XEmacs 21.5.28 "fuki" is released.

* XEmacs 21.5.28 "fuki" is released. "fuki" is the twenty-ninth in the VEGETABLE series. The successor to XEmacs 21.5.27 "fiddleheads", "fuki" collects a large number of small changes accumulated over several months. For short descriptions of the changes made, see the "Brief Summary of Changes" below. For detailed information about each change, refer to the ChangeLogs appended to this message. "fuki" is intended to provide a stable checkpoint as a foundation for increased activity and possible instability over the next few months. I would appreciate people checking that it downloads or cvs updates properly, and that it builds on their platforms. Please use M-x build-report to send build reports to the xemacs-buildreports list. If there are download or build problems reported within this week, I will correct those with a prompt release of 21.5.29 by June 1. Otherwise, I plan to try to make timely beta releases at intervals of 2-6 weeks, depending on development activity. As far as I know, there are two expected warnings in the C code, both when building etags and its derivatives. These have to do with backward compatibility requirements of the upstream maintainer, and will not be corrected unless he does. The following warnings in elisp compilation have known causes and may be ignored for this release: - Deprecation warnings for set-charset-registry (those are Aidan's until he disclaims responsibility) - Warnings about premature inlining or undefined functions with names starting with "fc-" in font.el and the reference to xft-font-regexp (those are mine). There are two unexpected failures in the mule-tests regression test. I believe these are bugs in formulation of the tests themselves; they may be ignored. All other warnings should be treated as bugs, and reported. Patches are welcome too, of course! * General information about XEmacs 21.5: This is the development line. The current series started with XEmacs 21.5.0 (an alias for XEmacs 21.4.0 "Solid Vapor", the first release in the current stable line). 21.5 is the code base for introduction of major new subsystems and fixes to design bugs that experience shows will introduce instability. So far the main effort has been on improved support for Unicode, updates to the build infrastructure, and development of new features in memory allocation. For general information about XEmacs, the developers, and the user community, see our home page, http://www.xemacs.org/ * XEmacs 21.5.28 is "beta" software. The usual "no warranty" disclaimer (see etc/WARRANTY, sections 10 and 11) applies. At this point in time, it is the version that most developers are using for their daily work. However, it is certain that many bugs remain and new ones will be introduced as development proceeds. Be sure to take care to save your work often and follow a regular backup regime. * Availability Anonymous ftp: ftp://ftp.xemacs.org/pub/xemacs/xemacs-21.5 See http://www.xemacs.org/Install/ for more information about building from source. If you already have a 21.5.27 source tree, a patchkit is available in xemacs-21.5.27-21.5.28.patch.gz. This does not update .elcs or .infos, they will be rebuilt when you make XEmacs. If you have an earlier version, you can repeatedly apply patchkits. Also, if you don't have the packages yet, see http://www.xemacs.org/Documentation/packageGuide.html. Anonymous (pserver) CVS: Anonymous CVS is available. We are very grateful to the staff at our host, SunSITE.dk, for a lot of help and quick reponse to all our requests. Take care that your Root is set correctly to CVSROOT=:pserver:cvs＠cvs.xemacs.org:/pack/xemacscvs On platforms with a Bourne shell and find available, something like for r in `find . -name Root`; do echo $CVSROOT > $r; done will convert your entire tree. To make it straightforward to access certain versions of the code, we have a standard tagging policy. To update to release 21.5.28, use the update flag "-r r21-5-28". To update to the current release without referring to it specifically, use the flag "-r r21-5-latest-beta". To update to the latest commits in CVS, use the flag "-A". Then rebuild XEmacs. For more details, see http://www.xemacs.org/Develop/cvsaccess.html . * Brief summary of Changes to XEmacs 21.5.28 "fuki" Brief summary of Changes to 21.5.28 "fuki" Major Features and Backward Incompatible Changes -- Fix: Check for missing dimensions for default face and window -- Aidan Kehoe, Mike Sperber -- Fix: Crash in PNG error handler -- Stephen J. Turnbull, Ron Isaacson -- Fix: Crash in device-matching-specifier-list -- Stephen J. Turnbull -- Fix: Crash in gtk/x_reset_modifier_mapping -- Aidan Kehoe -- Fix: Crash in linux_play_data_or_file on 64-bit linux -- Hans Graff, Vin Shelton -- Fix: Crash on double-close of fd in pdump -- Steve Higham, Vin Shelton -- Fix: Crash via buffer overflow in doc.c -- Aidan Kehoe, Fabrice Popineau -- Fix: Crash via buffer overrun in init_native_sound -- Jerry James -- Fix: Crash when deleting dialog via window manager -- Stephen J. Turnbull -- Fix: Device type of msprinter is non-window-system -- Aidan Kehoe -- Fix: Float formats overflow output buffer -- Aidan Kehoe -- Fix: Font menu on non-Mule -- Aidan Kehoe -- Fix: ISO 2022 decoding kludges -- Aidan Kehoe -- Fix: Make snarf-documentation robust to nonconforming DOC files -- Aidan Kehoe -- Fix: Non-existent charset ID is specified for a mule-to-unicode call -- Aidan Kehoe -- Fix: Sanity checks before accessing frame in x_set_frame_properties -- Aidan Kehoe -- Fix: Type of arguments to GaugeMercury (related to progress bar crash?) -- Stephen J. Turnbull -- Fix: init_native_sound is called on a msprinter device -- Aidan Kehoe -- Fix: set-buffer-file-coding-system now sets buffer-modified-flag by default -- Aidan Kehoe -- New: Echo area resizing -- Adrian Aichner, Stephen J. Turnbull User-Visible Bug Fixes and Minor Improvements -- Fix: Avoid delays in TTY-only builds by checking dispatch event queue -- Aidan Kehoe -- Fix: Avoid looking up variable bindings during GC -- Nix -- Fix: Buffer overruns in gnuclient and gnuslib -- Jerry James -- Fix: Remove references to unimplemented input methods from language environments -- Aidan Kehoe -- Fix: Find modules correctly -- Ville Skyttä, Mike Sperber, Stephen Turnbull -- Fix: Get X11 window property correctly on 64-bit platforms -- Stephen J. Turnbull, Mike Fabian, Takashi Iwai -- Fix: Handle coding system in file insertion in buffer properly -- Mike Sperber -- Fix: Info broke invariant `(equal buffer-file-truename (file-truename buffer-file-name))' -- Stephen J. Turnbull, Nelson Ferreira -- Fix: Lots of tweaking of Windows read-only handling -- Vin Shelton, Benson Margulies -- Fix: Make --unmapped work again -- Malcolm Purvis -- Fix: Make copying of char tables work -- Olivier Galibert, Stephen J. Turnbull -- Fix: Prevent C-z in a gnuclient frame from suspending the process -- Aidan Kehoe -- Fix: Regex for finding command nodes -- Jeff Miller -- Fix: Reversion should check buffer-file-coding-system -- Aidan Kehoe -- Fix: Syntax of guillemets -- Aidan Kehoe -- Fix: Window configuration should not restore window position unless requested -- Nix -- Improve: Better version info in etags -- Stephen Turnbull, Steve Youngs -- Improve: Case table, syntax table information for Cyrillic, Greek -- Aidan Kehoe -- Improve: Face initialization from X resources -- Aidan Kehoe -- Improve: Give x-compose-map an entry for sharp S -- Aidan Kehoe -- Improve: Handling of Asian "full-width" characters on TTY -- Aidan Kehoe -- Improve: Language environment detection from locale -- Aidan Kehoe -- Improve: Look harder for X11 locale-specific app-defaults files -- Malcolm Purvis -- Improve: Make read-quoted-char terminate, not error, on non-character keys -- Aidan Kehoe -- Improve: Sort abbrev database -- Adrian Aichner -- Improve: Support non-ISO Cyrillic keysyms -- Aidan Kehoe -- Improve: Sync etags to pot version 17.26; improve version report -- Stephen J. Turnbull, Steve Youngs -- Improve: Sync etags to pot version 17.32; move improved version report to configure, lose pot version report -- Stephen J. Turnbull -- Improve: Try to rationalize resources of Xft in lwlib -- Stephen J. Turnbull -- Improve: Use short list fixed string registries, not long list pruned by regexp for X11 fonts -- Aidan Kehoe -- Improve: charsets-in-region now in C -- Aidan Kehoe -- Improve: what-cursor-position gives Unicode and Mule information on non-ASCII -- Aidan Kehoe -- New: Add raw string support like SXEmacs -- Aidan Kehoe -- New: Find C source files for Lisp primitives -- Aidan Kehoe -- New: Just-in-time definition of Unicode characters and X keysyms -- Aidan Kehoe -- Update: New package download sites -- Adrian Aichner -- Update: Sync etags to pot_etags_version 17.19 -- Stephen J. Turnbull -- Update: Sync pop[ch] to Emacs -- Jerry James -- Update: Update to 2003 version of ISO 8859-7 table -- Aidan Kehoe Build Infrastructure -- Fix: Allow lowercase file names in mswin SDK -- Benson Margulies -- Fix: Avoid using Motif for cygwin -- Dr. Volker Zell -- Fix: C is not C++ -- Robert Pluim -- Fix: Check for u_int*_t typedefs -- Dr. Volker Zell -- Fix: Correct help strings, remove obsolete configure.* files -- Stephen J. Turnbull -- Fix: Documentation of --with-error-checking options -- Stephen J. Turnbull -- Fix: Don't call ichar_to_unicode on non-MULE -- Mike Sperber -- Fix: Enable SYSTEM_MALLOC on ppc64, alpha and ia64 systems -- Malcolm Purvis -- Fix: Make user-defined directories work in configure -- Mike Sperber -- Fix: Register --enable and --with forms with the option checking list -- Malcolm Purvis -- Fix: Remove Autoconf 2.13-isms -- Stephen J. Turnbull -- Fix: Revert workaround for now-fixed Cygwin bug of missing d_ino field -- Dr. Volker Zell -- Fix: Typo in configure shell syntax -- Jerry James -- Fix: Use $EGREP, not egrep -- Malcolm Purvis -- Fix: Use new Cygwin XPM lib name -- Rick Rankin -- Fix: Work around m4_cdr change -- Malcolm Purvis -- Fix: `config.status --recheck' bug in Autoconf 2.60/2.61 -- Malcolm Purvis -- Fix: char type correctness -- Vin Shelton, Aidan Kehoe -- Fix: configure debugger init files correctly -- Stephen J. Turnbull -- Improve: Conditionalize GCC warnings on version -- Jerry James -- Improve: Create $srcdir/src/depend if missing -- Stephen J. Turnbull -- Improve: Detect Intel Macs, use POSIX-style write barrier in NewGC -- Marcus Crestani -- Improve: Refactor Canna detection -- Stephen J. Turnbull -- Improve: Setup kit configuration -- Vin Shelton improvement -- debugger init sources can use XCOMM convention -- New: Check for utilities that are not on "normal" Solaris PATH -- Stephen J. Turnbull, S L Baur Documentation -- Fix: Description of kanji read syntax, size of Ichar (now 21 bits) -- Aidan Kehoe -- Fix: Improve docstrings -- Stephen J. Turnbull, Aidan Kehoe -- Fix: Make manual conform to reality -- Malcolm Purvis, Robert Pluim, Aidan Kehoe, Stephen J. Turnbull -- Fix: New compatibility alias string-to-char-list -- Vin Shelton, Ville Skyttä -- Fix: Typo fixes in manual -- Stephen J. Turnbull, Aidan Kehoe -- Fix: Update FSF address -- Jerry James, Stephen Turnbull -- Fix: Update copyrights -- Stephen J. Turnbull -- Improve: Small rephrasing in TUTORIAL.de -- Adrian Aichner, hroptatyr -- New: Describe X errors when using ssh in PROBLEMS -- Malcolm Purvis -- New: Describe XFree86 crash in PROBLEMS -- Stephen J. Turnbull, Aidan Kehoe -- New: Describe how X11R7 loses x11/bitmaps/gray in PROBLEMS -- Stephen J. Turnbull -- Update: Addresses in files and ChangeLogs -- Adrian Aichner, Stephen J. Turnbull Lisp API -- Fix: Catch error in cloning language environment -- Aidan Kehoe -- Fix: Make shell-command use requested output buffer -- Jerry James -- fix: xft-version is now DEVAR_CONST_INT. -- Improve: Face initialization and specification -- Aidan Kehoe -- Improve: Make `split-char' available in no-mule -- Aidan Kehoe -- Improve: New unicode-type property of Unicode coding systems -- Aidan Kehoe -- Improve: Reorganize Mule lisp -- Aidan Kehoe -- New: Full access to FcConfig objects -- Stephen J. Turnbull -- New: ccl-compile-mule-to-unicode, ccl-compile-unicode-to-mule, ccl-dump-mule-to-unicode, ccl-dump-unicode-to-mule -- Aidan Kehoe -- New: deprecate set-charset-registry, charset-registry -- Aidan Kehoe -- New: font-lock-keywords-alist, font-lock-removed-keywords-alist, font-lock-add-keywords, font-lock-update-removed-keyword-alist, font-lock-remove-keywords -- Adrian Aichner -- New: set-charset-registries, charset-registries -- Aidan Kehoe -- New: x-coverage-instantiator specifier tag identifies Unicode fallbacks -- Aidan Kehoe -- Update: Sync font-lock-add-keywords and font-lock-remove-keywords from GNU Emacs. -- Adrian Aichner Internal API and Implementation -- Fix: Dead code elimination in font-mgr -- Stephen J. Turnbull -- Fix: Eliminate buffer overflows and NULL pointer references -- Jerry James, Adrian Aichner -- Fix: Memory leaks -- Jerry James -- Fix: Mule correctness with strings passed to warning functions -- Aidan Kehoe -- Fix: Syntax errors -- Aidan Kehoe -- Fix: Typo in string_direct_data_description -- Marcus Crestani -- Fix: Use defined APIs -- Stephen J. Turnbull, Aidan Kehoe -- Fix: Warning elimination -- Stephen J. Turnbull -- Improve: Cache Xrm coding system -- Aidan Kehoe -- Improve: GCPRO correctness -- Aidan Kehoe -- Improve: NewGC fault handling -- Marcus Crestani -- Improve: Reorganize and document separate_textual_runs -- Olivier Galibert, Stephen J. Turnbull -- Improve: 21-bit Mule chars to support JIT charsets -- Aidan Kehoe -- Update: Get Unicode tables with permissive license -- Stephen J. Turnbull, Mike Fabian Testing and Debugging -- Improve: Sort results of charset-in-* before comparing to expected. -- Aidan Kehoe -- Improve: gdbinit.in recognizes new FcConfig objects -- Stephen J. Turnbull -- New: Silence-Message macro in test-harness.el -- Stephen J. Turnbull -- New: Test for ISO-2022 decoder not choking on invalid UTF-8 -- Aidan Kehoe -- New: Test for copy-syntax-table bug -- Stephen Turnbull, Ralf Angeli -- New: Test for correct usage of escape-quoted cookie -- Aidan Kehoe -- New: Test for availability of input methods referenced in language environments -- Aidan Kehoe -- New: Test raw strings -- Aidan Kehoe -- New: Test Unicode non-BMP support -- Aidan Kehoe -------- ChangeLog Entries from ChangeLog -------- 2007-05-21 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * XEmacs 21.5.28 "fuki" is released. 2007-05-18 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * configure.ac (error-checking): Move comma misplaced by sorting. * configure: Regenerate. 2007-05-18 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * configure.ac (error-checking): Reorder arguments and sync doctring. 2007-03-14 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * PROBLEMS: describe crash when inserting or displaying a TAB. Thanks to Aidan Kehoe for the diagnosis. 2006-12-24 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * configure.ac (ar): Sun has it but hides it; check for it. (ssize_t): Use the modern check and document todo. 2007-02-16 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * configure.ac (canna): Move #define of CANNA_NEW_WCHAR_AWARE to config.h. Use -DCANNA_NEW_WCHAR_AWARE since check for RK.h fails otherwise. Refactor into loops over orthogonal tweaks (prefix and define). Use have_canna to track detection success, not with_canna. Add AC_MSG_WARNs documenting autoconf's bogosity (ours, too). 2007-01-01 Malcolm Purvis <malcolmp(a)xemacs.org&gt; * aclocal.m4 (XE_SHLIB_STUFF): Use $EGREP instead of egrep. 2006-12-28 Malcolm Purvis <malcolmp(a)xemacs.org&gt; * configure.ac: Ensure ac_configure_args contains a leading space to work around problems with 'config.status --recheck' under autoconf 2.60 and 2.61. 2006-12-27 Malcolm Purvis <malcolmp(a)xemacs.org&gt; * configure.ac (XE_MERGED_ARG): Register --enable and --with forms with the option checking list. 2006-12-19 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * Makefile.in.in (${srcdir}/configure): Generated from configure.ac. 2006-12-06 Dr. Volker Zell <Dr.Volker.Zell(a)oracle.com&gt; * configure.ac: Check for u_int*_t typedefs. 2006-12-06 Dr. Volker Zell <Dr.Volker.Zell(a)oracle.com&gt; * configure.ac: Avoid using Motif for cygwin. 2006-11-23 Mike Sperber <mike(a)xemacs.org&gt; * configure.ac (XE_EXPAND_VARIABLE): Fully expand the various directories before comparing them for figuring out which of them are user-defined. Use XE_EPXAND_VARIABLE macro created for this purpose where applicable. 2006-12-07 Rick Rankin <rick.rankin(a)yahoo.com&gt; * configure.ac: Add new library name for cygwin's xpm. 2006-12-07 Rick Rankin <rick.rankin(a)yahoo.com&gt; * configure.ac: Add new library name for cygwin's xpm. 2006-10-30 Malcolm Purvis <malcolmp(a)xemacs.org&gt; * configure.ac (XE_CDR): m4_cdr differs in autoconf 2.59 and 2.60. (XE_EXPAND_COMPLEX_OPTIONS): Use XE_CDR, not m4_cdr. 2006-07-18 Marcus Crestani <crestani(a)xemacs.org&gt; * configure.ac: Intel-based Macs can use POSIX-style write barrier. Use $machine to determine this. 2006-07-20 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * configure.ac (system-packages,legacy-packages): Fix help strings. * configure.in: cvs remove. * configure.usage: cvs remove. 2006-06-20 Jerry James <james(a)xemacs.org&gt; * configure.ac: When gcc is used, capture the values of __GNUC__ and __GNUC_MINOR__. Use them to conditionally add warning flags that are not supported by all versions of gcc. 2006-06-23 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * configure.ac (Build Makefile.in's from Makefile.in.in's): Create $srcdir/src/depend if it's missing. 2006-06-05 Jerry James <james(a)xemacs.org&gt; * configure.ac: Change "if -z" to "if test -z". 2006-05-23 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * configure.ac: Look for debugger init file templates in etc/, not src/. Debugger init files depend on config.h; recreate on every configure. Support XCOMM convention. Update FSF copyright. 2006-05-17 Malcolm Purvis <malcolmp(a)xemacs.org&gt; * PROBLEMS: Add entry concerning X errors when using ssh. 2006-05-17 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * PROBLEMS: X11R7 loses x11/bitmaps/gray. -------- ChangeLog Entries from etc/ChangeLog -------- 2007-05-21 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * XEmacs 21.5.28 "fuki" is released. 2007-05-01 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * gdbinit.in (pobj): Nuke fc_fontset and fc_objectset, add fc_config. 2007-03-24 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * ETAGS.EBNF: Sync to pot_etags_version 17.19. * etags.1: Sync to pot_etags_version 17.19. * ChangeLog.etags: New. Copy of upstream ChangeLog. 2007-01-28 Stephen J. Turnbull <stephen(a)xemacs.org&gt; Thanks to Mike Fabian <mfabian(a)suse.de&gt; for bringing the issue to my attention. * unicode/unicode-consortium/COPYING: * unicode/unicode-consortium/oreilly.html * unicode/unicode-consortium/unicode-consortium.html New files clarifying copying conditions for mapping tables. * unicode/unicode-consortium/GB12345.TXT * unicode/unicode-consortium/BIG5.TXT * unicode/unicode-consortium/CNS11643.TXT * unicode/unicode-consortium/HANGUL.TXT * unicode/unicode-consortium/JIS0201.TXT * unicode/unicode-consortium/JIS0208.TXT * unicode/unicode-consortium/JIS0212.TXT * unicode/unicode-consortium/JOHAB.TXT * unicode/unicode-consortium/KSC5601.TXT * unicode/unicode-consortium/KSX1001.TXT * unicode/unicode-consortium/OLD5601.TXT * unicode/unicode-consortium/SHIFTJIS.TXT Mapping tables downloaded from a Unicode, Inc. URL with permissive conditions. 2006-12-21 Aidan Kehoe <kehoea(a)parhasard.net&gt; * unicode/unicode-consortium/8859-7.TXT: Update the mapping to the 2003 version of ISO 8859-7. 2006-11-11 Aidan Kehoe <kehoea(a)parhasard.net&gt; * NEWS: Add information on set-charset-registries, and the change to define-specifier-tag. 2006-10-08 Adrian Aichner <adrian(a)xemacs.org&gt; * TUTORIAL.de: Small rephrasing as suggested by hroptatyr on #xemacs. 2006-07-19 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * etags.1: Sync to pot_etags_version 17.19. Thanks to Morgon Kanter <morgon(a)surgo.net&gt;. 2006-05-23 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * gdbinit.in: * dbxrc.in: Moved from src/. Update FSF address. Improve comments. * gdbinit.in: Use XCOMM for "don't edit" comment and copyright. -------- ChangeLog Entries from lib-src/ChangeLog -------- 2007-05-21 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * XEmacs 21.5.28 "fuki" is released. 2007-03-24 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * etags.c (print_version): More precise version info. Suggested by Steve Youngs. (print_help): Use #ifdef; PRINT_UNDOCUMENTED_OPTIONS_HELP may be undefined. 2007-03-24 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * etags.c: Sync to pot_etags_version 17.26. 2006-11-24 Benson Margulies <benson(a)dchbk.us&gt; * make-mswin-unicode.pl: allow for lower case header file names in the current platform SDK. 2006-08-11 Jerry James <james(a)xemacs.org&gt; * pop.h: Sync with Emacs. * pop.c: Ditto. 2006-08-08 Jerry James <james(a)xemacs.org&gt; * gnuslib.c (disconnect_from_server): shutdown() has been fine on Linux for a long time now; use it. Also, don't use length to access the buffer unless it is positive, not just nonzero. * gnuclient.c (filename_expand): Initialize the last array element to get a valid C string in case of overflow. Use strncat to avoid buffer overruns. * gnuclient.c (main): Use strncpy to avoid buffer overruns. 2006-07-21 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * etags.c: Xemacs -> XEmacs. 2006-07-19 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * etags.c: Sync to pot_etags_version 17.19. 2006-07-07 Jerry James <james(a)xemacs.org&gt; * config.values.in: Regenerate. -------- ChangeLog Entries from lisp/ChangeLog -------- 2007-05-21 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * XEmacs 21.5.28 "fuki" is released. 2007-05-20 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule/devanagari.el ("Devanagari"): * mule/ethiopic.el ("Ethiopic"): * mule/hebrew.el ("Hebrew"): * mule/lao.el ("Lao"): * mule/tibetan.el ("Tibetan"): * mule/vietnamese.el ("Vietnamese"): The Hebrew input method we don't provide, for less than stellar reasons; we should reconsider when and if we get bidi support. The other input methods require changes to mule-category.el that 21.4 doesn't have, and as such supporting them in the packages is mostly impossible. I've commented out the references to them in the corresponding language environment. * mule/mule-category.el (define-category): Add an optional argument to define-category, taken from GNU; it's still only a partial implementation, though. This whole file is chaos. * mule/mule-category.el (make-category-table): New. Trivial implementation, for GNU compatibility. 2007-05-20 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * gutter-items.el (progress-feedback-with-label): Clarify docstring. 2007-05-13 Adrian Aichner <adrian(a)xemacs.org&gt; * abbrev.el: Sort abbrev-table-name-list entries by name. Unlike GNU Emacs we keep tables sorted internally too, not only when writing them by `write-abbrev-file'. * abbrev.el (define-abbrev-table): Sort abbrev-table-name-list by table names, so that `insert-abbrevs', `list-abbrevs', and `write-abbrev-file' all present them in the same order. * abbrev.el (insert-abbrev-table-description): Removed. Losely synced to abbrev.c from GNU Emacs. 2007-04-30 Aidan Kehoe <kehoea(a)parhasard.net&gt; * code-files.el (set-buffer-file-coding-system): Make set-buffer-file-coding-system update the buffer's modified flag. Also make it accept a new flag, NOMODIFY, taken from GNU, to suppress this behaviour. * code-files.el (insert-file-contents): Use the NOMODIFY flag when calling set-buffer-file-coding-system. 2007-05-12 Aidan Kehoe <kehoea(a)parhasard.net&gt; * bytecomp.el (byte-compile-insert-header): Check for any Unicode escapes in the source file text when deciding whether Mule support is necessary for it, and whether to use escape-quoted as the .elc coding system. Thanks to Stephen for the suggestion as to how to ignore appearances in comments. 2007-05-01 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * dumped-lisp.el (preloaded-file-list): Move resize-minibuffer before simple. * resize-minibuffer.el: Remove CVS "Id" cookie.. * resize-minibuffer.el (resize-minibuffer-mode): Remove autoload. 2007-04-30 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * simple.el (raw-append-message): Improve resizing of echo area --- now obeys resize-minibuffer conventions. * resize-minibuffer.el (resize-minibuffer-idle-height): New. * simple.el (undisplay-echo-area-resize-window-allowed): New. * simple.el (undisplay-echo-area-resize-window): New. Add function to shrink echo area when idle. (incomplete) * simple.el (log-message-ignore-regexps): * simple.el (undisplay-echo-area-function): * simple.el (clear-message): * simple.el (append-message): * simple.el (display-message): Improve docstrings. 2007-04-22 Aidan Kehoe <kehoea(a)parhasard.net&gt; * specifier.el (device-type-matches-spec): Add `msprinter' to the type of devices that are not window systems. * specifier.el (derive-device-type-from-tag-set): Remove a superflous empty line. * specifier.el (derive-device-type-from-locale-and-tag-set): If CURRENT-DEVICE is provided, only ever return its type (if it matches TAG-SET) or nil (if it doesn't). 2007-01-02 Aidan Kehoe <kehoea(a)parhasard.net&gt; * cus-face.el (custom-set-face-update-spec): Fix some formatting. * faces.el (reset-face): reset-face resets other faces to behave like the default face--it shouldn't do anything if it's handed the default face. * font-menu.el: * font-menu.el (font-menu-set-font): If the font was initialised from X resources (the tag-set contains 'x-resource) pretend to Custom that it has responsibility for those settings. * x-faces.el: Add a couple of fontconfig functions to the globally-declare-fboundp, to eliminate a couple of byte-compile warnings. * x-faces.el ('x-resource)): New specifier tag, used to mark when a face's font or other attributes have been initialised from X resources. * x-faces.el (x-init-face-from-resources): Use the new specifier tag; also, instead of using a fragile explicit list of what would incidentally be specified for the X11 Unicode fonts in faces.c, pay attention to the new specifier tag created in that file for the specific purpose of grouping those instantiators together. 2007-04-22 Aidan Kehoe <kehoea(a)parhasard.net&gt; * x-font-menu.el (x-reset-device-font-menus-core): Only call charset-registries in the font menu if it's available. Restores font menu functionality on non-Mule. 2007-03-29 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * diagnose.el (show-object-memory-usage-stats): Fix docstring typo. 2007-01-23 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * cus-face.el (custom-theme-reset-faces): Move code after docstring. 2007-01-22 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * subr.el (lambda): Improve docstring. 2007-01-21 Ville Skyttä <scop(a)xemacs.org&gt; * mule/canna-leim.el (canna-activate): Look for canna_api, not canna/canna_api. 2007-01-27 Mike Sperber <mike(a)xemacs.org&gt; * setup-paths.el (paths-module-load-path-depth): Add. * setup-paths.el (paths-construct-module-load-path): For `module-load-path', use `paths-module-load-path-depth', not `paths-core-load-path-depth'. 2006-12-30 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule/cyrillic.el ("Cyrillic-ISO"): * mule/english.el ("English"): * mule/greek.el ("Greek"): Specify a native-coding-system for each of these language environments--since that is what is examined and relevant when sniffing the locale at startup. * mule/mule-cmds.el (create-variant-language-environment): A language environment's coding-priority is a list of coding systems, not coding categories. Treat it as such. 2006-12-21 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule/cyrillic.el: * mule/cyrillic.el (iso-8859-5): * mule/cyrillic.el (cyrillic-koi8-r-encode-table): Add syntax, case support for Cyrillic; make some parentheses more Lispy. * mule/european.el: Content moved to latin.el, file deleted. * mule/general-late.el: If Unicode tables are to be loaded at dump time, do it here, not in loadup.el. * mule/greek.el: Add syntax, case support for Greek. * mule/latin.el: Move the content of european.el here. Change the case table mappings to use hexadecimal codes, to make cross reference to the standards easier. In all cases, take character syntax from similar characters in Latin-1 , rather than deciding separately what syntax they should take. Add (incomplete) support for case with Turkish. Remove description of the character sets used from the language environments' doc strings, since now that we create variant language environments on the fly, such descriptions will often be inaccurate. Set the native-coding-system language info property while setting the other coding-system properties of the language. * mule/misc-lang.el (ipa): Remove the language environment. The International Phonetic _Alphabet_ is not a language, it's inane to have a corresponding language environment in XEmacs. * mule/mule-cmds.el (create-variant-language-environment): Also modify the coding-priority when creating a new language environment; document that. * mule/mule-cmds.el (get-language-environment-from-locale): Recognise that the 'native-coding-system language-info property can be a list, interpret it correctly when it is one. 2006-12-21 Aidan Kehoe <kehoea(a)parhasard.net&gt; * coding.el (coding-system-category): Use the new 'unicode-type property for finding what sort of Unicode coding system subtype a coding system is, instead of the overshadowed 'type property. * dumped-lisp.el (preloaded-file-list): mule/european.el has been removed. * loadup.el (really-early-error-handler): Unicode tables loaded at dump time are now in mule/general-late.el. * simple.el (count-lines): Add some backslashes to parentheses in docstrings to help fontification along. * simple.el (what-cursor-position): Wrap a line to fit in 80 characters. * unicode.el: Use the 'unicode-type property, not 'type, for setting the Unicode coding-system subtype. 2006-12-19 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * find-paths.el (paths-construct-emacs-directory): (paths-for-each-emacs-directory): (paths-find-site-directory): (paths-find-site-directories): (paths-for-each-version-directory): (paths-find-version-directory): (paths-find-version-directories): (paths-find-architecture-directory): Document ROOT(S) argument correctly. 2006-12-17 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule/cyrillic.el: Add case table, syntax table information for cyrillic-iso8859-5. 2006-12-17 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule/greek.el: Support case tables for greek-iso8859-7; make different choices on the relevant syntax. 2006-12-17 Aidan Kehoe <kehoea(a)parhasard.net&gt; * x-faces.el (x-init-face-from-resources): Only retain the fallbacks for the default face if we're not running on XFT. 2006-12-11 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule/mule-cmds.el (create-variant-language-environment): Avoid an error when creating a modified lang environment; use the old environment's string name rather than its value when calling replace-in-string. 2006-12-06 Aidan Kehoe <kehoea(a)parhasard.net&gt; * simple.el (what-cursor-position): For non-ASCII characters, give details on what a character maps to in Unicode, and its Mule charsets and codes, instead of simply its integer code point in this XEmacs. 2006-11-30 Mike Sperber <mike(a)xemacs.org&gt; * code-files.el (insert-file-contents): Call the file-name handler from Lisp, not from `insert-file-contents-internal', which is too late. 2006-11-28 Mike Sperber <mike(a)xemacs.org&gt; * files.el (revert-buffer-internal): Determine the coding system while still in the original buffer, rather than the new one. 2006-11-29 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule/mule-charset.el: * mule/mule-charset.el (charset-registry): * mule/mule-charset.el (set-charset-registry): * mule/mule-charset.el (charset-registries): New. Make set-charset-registry, charset-registry obsolete; define a setf for charset-registries. 2006-11-28 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule/cyrillic.el ("Cyrillic-KOI8"): * mule/cyrillic.el ("Cyrillic-ALT"): Add information on the native coding system of the machine to the language environment definition for Cyrillic. * mule/general-late.el: New file, for dumped Mule code that needs to be run after the language support has been loaded. * mule/mule-cmds.el: * mule/mule-cmds.el (set-language-info-alist): Return the new language environment name instead of nil. * mule/mule-cmds.el (langenv-to-locale-hash): Removed. This was relevant because coding_system_of_xrm_database called get-language-environment-from-locale 1307 times on startup, so the hash table made a difference. I've changed c_s_o_x_d to normally not call Lisp, and that makes this caching unnecessary. * mule/mule-cmds.el (posix-charset-to-coding-system-hash): New. A map from charsets as found in POSIX locales, with non-alphanumeric character stripped, to XEmacs coding systems. * mule/mule-cmds.el (parse-posix-locale-string): New. Parse a POSIX locale string into a language, region, charset, modifiers quad. * mule/mule-cmds.el (create-variant-language-environment): New. Create a version of a language environment which differs in its name and in the associated coding systems from a given language environment. * mule/mule-cmds.el (get-language-environment-from-locale): Rework to better pay attention to the POSIX locale, and to create language environments on the fly if the coding system of a given language differs from that available in the environment. * mule/mule-cmds.el (set-language-environment-coding-systems): Update a comment. 2006-11-28 Aidan Kehoe <kehoea(a)parhasard.net&gt; * dumped-lisp.el (preloaded-file-list): Load mule/general-late when we're in a Mule build. 2004-06-28 Nix <nix(a)esperi.org.uk&gt; * cmdloop.el (truncate-command-history-for-gc): Delay execution of all things that look up variable bindings, via `enqueue-eval-event'. 2006-11-14 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * info.el (Info-find-file-node, Info-insert-dir) (Info-read-subfile, Info-insert-file-contents): Maintain invariant `(equal buffer-file-truename (file-truename buffer-file-name))'. Thanks to Nelson Ferreira <nelson.ferreira(a)ieee.org&gt; for report and discussion. 2006-11-15 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule/mule-charset.el: * mule/mule-charset.el (charsets-in-string): Implement it in terms of charsets-in-region. * mule/mule-charset.el (charsets-in-region): Removed. It's now in C. 2006-11-07 Aidan Kehoe <kehoea(a)parhasard.net&gt; * unicode.el: Re-introduce ccl-encode-to-ucs-2; it's still being used by the JIT-UCS charsets, despite what I thought. Thank you Ilya. 2006-11-05 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule/arabic.el (arabic-digit): * mule/arabic.el (arabic-1-column): * mule/arabic.el (arabic-2-column): * mule/chinese.el (make-chinese-cns11643-charset): * mule/chinese.el (chinese-sisheng): * mule/english.el (ascii-right-to-left): * mule/ethiopic.el (ethiopic): * mule/european.el (latin-iso8859-14): * mule/european.el (latin-iso8859-16): * mule/indian.el (indian-is13194): * mule/indian.el (indian-1-column): * mule/indian.el (indian-2-column): * mule/japanese.el (japanese-jisx0213-1): * mule/japanese.el (japanese-jisx0213-2): * mule/lao.el (lao): * mule/misc-lang.el (ipa): * mule/mule-charset.el: * mule/thai-xtis.el (thai-xtis): * mule/tibetan.el (tibetan-1-column): * mule/tibetan.el (tibetan): * mule/vietnamese.el (vietnamese-viscii-lower): * mule/vietnamese.el (vietnamese-viscii-upper): Stop using the `registry' charset property; use `registries' instead. The difference is that registries is an ordered vector of X11 registries and encodings rather than a regexp; this means we can leave the matching to the X11 server, avoiding transferring huge amounts of data (perhaps across the network!) in order to do a regexp search on it. * mule/mule-charset.el (charset-registries): New. charset-registries returns the registries of a charset; * mule/mule-charset.el (set-charset-registry): Moved here from C. 2006-11-05 Aidan Kehoe <kehoea(a)parhasard.net&gt; * faces.el (face-property-matching-instance): Simplify. * faces.el (face-font-instance): Document CHARSET. * faces.el (set-face-font): Give more details on common values for font instantiators, LOCALEs. * unicode.el: Remove a few comments that were only relevant to GNU Emacs. * unicode.el (decode-char): * unicode.el (encode-char): Document CODE, CHAR using uppercase, since they're parameters. Update commentary on GNU's mule-unicode charsets and how we've solved the same problem. * x-faces.el (x-init-face-from-resources): Retain some of the fallbacks in the generated default face, since it doesn't make sense to try Andale Mono's ISO-10646-1 encoding for Amharic or Thai. * x-font-menu.el (charset-registries): * x-font-menu.el (x-reset-device-font-menus-core): Use charset-registries instead of charset-registry. 2006-11-02 Adrian Aichner <adrian(a)xemacs.org&gt; * font-lock.el: Sync font-lock-add-keywords and font-lock-remove-keywords from GNU Emacs. * font-lock.el (font-lock-keywords-alist): New. * font-lock.el (font-lock-removed-keywords-alist): New. * font-lock.el (font-lock-add-keywords): New. * font-lock.el (font-lock-update-removed-keyword-alist): New. * font-lock.el (font-lock-remove-keywords): New. 2006-10-28 Adrian Aichner <adrian(a)xemacs.org&gt; * simple.el (raw-append-message): Implement minibuffer resizing based on requirements of echo area content. 2006-10-28 Nix <nix(a)esperi.org.uk&gt; * window-xemacs.el (window-configuration-includes-position): New. * window-xemacs.el (window-configuration-equal): Use it: window configurations with distinct positions are equal by default. * window-xemacs.el (really-set-window-configuration): Do not restore window positions unless requested. 2006-10-30 Malcolm Purvis <malcolmp(a)xemacs.org&gt; * startup.el (display-splash-screen): No longer bring buffer the front. Fixes problems with -unmapped. * startup.el (xemacs-splash-buffer): Pop splash buffer to the front here instead. 2006-10-27 Aidan Kehoe <kehoea(a)parhasard.net&gt; * files.el (revert-buffer-internal): When reverting, use the buffer-file-coding-system of the file's buffer when determing whether something has changed on disk. 2006-10-14 Jeff Miller <jmiller(a)xemacs.org&gt; * info.el (Info-find-emacs-command-nodes): fix regex for find command node. 2006-08-10 Vin Shelton <acs(a)xemacs.org&gt; * subr.el: Move string-to-char-list to obsolete.el. * obsolete.el (string-to-char-list): Move obsolete definition here, so dumping can succeed. 2006-08-07 Ville Skyttä <scop(a)xemacs.org&gt; * subr.el (string-to-char-list): New backwards compatibility alias for `string-to-list'. 2006-08-07 Aidan Kehoe <kehoea(a)parhasard.net&gt; * frame.el (suspend-or-iconify-emacs): * frame.el (suspend-emacs-or-iconify-frame): Elaborate on what the functions do, hopefully preventing misunderstandings in the future. * keydefs.el (global-tty-map): "\C-z" is suspend-or-iconify-emacs on the TTY, not suspend-emacs. Fixes problems that arose where pressing C-z in a gnuclient frame suspended the whole process. 2006-08-05 Aidan Kehoe <kehoea(a)parhasard.net&gt; * lisp.el (forward-sexp): Handle raw strings specially just as we do structures. Fixes problems evaluating them in *scratch*. 2006-08-04 Aidan Kehoe <kehoea(a)parhasard.net&gt; * iso8859-1.el: Move the symbol table modification to three lines at the end of syntax.c. 2006-07-16 Aidan Kehoe <kehoea(a)parhasard.net&gt; * loadhist.el (symbol-file): Use `source-directory,' not `build-root' for the full pathnames of C files. * loadup.el: * update-elc.el: Rename `build-root', `source-root' to `build-directory', `source-directory' respectively, the latter for compatibility with the FSF; give them and `source-lisp' docstrings, change the defvars to defconsts where possible. * make-docfile.el (build-root): Removed. * make-docfile.el (build-directory): New. Rename build-root. * update-elc.el (build-root): Removed. * update-elc.el (build-directory): New. * update-elc.el (source-root): Removed. * update-elc.el (source-directory): New. Rename build-root, source-root. 2006-07-13 Aidan Kehoe <kehoea(a)parhasard.net&gt; * unicode.el (decode-char): * unicode.el (encode-char): Specify the SHOW-ARGS properly in a couple of assertions. 2006-06-30 Jerry James <james(a)xemacs.org&gt; * process.el (shell-command): If a specific output buffer was requested, use it. 2006-06-25 Aidan Kehoe <kehoea(a)parhasard.net&gt; * x-init.el (x-initialize-keyboard): No longer warn about Unicode keysyms without a mapping, since we now have valid mappings for all Unicode keysyms. 2006-06-25 Aidan Kehoe <kehoea(a)parhasard.net&gt; * cmdloop.el (read-quoted-char): Don't error on non-character keys, accept them as terminating the sequence. 2006-06-18 Aidan Kehoe <kehoea(a)parhasard.net&gt; * x-compose.el (compose-diaeresis-map): Add an entry to map dead_diaeresis + s to ß (LATIN SMALL LETTER SHARP S). 2006-06-03 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule/mule-ccl.el: "X Emacs" -> "XEmacs" * mule/mule-ccl.el (ccl-compile-mule-to-unicode): New. * mule/mule-ccl.el (ccl-compile-unicode-to-mule): New. * mule/mule-ccl.el (ccl-dump-mule-to-unicode): New. * mule/mule-ccl.el (ccl-dump-unicode-to-mule): New. * mule/mule-ccl.el (define-ccl-program): Add two new CCL commands, and commands to describe them; document them. 2006-06-03 Aidan Kehoe <kehoea(a)parhasard.net&gt; * unicode.el: * unicode.el (featurep): Define a CCL program to translate characters to Unicode, for use in redisplay with the fall-back Unicode charsets. * x-init.el (x-initialize-keyboard): Only warn about unknown Unicode keysyms if Mule is available; we assume that people who've chosen non-Mule are okay with losing data. 2006-06-03 Adrian Aichner <adrian(a)xemacs.org&gt; * package-get.el (package-get-download-sites): Welcome nl.xemacs.org and Thanks! * package-get.el (package-get-pre-release-download-sites): Ditto. 2006-05-24 Jerry James <james(a)xemacs.org&gt; * autoload.el (batch-update-autoloads): Only write to generated-autoload-file. 2006-05-23 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * code-cmds.el: Update FSF copyright; improve header comment. (coding-system-change-eol-conversion): Clarify, add sanity check. Fix keyword mismatch between function arg and subsidiary accessor. -------- ChangeLog Entries from lwlib/ChangeLog -------- 2007-05-21 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * XEmacs 21.5.28 "fuki" is released. 2007-05-20 Stephen J. Turnbull <stephen(a)xemacs.org&gt; Gauge values are signed integers (ints). * xlwgauge.h: Get rid of references to Cardinal in comment. (XawGaugeGetValue): Declare return value as int. (XawGaugeSetValue): Declare value as int. * xlwgauge.c (GaugeGetValue): Declare value as int. (GaugeMercury): Declare val0 and val1 as int. Remove redundant casts. (XawGaugeGetValue): Declare return value as int. (XawGaugeSetValue): Declare value as int. 2007-05-17 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * lwlib-Xaw.c (wm_delete_window): Iterate over children of shell (there may be more than one) to find our widget. 2006-11-17 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * xlwmenu.c (xlwMenuResources): * xlwmenu.c (XlwMenuInitialize): * xlwmenuP.h (_XlwMenu_part): * xlwtabs.c (resources): * xlwtabs.c (TabsInit): * xlwtabsP.h (TabsPart): New xftFontName member in each widget part struct, corresponds to xftFont String resource. Initialize renderFont private member from fcFontName if non-NULL, otherwise initialize from xftFontName. * xlwtabs.h: Update parameter table. 2006-06-16 Jerry James <james(a)xemacs.org&gt; * lwlib-Xlw.c (xlw_scrollbar_callback): Do not dereference instance before checking whether it is NULL. * xlwmenu.c (xlw_map_menu): Prevent uninitialized access to root and waste. -------- ChangeLog Entries from man/ChangeLog -------- 2007-05-21 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * XEmacs 21.5.28 "fuki" is released. 2007-05-15 Aidan Kehoe <kehoea(a)parhasard.net&gt; * internals/internals.texi: Replace an inaccurate description of the read syntax of a Kanji character with one using the recently-added Unicode escapes. Also update the size of an Ichar; they're now 21-bit integers, not 19-bit integers. 2007-04-30 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * internals/internals.texi (Creating a New Console/Device/Frame Type): Typo fix. 2007-01-01 Malcolm Purvis <malcolmp(a)xemacs.org&gt; * internals/internals.texi (Ben's README): Use 'grep -F' instead of fgrep. 2006-11-07 Robert Pluim <rpluim(a)gmail.com&gt; * lispref/os.texi (User Identification): The code uses HOMEPATH, not HOMEDIR. 2006-11-29 Aidan Kehoe <kehoea(a)parhasard.net&gt; * xemacs/custom.texi (Face Resources): Mention that the user should use full XLFD forms for specifying fonts, and that Mule builds reject the short forms by default. Also mention the work-around to the latter design choice that was implemented for Ilya. 2006-11-16 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * internals/internals.texi (Better Rendering Support -- Configuration with the Interim Patches): Fix examples of configuration via X resources. (Better Rendering Support -- Implementation): Fix description of xftFont resources, introduce fcFontName resource. 2006-11-11 Aidan Kehoe <kehoea(a)parhasard.net&gt; * lispref/faces.texi (Face Convenience Functions): Add information on how to specify a face's font for a given Mule charset. * lispref/specifiers.texi (Specifiers): * lispref/specifiers.texi (Simple Specifier Usage): * lispref/specifiers.texi (Specifiers In-Depth): * lispref/specifiers.texi (Specifier Tag Functions): * lispref/specifiers.texi (Specifier Instantiation Functions): Update the documentation of specifiers to reflect the new support for Mule character sets and associating tags with them. 2006-08-05 Aidan Kehoe <kehoea(a)parhasard.net&gt; * lispref/objects.texi (String Type): Give details of the raw string syntax, taken from SXEmacs and Python. 2006-07-19 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * new-users-guide/edit.texi (Insert): Document bogosity in vendor labeling of DEL key. (Numeric Argument): Remove spurious RETs from keystroke examples. Thanks to Michael C. Wescott <wescott(a)sc.rr.com&gt;. 2006-07-16 Aidan Kehoe <kehoea(a)parhasard.net&gt; * xemacs/custom.texi (File Variables): Make it clearer that file variables are buffer-local. * xemacs/custom.texi (Faces): "must be encoding" -> "must be encoded". 2006-07-16 Aidan Kehoe <kehoea(a)parhasard.net&gt; * lispref/variables.texi (Creating Buffer-Local): Mention that buffer-local variables are created when file local variables are set. 2006-07-08 Aidan Kehoe <kehoea(a)parhasard.net&gt; * internals/internals.texi (Internal String Encoding): Mention that UTF-8 would be a reasonable alternative encoding. * internals/internals.texi (Internal Character Encoding): Re-arrange the description of characters to deal with 21-bit characters. 2006-06-03 Aidan Kehoe <kehoea(a)parhasard.net&gt; * lispref/mule.texi (CCL Syntax): * lispref/mule.texi (CCL Statements): Describe the mule-to-unicode and unicode-to-mule statements; rename the section they are described in. 2006-05-17 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * xemacs-faq.texi (Q2.2.3): New node. (Q2.2.2, Q2.3.1): Fix navigation references. (Top, Installation): Add to menus. -------- ChangeLog Entries from modules/ChangeLog -------- 2007-05-21 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * XEmacs 21.5.28 "fuki" is released. 2007-02-16 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * canna/canna_api.c: Move CANNA_NEW_WCHAR_AWARE to config.h. Clean up ancient cruft for IROHA (Canna v.1) support. 2007-02-08 Adrian Aichner <adrian(a)xemacs.org&gt; * postgresql/postgresql.c: Update Steve L. Baur's address on his request. * postgresql/postgresql.h: Ditto. -------- ChangeLog Entries from netinstall/ChangeLog -------- 2007-05-21 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * XEmacs 21.5.28 "fuki" is released. -------- ChangeLog Entries from nt/ChangeLog -------- 2007-05-21 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * XEmacs 21.5.28 "fuki" is released. 2006-12-11 Vin Shelton <acs(a)xemacs.org&gt; * config.inc.samp: Added BUILD_FOR_SETUP_KIT. * xemacs.mak: Use BUILD_FOR_SETUP_KIT to define OK_TO_USE_MSVCRTD. -------- ChangeLog Entries from nt/installer/Wise/ChangeLog -------- 2007-05-21 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * XEmacs 21.5.28 "fuki" is released. -------- ChangeLog Entries from src/ChangeLog -------- 2007-05-21 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * XEmacs 21.5.28 "fuki" is released. 2007-05-13 Adrian Aichner <adrian(a)xemacs.org&gt; * abbrev.c: Sort abbreviations by name, similar to GNU Emacs. * abbrev.c (write_abbrev): Losely ported from GNU Emacs. * abbrev.c (describe_abbrev): Ditto. * abbrev.c (Finsert_abbrev_table_description): Ditto. 2007-05-03 Vin Shelton <acs(a)xemacs.org&gt; * dumper.c (pdump): Don't close an already-closed file descriptor. Patch from Steve Higham. 2007-05-17 Vin Shelton <acs(a)xemacs.org&gt; * linuxplay.c (linux_play_data_or_file): Fix playing sound on 64-bit linux. Pathc from Hans de Graaff. 2007-05-18 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * config.h.in (ERROR_CHECK_BYTE_CODE): Alphabetize ERROR_CHECK_*. 2007-04-30 Aidan Kehoe <kehoea(a)parhasard.net&gt; * unicode.c: * unicode.c (encode_unicode_char_1): * unicode.c (unicode_convert): Support non-BMP characters in UTF-16. 2007-05-12 Aidan Kehoe <kehoea(a)parhasard.net&gt; * event-Xt.c (x_reset_modifier_mapping): * event-gtk.c (gtk_reset_modifier_mapping): Zero out the device's modifier map once we've freed it, to prevent a double free on a re-entrant call. 2007-05-01 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * font-mgr.c (Qfontsetp): Remove unused declaration. * font-mgr.c (xft-version): Make it a DEFVAR_CONST_INT. * font-mgr.c (fc-version): New: fontconfig.h version. 2007-01-02 Aidan Kehoe <kehoea(a)parhasard.net&gt; * faces.c: * faces.c (syms_of_faces): * faces.c (complex_vars_of_faces): New symbol and corresponding specifier tag, x-coverage-instantiator, used to group those fonts used for their extensive coverage for obscure characters in x-faces.el. 2007-04-16 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * redisplay-x.c (separate_textual_runs_nomule): Oops. We agreed that memcpy didn't work, and used it anyway. Fix it. 2007-04-15 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * font-mgr.c (DestroyFontsetP): New enum. * font-mgr.c (fontset_to_list): Add destroyp argument of that type. * font-mgr.c (Ffc_font_list): Add argument. * font-mgr.c (Ffc_font_sort): Add argument. * font-mgr.h (fc_config): Declare new Lisp object type. * font-mgr.c (print_fc_config): * font-mgr.c (finalize_fc_config): * font-mgr.c (fc_config_p): * font-mgr.c (Qfc_configp): Implement it. * font-mgr.c (syms_of_font_mgr): * font-mgr.c (complex_vars_of_font_mgr): * lrecord.h (lrecord_type_fc_config): Initialize it. * font-mgr.c (fc_config_create_using): New helper function. * font-mgr.c (FCSTRLIST_TO_LISP_USING): New helper macro. * font-mgr.c (Vfc_config_weak_list): Manage references to FcConfigs. * font-mgr.c (Ffc_get_version): * font-mgr.c (Ffc_config_create): * font-mgr.c (Ffc_config_get_current): * font-mgr.c (Ffc_config_set_current): * font-mgr.c (Ffc_config_up_to_date): * font-mgr.c (Ffc_config_build_fonts): * font-mgr.c (Ffc_config_get_config_dirs): * font-mgr.c (Ffc_config_get_font_dirs): * font-mgr.c (Ffc_config_get_config_files): * font-mgr.c (Ffc_config_get_cache): * font-mgr.c (Ffc_config_get_rescan_interval): * font-mgr.c (Ffc_config_set_rescan_interval): * font-mgr.c (Ffc_config_app_font_add_file): * font-mgr.c (Ffc_config_app_font_add_dir): * font-mgr.c (Ffc_config_app_font_clear): * font-mgr.c (Ffc_config_filename): * font-mgr.c (Ffc_init_load_config): * font-mgr.c (Ffc_init_load_config_and_fonts): * font-mgr.c (Ffc_init): * font-mgr.c (Ffc_init_reinitialize): Implemented operations. * font-mgr.c (Ffc_config_destroy): * font-mgr.c (Ffc_config_get_blanks): Stub operations. 2007-04-15 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * font-mgr.c: Update FSF copyrights. * font-mgr.c (print_fc_pattern): New facility for fc-pattern objects. Comment on need/implementation for equal/hash. * font-mgr.c (build_fcapi_string): New convenience macro. * font-mgr.c (Ffc_name_unparse): * font-mgr.c (Ffc_pattern_get): Use it. * font-mgr.c (Ffc_name_parse): * font-mgr.c (Ffc_font_match): Remove incorrect comments about memory leaks. 2007-04-01 Stephen J. Turnbull <stephen(a)xemacs.org&gt; Thanks to OG for review and Aidan Kehoe for useful comments. * redisplay-x.c (separate_textual_runs_mule): Reorganize and document. (separate_textual_runs): (separate_textual_runs_mule): (separate_textual_runs_xft_nomule): Update or add "theory of operation" header comments. (separate_textual_runs_xft_mule): Trailing whitespace removal. 2007-03-30 Olivier Galibert <galibert(a)pobox.com&gt; * redisplay-x.c (separate_textual_runs_xft_mule): Create. (separate_textual_runs_xft_nomule): Create. (separate_textual_runs_mule): Create. (separate_textual_runs_nomule): Create. (separate_textual_runs): Split the ifdef mess into 4 separate functions and simplify each. 2007-03-26 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * chartab.c (Fcopy_char_table): Use EQ, not ==. 2007-03-24 Olivier Galibert <galibert(a)pobox.com&gt; * chartab.c (Fcopy_char_table): Simplify the mirror table handling: never copy a mirror table, just make a new one and mark it dirty. Fixes http://article.gmane.org/gmane.emacs.xemacs.beta/17353 2007-01-22 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * eval.c (quote): (function): * fns.c (Frequire): Improve docstrings. 2007-02-21 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * objects-tty.c (tty_font_spec_matches_charset): Use Aidan's enum. * objects-xlike-inc.c (xft_find_charset_font): FC_WIDTH is obsolete and may be undefined. * sysdll.c (image_for_address): const cleanliness for Darwin. * fileio.c (Finsert_file_contents_internal): Remove unused gcpro5. 2007-02-17 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * glyphs-eimage.c (png_instantiate_unwind): Avoid recursion. (png_instantiate): Initialize setjmp_buffer early, and avoid recursive entry to error handler. 2007-02-19 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * specifier.c (Fdevice_matching_specifier_tag_list): Don't take XCDR of possible non-cons. Avoids crash in VM introduced or unmasked by Aidan's 2007-02-06 patch. (setup_device_initial_specifier_tags): Nuke unused variable. 2007-02-18 Stephen J. Turnbull <stephen(a)xemacs.org&gt; Code by Mike FABIAN <mfabian(a)suse.de&gt;, Takashi Iwai <tiwai(a)suse.de&gt;. See xemacs-beta <s3thctmf46c.fsf(a)magellan.suse.de&gt;. Thanks! Documentation marshalled by me. * select-x.c (x_get_window_property): The buffer for property data in 32-bit format is an array of longs, which need not be 32-bit. Compute residual from partial reads and buffer sizes correctly for sizeof(long) == 8. * select-common.h: Refer to documentation in select-x.c. 2007-02-16 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * config.h.in: Move CANNA_NEW_WCHAR_AWARE here from canna_api.c. Remove crufty CANNA2 define, we can't support CANNA1 (IROHA). 2007-02-06 Aidan Kehoe <kehoea(a)parhasard.net&gt; * specifier.c (setup_device_initial_specifier_tags): Fix a bug where the mswindows specifier tag was matching X11 devices, because the format of the DEVICE_USER_SPECIFIED_TAGS list wasn't being respected correctly. 2007-02-06 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule-charset.c (complex_vars_of_mule_charset): iso8859-2 is not the X11 charset registry for the iso8859-4 charset, my mistake. 2007-02-05 Mike Sperber <mike(a)xemacs.org&gt; * fileio.c (Finsert_file_contents_internal): Clean up dead `handler' variable. 2007-01-27 Mike Sperber <mike(a)xemacs.org&gt; * faces.c (complex_vars_of_faces): Move declarations of fontptr and fonts out of #ifdef MULE, unbreaking the non-MULE Xft build. 2007-01-20 Aidan Kehoe <kehoea(a)parhasard.net&gt; * eval.c: Document in more detail what happens with byte-compilation, `function', and `quote'. * fns.c: Mention that `require' is evaluated both at byte-compile time and at runtime. 2007-01-06 Vin Shelton <acs(a)xemacs.org&gt; * fileio.c (check_writable): Check old-style readonly bit first. 2006-12-21 Aidan Kehoe <kehoea(a)parhasard.net&gt; * file-coding.c: Update the make-coding-system docstring to reflect unicode-type * general-slots.h: New symbol, unicode-type, since 'type was being overridden when accessing a coding system's Unicode subtype. * intl-win32.c: Backslash a few parentheses, to help fontification along. * intl-win32.c (complex_vars_of_intl_win32): Use the 'unicode-type symbol, not 'type, when creating the Microsoft Unicode coding system. * unicode.c (unicode_putprop): * unicode.c (unicode_getprop): * unicode.c (unicode_print): Using 'type as the property name when working out what Unicode subtype a given coding system is was broken, since there's a general coding system property called 'type. Change the former to use 'unicode-type instead. 2006-12-27 Malcolm Purvis <malcolmp(a)xemacs.org&gt; * s/linux.h: Enable SYSTEM_MALLOC on ppc64, alpha and ia64 systems. 2006-12-17 Aidan Kehoe <kehoea(a)parhasard.net&gt; * faces.c (complex_vars_of_faces): Don't use server-side fallbacks when building with XFT support. 2006-12-06 Dr. Volker Zell <Dr.Volker.Zell(a)oracle.com&gt; * config.h.in: New HAVE_U_INT*_T defines. * database.c: Only use u_int*_t typedefs if not already defined. 2006-12-08 Aidan Kehoe <kehoea(a)parhasard.net&gt; * specifier.c (charset_matches_specifier_tag_set_p): A charset's entry in Vcharset_tag_lists may be nil, if, when that charset was created, no tags with associated charset predicates existed. Accept this possibility, treat it as the tag not matching that charset. 2006-12-11 Aidan Kehoe <kehoea(a)parhasard.net&gt; * event-tty.c: * event-tty.c (emacs_tty_event_pending_p): * event-tty.c (reinit_vars_of_event_tty): Pay attention to the dispatch event queue, and input pending signals in emacs_tty_event_pending_p. Makes pure TTY builds more responsive. 2006-12-11 Aidan Kehoe <kehoea(a)parhasard.net&gt; * glyphs-eimage.c (png_warning_func): * glyphs-eimage.c (png_instantiate): * glyphs-eimage.c (tiff_warning_func): Decode external binary data as such before passing it to warn_when_safe(). 2006-12-09 Vin Shelton <acs(a)xemacs.org&gt; * sound.c (init_native_sound): Fix typo in call to GTK_DEVICE macro. 2006-12-07 Vin Shelton <acs(a)xemacs.org&gt; * fileio.c: Added cast to qxeGetNamedSecurityInfofix call to fix VC6 build. 2006-11-24 Benson Margulies <benson(a)dchbk.us&gt; * fileio.c: Change check_writable to use the full Win32 mechanism to check access. * intl-auto-encap-win32.c: Add GetNamedSecurityInfo * intl-auto-encap-win32.h: Add GetNamedSecurityInfo * intl-auto-encap-win32.h Add qxeGetNamedSecurityInfo. * intl-encap-win32.c: Add aclapi.h : GetNamedSecurityInfo * syswindows.h: Add aclapi.h 2006-12-06 Aidan Kehoe <kehoea(a)parhasard.net&gt; * text.c (Fsplit_char): Make split-char available on non-Mule builds, taking out a superfluous call to get-charset to make that possible. 2006-11-30 Mike Sperber <mike(a)xemacs.org&gt; * fileio.c (Finsert_file_contents_internal): Don't call the file-name handler for `insert-file-contents' from here, which is too late. Instead, do it from Lisp.SSper 2006-11-29 Dr. Volker Zell <Dr.Volker.Zell(a)oracle.com&gt; * sysdir.h: Revert workaround missing d_ino field from 'struct dirent' for Cygwin as it's back in again. 2006-11-29 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule-charset.c (Fset_charset_registries): Add an optional argument FORCE, to allow specifying badly-formed CHARSET_REGISTRY-CHARSET_ENCODING combinations. Re-enables the sort of hacks described in 96wt67fa3f.fsf(a)mo.msk.ru from Ilya Golubev. 2006-11-29 Aidan Kehoe <kehoea(a)parhasard.net&gt; * sound.c (init_native_sound): Only X11 and GTK devices can possibly not be on the console of the associated machine. Fixes a crash when init_native_sound is called on a msprinter device. 2006-11-28 Aidan Kehoe <kehoea(a)parhasard.net&gt; * device-x.c: * device-x.c (coding_system_of_xrm_database): Cache the last db argument and resulting coding system, and return them--instead of calling Lisp--if the DB is the same pointer arument as last time. * faces.c (default_face_font_info): * window.c (window_displayed_height): Behave more gracefully if called when we have no information about the dimensions of the default face and window. 2006-11-28 Aidan Kehoe <kehoea(a)parhasard.net&gt; * doprnt.c (emacs_doprnt_1): Ibyte -> Ascbyte, for the sake of the MSVC build. Thank you Vin. 2006-11-28 Aidan Kehoe <kehoea(a)parhasard.net&gt; * doprnt.c (emacs_doprnt_1): Integrate Sebastian Freundt's SXEmacs bug fix for cases where a format specifier overflows the allocated buffer with a float format string. 2006-11-23 Robert Pluim <rpluim(a)gmail.com&gt; * sysdep.c (strlwr): Don't intermix declarations and code. (wcslen): ditto 2006-11-23 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule-coding.c (iso2022_decode): Only take the lower seven bits of any eight-bit character that would be illegal in UTF-8, when handling ISO/IR 196 escapes. 2006-11-14 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * buffer.c (buffer-file-name): Document invariant. (buffer-file-truename): Sync wording to buffer-file-name. Thanks to Nelson Ferreira <nelson.ferreira(a)ieee.org&gt; for report and discussion. 2006-11-20 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule-ccl.c (CCL_MAKE_CHAR): * mule-ccl.c (ccl_driver): Eliminate a CCL bug with control-1 chars and write-multibyte-character--thank you for the report, Ilya--and eliminate a crash when a non-existent charset ID is specified for a mule-to-unicode call. 2006-11-18 Mike Sperber <mike(a)xemacs.org&gt; * redisplay-x.c (separate_textual_runs): Don't try to call ichar_to_unicode on non-MULE. 2006-11-18 Mike Sperber <mike(a)xemacs.org&gt; * window.c (window_pixel_height_to_char_height): Cater to the possibility that defheight may be 0. 2006-11-15 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule-charset.c: * mule-charset.c (Fcharsets_in_region): Added a charsets-in-region implementation in C. 2006-11-14 Aidan Kehoe <kehoea(a)parhasard.net&gt; * objects-xlike-inc.c (x_find_charset_font): Move the warn_when_safe call to where GCing is irrelevant; as things are it doesn't GC, but I'm more comfortable without that being relevant. * specifier.c (define_specifier_tag): Remove a couple of lines added for the sake of debugging; add a GCPRO1 for the allocated vector, since the call_trapping_problems can GC. 2006-11-12 Aidan Kehoe <kehoea(a)parhasard.net&gt; * charset.h: * mule-charset.c (set_charset_registries): Provide a C-accessible version of set-charset-registries that doesn't error. Called from x_find_charset_font. * faces.c (ensure_face_cachel_contains_charset): Correct my spelling. * faces.c (update_EmacsFrame): Don't update the frame if it isn't live. * faces.h: Add some documentation on FACE_FONT. * frame-gtk.c (gtk_update_frame_external_traits): * frame-x.c (x_update_frame_external_traits): In the event that FACE_FONT has deleted the frame, don't manipulate it further in update_frame_external_traits. * mule-charset.c: * mule-charset.c (Fset_charset_registries): Don't allow XLFD wildcards in charset registries. Call the factored-out C-callable version instead of implementing the guts of the function here. * objects-gtk.c: #include "charset.h" * objects-xlike-inc.c (x_find_charset_font, gtk_find_charset_font): In the event that the charset is ASCII and we haven't matched anything up to now, even with a pattern of "*", add "iso8859-1" to the charset's registry and try again. * window.c (window_pixel_width_to_char_width): The default width of a face may be zero; only divide by it if it's nonzero. 2006-11-11 Aidan Kehoe <kehoea(a)parhasard.net&gt; * specifier.c: Update the specifier-matching-instance documentation to reflect the new format of font-specifier MATCHSPECs. 2006-11-11 Aidan Kehoe <kehoea(a)parhasard.net&gt; * specifier.c: VM calls device-matching-specifier-tag-list; my taking it out of non-debug builds was misjudged, this change puts it back in. I've also reverted a couple of incidental and wrong whitespace changes. 2006-11-09 Aidan Kehoe <kehoea(a)parhasard.net&gt; * specifier.c (specifier_instance_from_inst_list): Accept symbols as well as charset objects as the cons of a font-related specifier matchspec. Thank you Ilya. 2006-11-07 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule-charset.c: Take the Qfinal declaration out of mule-charset.c; fixes the MSVC build. Sorry Vin! 2006-11-06 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule-charset.c (Fmake_charset): * objects-msw.c (mswindows_font_spec_matches_charset): * specifier.c (syms_of_specifier): Three changes to prevent the build dying with MSVC and with DEBUG_XEMACS turned off--thank you Vin. 2006-11-05 Aidan Kehoe <kehoea(a)parhasard.net&gt; * charset.h: Prefer the charset-registries property to the charset-registry property; accept the latter for compatibility, warning when its regexp functionality is used. * charset.h (XCHARSET_CCL_PROGRAM): * charset.h (XCHARSET_NAME): Make dummy versions of these available in non-Mule. * console-impl.h: * console-impl.h (struct console_methods): Rename the last parameter to a couple of methods; reformat their declarations. * faces.c: * faces.c (face_property_matching_instance): * faces.c (ensure_face_cachel_contains_charset): * faces.c (merge_face_cachel_data): * faces.c (reset_face_cachel): * faces.c (mark_face_cachels_as_not_updated): * faces.c (syms_of_faces): * faces.c (vars_of_faces): * faces.c (complex_vars_of_faces): Provide a DEBUG_FACES macro; use it to make debugging output available in debug builds. Implement multi-stage font lookup, assigning the stages names, not numbers. Re-implement the cachel->font_specified cache using the infrastructure for Lisp bit vectors. * faces.h: * faces.h (struct face_cachel): * faces.h (FACE_CACHEL_FONT_UPDATED): * faces.h (FACE_FONT): Re-implement the cachel->font_specified cache using the infrastructure for Lisp bit vectors. * font-mgr.h: Move some XFT debug macros here from objects-x.c. * general-slots.h: Provide a few new symbols for the multi-stage font resolution process. * intl.c (init_intl): Correct a comment. * lisp.h: Provide a macro to declare an inline lisp bit vector where the size is fixed. Make Qregistries available all over, not Qregistry. * mule-charset.c: * mule-charset.c (mark_charset): * mule-charset.c (print_charset): * mule-charset.c (make_charset): * mule-charset.c (Fmake_charset): * mule-charset.c (Fcharset_property): * mule-charset.c (Fset_charset_ccl_program): * mule-charset.c (syms_of_mule_charset): * mule-charset.c (complex_vars_of_mule_charset): * mule-charset.c (CHINESE_CNS_PLANE): Prefer the charset-registries property to the charset-registry property; accept the latter for compatibility, warning when its regexp functionality is used. * objects-gtk.c: * objects-gtk.c (gtk_font_spec_matches_charset): * objects-gtk.c (gtk_find_charset_font): * objects-msw.c (mswindows_find_charset_font): * objects-tty.c (tty_find_charset_font): Redeclare various functions to work with the multi-stage lookup process. Include objects-xlike-inc. * objects-x.c: Provide a DEBUG_OBJECTS macro; use it to make debugging output available in debug builds. * objects-x.c (x_initialize_font_instance): * objects-x.c (x_print_font_instance): * objects-x.c (xlistfonts_checking_charset): * objects-x.c (vars_of_objects_x): Don't regex match on the output of XListFonts; instead, use the fixed strings of the charset-registries to comparatively limit the IPC that will happen. Include objects-xlike-inc.c * objects-xlike-inc.c: * objects-xlike-inc.c (count_hyphens): New. How many ASCII minus characters in a string? * objects-xlike-inc.c (xlistfonts_checking_charset): * objects-xlike-inc.c (mule_to_fc_charset): * objects-xlike-inc.c (xft_find_charset_font): * objects-x.c (x_find_charset_font): Move some methods here to share them with GTK. * objects.c (print_font_instance): * objects.c (font_spec_matches_charset): * objects.c (font_validate_matchspec): * objects.c (font_instantiate): Redeclare some methods to take enums rather than numeric stages. * objects.h (EXFUN): Make Fregexp_quote available to mule-charset.c * redisplay-x.c: * redisplay-x.c (separate_textual_runs): Make this slightly faster, cleaner. Make it accept a face cachel pointer argument, and check it as to whether a given charset should be translated to UCS-2 before redisplay. * specifier.c: * specifier.c (charset_matches_specifier_tag_set_p): * specifier.c (define_specifier_tag): * specifier.c (Fdefine_specifier_tag): * specifier.c (setup_device_initial_specifier_tags): * specifier.c (setup_charset_initial_specifier_tags): * specifier.c (specifier_instance_from_inst_list): * specifier.c (syms_of_specifier): * specifier.c (vars_of_specifier): * specifier.h: Extend specifiers to allow limiting their applicability by using charset predicates. Document this. Run indent-region on the file, at Stephen's suggestion. * unicode.c (unicode_to_ichar): * unicode.c (syms_of_unicode): * unicode.c (vars_of_unicode): Use unicode-registries, a dumped vector, as the charset-registries of the on-the-fly JIT charsets. 2006-11-01 Adrian Aichner <adrian(a)xemacs.org&gt; * sysdep.c (wcslen): Check for NULL pointer. * sysdep.c (strlwr): Ditto. * nt.c (mswindows_getdcwd): Ditto (actual cause of reported crash). * intl-win32.c (wcscmp): Ditto. * intl-win32.c (wcslen): Ditto. * intl-win32.c (wcsncpy): Ditto. * intl-win32.c (wcscpy): Ditto. * intl-win32.c (wcsdup): Ditto. * fileio.c (Ffile_name_directory): Return Qnil when mswindows_getdcwd returns NULL working directory. 2006-10-30 Malcolm Purvis <malcolmp(a)xemacs.org&gt; * device-x.c (x_init_device): Look in more directories when searching for the locale specific app-defaults files. 2006-10-28 Aidan Kehoe <kehoea(a)parhasard.net&gt; * event-xlike-inc.c: Rework the X11 keysym support to allow the use of Cyrillic keysyms that are not in ISO-8859-5. 2006-10-27 Aidan Kehoe <kehoea(a)parhasard.net&gt; * fileio.c (Finsert_file_contents_internal) : Clarify that we follow a more correct but more expensive design for buffers where the on-disk representation doesn't correspond directly to the XEmacs internal representation. 2006-08-29 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * eval.c (Fcatch): Correct syntax of `throw' in docstring. * cmds.c (Fforward_line): Document that return can be negative. 2006-08-11 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * frame.c (mouse-motion-handler): Document relation to hooks. 2006-08-24 Aidan Kehoe <kehoea(a)parhasard.net&gt; * console-tty-impl.h (struct tty_console): New flag; multiple_width, specifying whether East Asian characters take up two columns on this terminal. The macro to access it expands to a constant under non-Mule, so anything conditionalising on it will be optimised away. * console-tty.c: * console-tty.c (tty_init_console): Initialise tty_con->multiple_width; zero on non-Mule, one on Mule. * console-tty.c (Fconsole_tty_multiple_width): New. * console-tty.c (Fset_console_tty_multiple_width): New. * console-tty.c (syms_of_console_tty): Make them available outside console-tty.c. * redisplay-tty.c (tty_text_width): * redisplay-tty.c (tty_output_ibyte_string): Check that the relevant console has multiple-width characters before using something other than the number of characters in a string for the text width of that string. * text.c: * text.c (ibyte_string_displayed_columns): * text.c (ichar_string_displayed_columns): Remove some Mule conditionals; add some sanity-checking. 2006-08-23 Aidan Kehoe <kehoea(a)parhasard.net&gt; * syntax.c (complex_vars_of_syntax): Guillemets are now punctuation, not parentheses. AucTeX does this, and it's infinitely less annoying if you're dealing with German text or mixed German and French. 2006-08-03 Jerry James <james(a)xemacs.org&gt; * keymap.c (where_is_recursive_mapper): Use the freshly allocated storage instead of leaking it. 2006-08-06 Aidan Kehoe <kehoea(a)parhasard.net&gt; * doc.c (Fsnarf_documentation): Add an assertion and a comment, rather than crashing in the 76584th call of hash_string when someone's using a DOC file that doesn't conform to format. 2006-08-05 Aidan Kehoe <kehoea(a)parhasard.net&gt; * syntax.c (complex_vars_of_syntax): String literals are char[], not unsigned char[]. Cast them when passing to a function that takes UExtbyte *. 2006-08-05 Aidan Kehoe <kehoea(a)parhasard.net&gt; * lread.c (read_unicode_escape): Refactor this code out from read_escape, since it's now called from read_string as well. * lread.c (read_escape): Call read_unicode_escape instead of using inline code, * lread.c (read_string): Refactor out from read1, provide raw and honor_unicode options. * lread.c (read_raw_string): Added, a function that calls read_string with the correct arguments for a raw string. * lread.c (read1): Pass raw strings to read_raw_string; pass strings to read_string. 2006-08-04 Aidan Kehoe <kehoea(a)parhasard.net&gt; * event-tty.c (emacs_tty_next_event): Check dispatch_event_queue for pending events, since we add to that in drain_tty_devices(). Fixes dropped key sequences on TTY builds. 2006-08-04 Aidan Kehoe <kehoea(a)parhasard.net&gt; * syntax.c (Fsyntax_table_p): Make the docstring better reflect the code. * syntax.c (define_standard_syntax): Take a const UExtbyte * as the first argument, not a const char * * syntax.c (complex_vars_of_syntax): Use a macro instead of repeating code; don't redundantly set the syntax of the alphanumeric characters to ?w; define syntax for Latin 1 characters here instead of in Lisp; guillemets are string delimiters, not parentheses. 2006-07-18 Marcus Crestani <crestani(a)xemacs.org&gt; * .cvsignore: Add .dbxrc.in and .gdbinit.in. 2006-07-18 Marcus Crestani <crestani(a)xemacs.org&gt; * gc.c (kkcc_backtrace): Print adresses as pointers. * vdb-posix.c (vdb_fault_handler): Print adresses as pointers. 2006-07-18 Marcus Crestani <crestani(a)xemacs.org&gt; * alloc.c: Fix typo in string_direct_data_description. 2006-07-07 Jerry James <james(a)xemacs.org&gt; * sound.c (init_native_sound): Make the target of a strcpy be exactly the right size to hold the copied string. 2006-07-08 Aidan Kehoe <kehoea(a)parhasard.net&gt; * lisp.h: * redisplay.c (add_octal_runes): * syntax.h: * text.c Change some comments to reflect a 21-bit character space. * text.c (non_ascii_valid_ichar_p): Check that no character value is greater than 2^^21, not 2^^19. This fixes the Mule build when error-checking is turned on. 2006-07-08 Aidan Kehoe <kehoea(a)parhasard.net&gt; * symbols.c (Fsubr_name): Use the CHECK_SUBR macro instead of the GNU code's explicit if statement. 2006-07-08 Aidan Kehoe <kehoea(a)parhasard.net&gt; * charset.h: Move to 7 bits instead of 5 for the first field of a character. 2006-07-07 Aidan Kehoe <kehoea(a)parhasard.net&gt; * config.h.in: Don't declare inline macros as `extern' if we're building a standalone program, because their extern definitions in inline.c are unlikely to be included. 2006-06-24 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * gc.c (gc_mark_root_set): Quiet GCC 4 whining about unused values. 2006-06-29 Jerry James <james(a)xemacs.org&gt; * scrollbar-gtk.c (gtk_free_scrollbar_instance): Compare instance->scrollbar_data against NULL before using it. * scrollbar-msw.c (mswindows_free_scrollbar_instance): Compare sb->scrollbar_data against NULL before using it. 2006-06-22 Jerry James <james(a)xemacs.org&gt; * redisplay-gtk.c (gtk_output_display_block): Fix a Dynarr leak. Don't create the buffer if there is nothing to do. * redisplay-msw.c (mswindows_output_display_block): Ditto. * redisplay-output.c (redisplay_output_layout): Ditto. * redisplay-tty.c (tty_output_display_block): Ditto. * redisplay-x.c (x_output_display_block): Ditto. 2006-06-23 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * font-mgr.c (extract_fcapi_string): (fc_intern): (Ffc_name_parse): (Ffc_name_unparse): (Ffc_pattern_add): (Ffc_pattern_del): (Ffc_pattern_get): (string_list_to_fcobjectset): fc_intern and extract_fcapi_string should return [const] Extbyte *. Make it so, update callers. 2006-05-26 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * objects-x.c (x_initialize_font_instance): (x_font_instance_truename): (charset_table): (x_find_charset_font): Pander to GCC4 signed character paranoia. 2006-06-19 Jerry James <james(a)xemacs.org&gt; * device-x.c (x_IO_error_handler): Do not dereference d if it is NULL. * dialog-x.c (dbox_selection_callback): Ensure f is non-NULL, then dereference it, not the other way around. * emacs.c (main_1): restart is always 0 here. * extents.c (detach_all_extents): Call extent_list_delete_all with a non-NULL parameter only. * glyphs-widget.c (widget_query_geometry): Guard against possibly NULL width and height. * window.c (change_window_height): Restore use of CURCHARSIZE removed by 2006-06-16 change, to preserve the abstraction. 2006-06-19 Jerry James <james(a)xemacs.org&gt; * frame-x.c (x_set_frame_properties): Remove casts to silence GCC warnings about a missing sentinel. 2006-06-16 Jerry James <james(a)xemacs.org&gt; * dgif_lib.c (DGifCloseFile): Do not dereference GifFile before checking if it is NULL. Also fix a memory leak. * fileio.c (Finsert_file_contents_internal): Remove dead code. * input-method-xlib.c (XIM_SetGeometry): Do not dereference f or xic before checking if they are NULL. * md5.c (Fmd5): Check whether Lstream_read encountered an error. * nas.c (Err): Fix a memory leak. * scrollbar-x.c (x_free_scrollbar_instance): Do not dereference instance->scrollbar_data before checking if it is NULL. * text.c (eicmp_1): Move assertions to before the point where they must be true for correctness. * vdb-posix.c (vdb_fault_handler): Guard against a return from ABORT(). * window.c (change_window_height): Skip always true comparison in the expansion of CURCHARSIZE. 2006-06-16 Jerry James <james(a)xemacs.org&gt; * alloc.c: Don't add MODULE_DEFINABLE_TYPE_COUNT to countof (lrecord_implementations_table); the latter is already big enough. 2006-06-18 Aidan Kehoe <kehoea(a)parhasard.net&gt; * event-xlike-inc.c: Add support for three publishing keysyms with clear Unicode mappings. They would have been included in my commit of 2005-06-26 had Markus Kuhn known about them. 2006-06-13 Jerry James <james(a)xemacs.org&gt; * lisp-union.h: Cast away bit restrictions on values retrieved from the Lisp_Object union. 2006-06-14 Aidan Kehoe <kehoea(a)parhasard.net&gt; * mule-ccl.c: Don't declare ccl-program as a symbol here, leave that to general-slots.h. * unicode.c (unicode_to_ichar): Use qxesprintf, not snprintf, change some types to allow complilation on Win32. 2006-06-03 Aidan Kehoe <kehoea(a)parhasard.net&gt; * charset.h: * charset.h (struct Lisp_Charset): * charset.h (CHARSET_ENCODE_AS_UTF_8): * charset.h (XCHARSET_ENCODE_AS_UTF_8): Add a flag `encode-as-utf-8' to the Mule charset structure; if set, it's an indication to ISO 2022-oriented coding systems that the characters of that charset should be encoded using the ISO-IR 196 UTF-8 escape syntax, since they're not members of any other well-known character set we're aware of. Make enum unicode_type, encode_unicode_char and Funicode_to_char available outside of unicode.c * lread.c: * event-xlike-inc.c: Use the charset.h declaration of Funicode_to_char, don't declare it ourselves. My XFree86 installation has taken to passing me ASCII characters using the Unicode keysyms; accept them too. * general-slots.h: Make `ccl-program' and `encode-as-utf-8' available as symbols generally. * mule-ccl.c: Add CCL_MuleToUnicode, CCL_UnicodeToMule, implement them, enable and debug CCL_MAKE_CHAR, have CCL_WriteMultibyteChar2 segfault less, fix some grammar. * mule-charset.c (make_charset): * mule-charset.c (Fmake_charset): * mule-charset.c (Fcharset_property): * mule-charset.c (complex_vars_of_mule_charset): Require the encode_as_utf_8 property when calling make_charset (); accept it when creating a charset from Lisp in Fmake_charset. * mule-coding.c: * mule-coding.c (dynarr_add_2022_one_dimension): * mule-coding.c (dynarr_add_2022_two_dimensions): Add two convenience functions for iso2022_decode, to abstract out writing UTF-8 a little. * mule-coding.c (enum iso_esc_flag): Add one more state to reflect the existence of the UTF-8 escape. * mule-coding.c (struct iso2022_coding_stream): Add a counter variable to the state to permit handling variable-length UTF-8. * mule-coding.c (parse_iso2022_esc): Update the function to work with ISO_STATE_UTF_8; only the ESC % ＠ escape is processed in that state, everything else is ignored and passed through by the error handler. * mule-coding.c (iso2022_decode): * mule-coding.c (iso2022_designate): * mule-coding.c (iso2022_encode): Handle the UTF-8 escape sequences in reading and in writing ISO 2022. * redisplay-x.c (separate_textual_runs): Add a comment to the effect that the dimension stuff breaks when using CCL programs and registries to map to a bigger charset. * unicode.c: Add support for creating new characters on the fly as unknown Unicode code points are encountered. * unicode.c (get_free_codepoint): New. * unicode.c (unicode_to_ichar): Reworked to create new code points on the fly. * unicode.c (Funicode_to_char): Update the docstring. * unicode.c (struct unicode_coding_system): Move enum unicode_type into charset.h. * unicode.c (encode_unicode_char): encode_unicode_char isn't static any longer, mule-coding.c uses it. * unicode.c (syms_of_unicode): Make a couple of symbols available to unicode.c * unicode.c (vars_of_unicode): Tell the garbage collector about current_jit_charset, initialise it. 2006-05-25 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * objects-x.c (x_find_charset_font): Keep local names local; wrap the spurious declaration in a block, don't move it. 2006-05-25 Aidan Kehoe <kehoea(a)parhasard.net&gt; * file-coding.c (output_bytes_in_ascii_and_hex): I installed GCC 4.0 (which may not even have been necesssary, -pedantic might have done it) and it showed me that the problem my last commit intended to address was in eistr_ext rather than stderr_out. 2006-05-16 Andrey Slusar <anrays(a)gmail.com&gt; * objects-x.c (x_find_charset_font): Fix build with gcc 2.95. 2006-05-23 Aidan Kehoe <kehoea(a)parhasard.net&gt; * file-coding.c (output_bytes_in_ascii_and_hex): Fix a Win32 build failure introduced by my last commit. 2006-05-23 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * .gdbinit.in: Moved to etc. * .dbxrc.in: Moved to etc. 2006-05-20 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * tests.c (Ftest_data_format_conversion): Fix no-mule autodetect bugs. Update FSF copyright. 2006-05-23 Aidan Kehoe <kehoea(a)parhasard.net&gt; * frame-x.c (x_set_frame_properties): Check that the frame is live and that the device is not being deleted before attempting anything. Avoids calling X code if x_IO_error_handler has seen a broken pipe. 2006-05-22 Aidan Kehoe <kehoea(a)parhasard.net&gt; * file-coding.c (output_bytes_in_ascii_and_hex): Pass stderr_out internally-formatted data, don't use an external format. 2006-05-21 Aidan Kehoe <kehoea(a)parhasard.net&gt; * doc.c (extract_object_file_name): * doc.c (unparesseuxify_doc_string): Leave sufficient space for the '\0' sentinel when reading into the buffer. The bug in unparesseuxify_doc_string had been there for ten years at least, but it was Fabrice Popineau's investigation of the code on the same model in extract_object_file_name that provoked its discovery. Thank you Fabrice! -------- ChangeLog Entries from tests/ChangeLog -------- 2007-05-21 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * XEmacs 21.5.28 "fuki" is released. 2007-05-20 Aidan Kehoe <kehoea(a)parhasard.net&gt; * automated/mule-tests.el: Make the file name coding system tests work on OS X. Check various slots of the language environment structure to make sure they're well formed--we've been shipping without a Turkish or Latin-10 input method for years, for example, which is an error. 2007-04-30 Aidan Kehoe <kehoea(a)parhasard.net&gt; * automated/mule-tests.el (featurep): Minimal tests of the non-BMP UTF-16 support. 2007-05-12 Aidan Kehoe <kehoea(a)parhasard.net&gt; * automated/mule-tests.el: * automated/mule-tests.el (bytecomp): Require it, since we're testing its Unicode support. * automated/mule-tests.el (featurep): Assert that the escape-quoted coding cookie is added when needed, and ignored when not. 2007-04-29 Aidan Kehoe <kehoea(a)parhasard.net&gt; * automated/mule-tests.el (featurep): Sort the results of charsets-in-region, charsets-in-string before comparing them to the previously-determined list of character sets. Eliminates a dependency on the algorithm charsets-in-{region,string} uses. 2007-03-26 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * automated/syntax-tests.el: Test for regression of bug fixed by Olivier Galibert <20070324221053.GA48218(a)dspnet.fr.eu.org&gt;. Test by Ralf Angeli http://article.gmane.org/gmane.emacs.xemacs.beta/17353. 2006-11-25 Aidan Kehoe <kehoea(a)parhasard.net&gt; * automated/iso-ir-196-test.el: * automated/mule-tests.el (featurep): Move the tested added in the previous commit to a separate file. 2006-11-23 Aidan Kehoe <kehoea(a)parhasard.net&gt; * automated/mule-tests.el (featurep): Add a test that ISO/IR 196 escape handling in ISO-2022-based charsets don't choke on invalid bytes in UTF-8 text. 2006-11-20 Aidan Kehoe <kehoea(a)parhasard.net&gt; * automated/mule-tests.el (featurep): Add tests to make sure the fixes to the CCL bugs I just checked in don't regress. 2006-08-05 Aidan Kehoe <kehoea(a)parhasard.net&gt; * automated/lisp-reader-tests.el: New file, imported from Martin Kuehl's SXEmacs commit; test the new raw string syntax, including the Unicode escapes, which SXEmacs doesn't have. 2006-06-24 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * automated/test-harness.el (Silence-Message): New macro. * automated/mule-tests.el: Use it. * automated/region-tests.el: Use it. * automated/tag-tests.el: Use it. 2006-06-24 Stephen J. Turnbull <stephen(a)xemacs.org&gt; * automated/test-harness.el: Improve handling of unexpected errors. 2006-06-03 Aidan Kehoe <kehoea(a)parhasard.net&gt; * automated/mule-tests.el: Add checks that several Unicode characters, expressed as UTF-8-encoded strings, are handled correctly by the UTF-8 support of the escape-quoted character set. * automated/mule-tests.el (unicode-code-point-to-utf-8-string): New. Convert a Unicode code point to the equivalent UTF-8 string. This is a naive implementation in Lisp. -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Ask not how you can "do" free software business; ask what your business can "do for" free software. Date: Mon, 21 May 2007 15:54:41 +0900 Message-ID: <87y7ji7jke.fsf(a)uwakimon.sk.tsukuba.ac.jp&gt; _______________________________________________ XEmacs-Beta mailing list XEmacs-Beta(a)xemacs.org http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta