-region): Add a new IGNORE-INVALID-SEQUENCESP argument, implement support for it using the 8-bit-fixed-invalid-sequences-skip-chars coding system property; remove some debugging messages. * mule/mule-coding.el (make-8-bit-coding-system): This function is dumped, autoloading it makes no sense. Document what happens when characters above #x7f are not specified, implement this. * mule/vietnamese.el: Correct spelling. tests/ChangeLog addition: 2009-02-07 Aidan Kehoe <kehoea@parhasard.net> * automated/query-coding-tests.el: Add FAILING-CASE arguments to the Assert calls, making #'q-c-debug mostly unnecessary. Remove #'q-c-debug. Add new tests that use the IGNORE-INVALID-SEQUENCESP argument to #'query-coding-region; rework the existing ones to respect it.

: #-*- Perl -*-

# Create skeleton build tree
#
# Copyright (C) 2003 Ben Wing.
#
# This file is part of XEmacs.
#
# XEmacs 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.
#
# XEmacs 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.
#
# Author: Ben Wing <ben@xemacs.org>
#
# Synched up with: Not in FSF.

eval 'exec perl -w -S $0 ${1+"$@"}'
    if 0;

use File::Copy;
use File::Basename;
use Cwd;

die "Creates a skeleton build tree for use with SOURCE_DIR in config.inc.

Usage: $0 PATH
" if ($#ARGV);

my $path = $ARGV[0];
# Sometimes perl sucks, too.  To get the equivalent of expand-file-name
# in a reliable way, you have to do really weird shit, it seems.
my $cwd = cwd ();
$0 =~ s|\\|/|g;
chdir (dirname ($0));
my $srcroot = dirname (cwd ());

# Convert the path to MS Windows format if we're running Cygwin Perl.
chomp ($srcroot = `cygpath -w $srcroot`) if ($^O eq "cygwin");
$srcroot =~ s|/|\\|g;

chdir ($cwd);

print "Creating skeleton build tree in $path\n";
mkdir $path if ! -e $path;
mkdir "$path/nt" if ! -e "$path/nt";
copy("$srcroot/nt/xemacs.mak", "$path/nt/xemacs.mak") if ! -e "$path/nt/xemacs.mak";

&HackFile ("config.inc.samp");
&HackFile ("config.inc") if -e "$srcroot/nt/config.inc";

sub HackFile
{
  my $file = $_[0];
  if (! -e "$path/nt/$file")
    {	
      open IN, "<$srcroot/nt/$file";
      open OUT, ">$path/nt/$file";

      while (<IN>)
	{
	  # Must hack away CRLF junk.  Perl sucks again.  Wouldn't it be
	  # nice if perl handled this right??  Really can't be that hard!!!
	  s/\r\n/\n/g;

	  # hack the SOURCE_DIR line to point back to the source.
	  s!^# SOURCE_DIR=.*!SOURCE_DIR=$srcroot!;

	  print OUT;
	}

      close IN;
      close OUT;
    }
}