[xft-sjt] Workaround crash in Label widget, build fixes

Informative patch to xft-sjt branch Fixes crash in Label widget and problems with building without Xft. Index: ChangeLog =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/ChangeLog,v retrieving revision 1.418 diff -u -r1.418 ChangeLog --- ChangeLog 14 Nov 2004 12:31:09 -0000 1.418 +++ ChangeLog 10 Dec 2004 03:34:47 -0000 ＠＠ -0,0 +1,11 ＠＠ +2004-12-10 Stephen J. Turnbull <stephen(a)xemacs.org&gt; + + * configure.usage: Various minor improvements. + +2004-12-09 Stephen J. Turnbull <stephen(a)xemacs.org&gt; + + * configure.in: Update FSF's copyright. Support --with-xft=LIST + syntax. Try to find a directory containing freetype/ftheader.h. + + * configure.usage (--with-xft): Update. + Index: lwlib/ChangeLog =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/lwlib/ChangeLog,v retrieving revision 1.64.2.1 diff -u -r1.64.2.1 ChangeLog --- lwlib/ChangeLog 3 Dec 2004 07:40:58 -0000 1.64.2.1 +++ lwlib/ChangeLog 10 Dec 2004 03:34:50 -0000 ＠＠ -0,0 +1,21 ＠＠ +2004-12-10 Stephen J. Turnbull <stephen(a)xemacs.org&gt; + + * lwlib-Xaw.c (xaw_create_button): + (xaw_create_progress): + (xaw_create_text_field): + Changed XtCreateManagedWidget to XtCreateWidget when child is + managed later in the same function. + + (debug_gauge): New variable. + (lw_debug_print_xt_arglist): + (lw_debug_print_class_resources): + New debug functions. + +2004-12-09 Stephen J. Turnbull <stephen(a)xemacs.org&gt; + + * xlwmenu.c (xlwMenuResources): Use "redundant" XftFont resource. + + * xlwtabs.c: + * xlwtabsP.h: + Use USE_XFT_TABS consistently. + Index: man/ChangeLog =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/man/ChangeLog,v retrieving revision 1.269 diff -u -r1.269 ChangeLog --- man/ChangeLog 16 Nov 2004 07:37:29 -0000 1.269 +++ man/ChangeLog 10 Dec 2004 03:34:53 -0000 ＠＠ -0,0 +1,19 ＠＠ +2004-11-26 Stephen J. Turnbull <stephen(a)xemacs.org&gt; + + * internals/internals.texi (Better Rendering Support -- + Configuration with the Interim Patches): + Add documentation of X resources, especially for GUI elements. + (Better Rendering Support -- Implementation): + More comments about specification and design issues. + (Better Rendering Support -- Current Status): + New subsubsection "Bugs Reported in sjt-xft". + +2004-11-26 Stephen J. Turnbull <stephen(a)xemacs.org&gt; + + * internals/internals.texi (Future Work -- Better Rendering Support): + New node. + (Top): + (Future Work): + (Future Work -- Lisp Engine Replacement): + Add pointers to new node. + Index: src/ChangeLog =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/ChangeLog,v retrieving revision 1.758.2.1 diff -u -r1.758.2.1 ChangeLog --- src/ChangeLog 27 Nov 2004 05:56:00 -0000 1.758.2.1 +++ src/ChangeLog 10 Dec 2004 03:35:20 -0000 ＠＠ -0,0 +1,26 ＠＠ +2004-12-10 Stephen J. Turnbull <stephen(a)xemacs.org&gt; + + * redisplay-x.c (x_text_width_single_run): Whitespace improvement. + +2004-12-09 Stephen J. Turnbull <stephen(a)xemacs.org&gt; + + * glyphs-widget.c (image_instantiator_progress_gauge): + (image_instantiator_format_create_glyphs_widget): + Fix typo guage->gauge. + + * emacs.c (main_1): Casting out devils (size_t). + + * compiler.h (USED_IF_XFT): New variant. + * objects-x.c (x_font_spec_matches_charset): + * redisplay-x.c (x_text_width_single_run): + Use it. + + * config.h.in: Support USE_XFT_TABS and USE_XFT_GAUGE. + +2004-11-28 Stephen J. Turnbull <stephen(a)xemacs.org&gt; + + * objects-x.c (x_initialize_font_instance): Make sure + FONT_INSTANCE_X_FONT is cleared for Xft fonts. Add comments. + Fiddle whitespace. + (x_print_font_instance): Clarify printed font instance. + Index: configure.in =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/configure.in,v retrieving revision 1.233.2.1 diff -u -r1.233.2.1 configure.in --- configure.in 25 Nov 2004 12:43:52 -0000 1.233.2.1 +++ configure.in 10 Dec 2004 03:34:49 -0000 ＠＠ -3,7 +3,7 ＠＠ [#### Configuration script for XEmacs. Largely divergent from FSF. #### Guess values for system-dependent variables and create Makefiles. #### Generated automatically using autoconf version] AC_ACVERSION [ -#### Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000 Free Software Foundation, Inc. +#### Copyright (C) 1992-2004 Free Software Foundation, Inc. #### Copyright (C) 1993-1995 Board of Trustees, University of Illinois. #### Copyright (C) 1996, 1997 Sun Microsystems, Inc. #### Copyright (C) 1995, 1996, 2002, 2003, 2004 Ben Wing. ＠＠ -373,7 +373,10 ＠＠ need_modules_common='' with_menubars='' dnl These conflict, try to detect incompatibility early -with_xft_menubars='no' +with_xft_emacs='yes' +with_xft_menubars='yes' +with_xft_tabs='yes' +with_xft_gauge='yes' with_xfs='no' with_scrollbars='' with_widgets='' ＠＠ -533,7 +536,6 ＠＠ with_hesiod | \ with_dnet | \ with_infodock | \ - with_xft | \ with_netinstall | \ with_ipv6_cname | \ external_widget | \ ＠＠ -838,9 +840,6 ＠＠ l | lu | luc | luci | lucid ) val=lucid ;; mo | mot | moti | motif ) val=motif ;; a | at | ath | athe | athen | athena ) val=athena ;; - dnl not in xft reloaded #3 - dnl #### for menubars only; rethink this interface - x | xf | xft ) val=xft ;; n | no | non | none ) val=no ;; y | ye | yes ) val=yes ;; dnl Explicit --with-widgets on command line means yes. ＠＠ -853,6 +852,51 ＠＠ eval "$opt=\"$val\"" ;; + dnl Has the user requested Xft support? + dnl Borrowed from the sound code + dnl not in xft reloaded #3 + "with_xft" ) + dnl values is a subset of all,emacs,menubars,tabs,gauge + dnl or their negatives: none,noemacs,nomenubars,notabs,nogauge + for x in `echo "$val" | sed -e 's/,/ /g'` ; do + case "$x" in + dnl all and none are only permitted as the first in the list. + n | no | non | none ) new_fdefault=no ;; + a | al | all | both ) new_fdefault=yes ;; + + emacs ) with_xft_emacs=yes ;; + # noemacs makes no sense + + menubars ) with_xft_menubars=yes with_xft_emacs=yes ;; + nomenubars ) with_xft_menubars=no ;; + + tabs ) with_xft_tabs=yes with_xft_emacs=yes ;; + notabs ) with_xft_tabs=no ;; + + gauge ) with_xft_gauge=yes with_xft_emacs=yes ;; + nogauge ) with_xft_gauge=no ;; + + * ) bogus_xft=yes ;; + esac + if test "$bogus_xft" -o \ + \( -n "$new_fdefault" -a -n "$xft_notfirst" \) ; then + types="\`all', \`none', \`(no)menubars', \`no(tabs)', \`(no)gauge', \`emacs'." + USAGE_ERROR(["Valid types for the \`--$optname' option are: + $types. +Option \`all' or \`none' must be first in the list. +All of \`menubars', \`tabs', and \`gauge' force \`emacs' (buffer text) on. +Use \`none,emacs' to have only the buffer text supporting Xft."]) + elif test -n "$new_fdefault" ; then + with_xft_emacs=$new_fdefault + with_xft_menubars=$new_fdefault + with_xft_tabs=$new_fdefault + with_xft_gauge=$new_fdefault + new_fdefault= # reset this + fi + xft_notfirst=true + done + ;; + dnl Obsolete legacy argument? Warn, but otherwise ignore. "use_minimal_tagbits" | \ "use_indexed_lrecord_implementation" | \ ＠＠ -881,7 +925,7 ＠＠ dnl Check for incompatible settings dnl ------------------------------- -if test "$with_xfs" = "yes" -a "$with_menubars" = "xft"; then +if test "$with_xfs" = "yes" -a "$with_xft_menubars" = "yes"; then USAGE_ERROR("XFS and Xft in the menubars are incompatible!") fi ＠＠ -3249,7 +3293,22 ＠＠ dnl include xft/AA support? dnl #### need to check for includes here (especially, freetype.h for v.2) - if test "$with_xft" = "yes"; then + if test "$with_xft_emacs" = "yes"; then + AC_CHECKING([for Xrender, fontconfig, and Xft]) + xft_includes_found=no + AC_CHECK_HEADER(freetype/config/ftheader.h,xft_includes_found=yes) + + dnl do this for /usr/include, /sw/include, and /opt/local/include, too? + freetype_include_top="/usr/X11R6/include/freetype2" + if test "$xft_includes_found" = "no" -a "$with_xft_emacs" != "no" -a \ + -d $freetype_include_top; then + save_c_switch_site="$c_switch_site" + c_switch_site="$c_switch_site -I$freetype_include_top" + AC_CHECK_HEADER(freetype/config/ftheader.h,xft_includes_found=yes) + if test "$xft_includes_found" != "yes"; then + XE_DIE(["Unable to find headers for --with-xft"]) + fi + fi AC_CHECK_LIB(Xrender, XRenderQueryExtension, XE_PREPEND(-lXrender, libs_x), [XE_DIE(["Unable to find libXrender for --with-xft"])]) AC_CHECK_LIB(fontconfig, FcPatternCreate, XE_PREPEND(-lfontconfig, libs_x), ＠＠ -4022,13 +4081,13 ＠＠ with_menubars="lucid" ;; esac -dnl "$with_menubars" = "xft" dnl this is not in xft reloaded #3 -if test "$with_menubars" = "xft" ; then - if test "$with_xft" = "yes" ; then +if test "$with_xft_menubars" != "no" ; then + if test "$with_xft_emacs" = "yes" -a "$with_menubars" != "no" ; then with_xft_menubars="yes" + else + with_xft_menubars="no" fi - with_menubars="lucid" fi case "$with_dialogs" in "" | "yes" | "lucid" ) ＠＠ -4049,6 +4108,24 ＠＠ fi ;; esac +dnl this is not in xft reloaded #3 +if test "$with_xft_tabs" != "no" ; then + if test "$with_xft_emacs" = "yes" -a "$with_widgets" != "no" ; then + with_xft_tabs="yes" + else + with_xft_tabs="no" + fi +fi + +dnl this is not in xft reloaded #3 +if test "$with_xft_gauge" != "no" ; then + if test "$with_xft_emacs" = "yes" -a "$with_widgets" != "no" ; then + with_xft_gauge="yes" + else + with_xft_gauge="no" + fi +fi + all_widgets="$with_menubars $with_scrollbars $with_dialogs $with_toolbars $with_widgets" case "$all_widgets" in ＠＠ -4138,6 +4215,9 ＠＠ dnl this is not in xft reloaded #3 test "$with_xft_menubars" = "yes" && AC_DEFINE(USE_XFT_MENUBARS) +dnl these are new in sjt-xft +test "$with_xft_tabs" = "yes" && AC_DEFINE(USE_XFT_TABS) +test "$with_xft_gauge" = "yes" && AC_DEFINE(USE_XFT_GAUGE) dnl ---------------------- dnl Mule-dependent options ＠＠ -5565,6 +5645,17 ＠＠ athena ) echo " Using Athena native widgets." ;; msw ) echo " Using MS-Windows native widgets." ;; esac +dnl this is not in xft reloaded #3 +if test "$with_xft_tabs" = "yes"; then + echo " - Using Xft to render antialiased fonts in tab controls." + echo " WARNING: This feature will be replaced with a face." +fi +dnl this is not in xft reloaded #3 +if test "$with_xft_gauge" = "yes"; then + echo " - Using Xft to render antialiased fonts in progress bars." + echo " WARNING: This feature will be replaced with a face." + echo " WARNING: This feature not yet implemented; setting ignored." +fi if test "$with_dragndrop" = yes; then echo " Compiling in support for Drag'n'Drop (EXPERIMENTAL)." echo " - Drag'n'Drop prototype: $dragndrop_proto." Index: configure.usage =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/configure.usage,v retrieving revision 1.36.2.1 diff -u -r1.36.2.1 configure.usage --- configure.usage 25 Nov 2004 12:43:53 -0000 1.36.2.1 +++ configure.usage 10 Dec 2004 03:34:49 -0000 ＠＠ -2,10 +2,10 ＠＠ Set compilation and installation parameters for XEmacs, and report. -Note that for most of the following options, you can explicitly enable -them using `--OPTION=yes' and explicitly disable them using `--OPTION=no'. -This is especially useful for auto-detected options. -The option `--without-FEATURE' is a synonym for `--with-FEATURE=no'. +You can explicitly enable most of the following options using `--OPTION=yes' +and explicitly disable them using `--OPTION=no'. This is especially useful +for auto-detected options. The option `--without-FEATURE' is a synonym for +`--with-FEATURE=no'. Options marked with a (*) are auto-detected. ＠＠ -40,8 +40,8 ＠＠ all default compiler flags except those that control warnings. --cflags-warning=FLAGS Override compiler flags used to control warnings. - Normally, don't set this, as XEmacs already turns on - the maximum safe warning level. + Normally, use configure's defaults, and report + warnings as bugs to xemacs-beta(a)xemacs.org. --cpp=PROG C preprocessor to use (e.g. /usr/ccs/lib/cpp or cc -E) --cppflags=FLAGS C preprocessor flags (e.g. -I/foo or -Dfoo=bar) --libs=LIBS Additional libraries (e.g. -lfoo) ＠＠ -53,7 +53,7 ＠＠ 'site-libraries' --site-runtime-libraries=PATH List of ALL directories to search for dynamically - linked libraries at run time + linked libraries at run time. --dynamic=yes Link dynamically if supported by system. --dynamic=no Force static linking on systems where dynamic linking is the default. ＠＠ -85,33 +85,33 ＠＠ --docdir=DIR --moduledir=DIR You may also control individually where various - parts of XEmacs are installed. Note that we - recommend against explicitly setting any of these - variables. See the INSTALL file for a complete list + parts of XEmacs are installed. We recommend that you + accept configure's computed defaults for all of these + variables. See the INSTALL file for descriptions plus the reasons we advise not changing them. Run-time path-searching options: -------------------------------- ---with-site-lisp=yes Allow for a site-lisp directory in the XEmacs hierarchy - searched before the installation packages. +--with-site-lisp=yes Enable the formerly standard site-lisp directory in + the XEmacs hierarchy. It precedes the installation + packages in `load-path'. Now deprecated, with + installation of site LISP in the site-packages + hierarchy recommended. --with-site-modules=no Disable site-modules directory in the XEmacs hierarchy, which is searched before the installation modules. ---package-path=PATH Directories to search for packages to dump with xemacs. - PATH splits into three parts separated by double - colons (::), an early, a late, and a last part, - corresponding to their position in the various - system paths: The early part is always first, - the late part somewhere in the middle, and the - last part at the very back. - Only the late part gets seen at dump time. +--package-path=PATH Directories to search for LISP packages. + PATH has three parts separated by double colons (::), + early, late, and last. Only the late part gets seen + at dump time. Use of the "last" component should be + avoided. If PATH has only one component, that component is late. If PATH has two components, the first is early, the second is late. ---infopath=PATH Directories to search for Info documents, info dir - and localdir files in case run-time searching +--infopath=PATH Fallback directories to search for Info documents, + info dir and localdir files when run-time searching for them fails. ＠＠ -134,7 +134,6 ＠＠ widgets emulate Motif (mostly) but are faster. *WARNING* The Motif menubar is currently broken. Lucid menubars are the default. - *Experimental* (xft). Xft menubar, requires --with-xft --with-scrollbars=TYPE Use TYPE scrollbars (lucid, motif, athena, or no). Lucid scrollbars are the default. --with-dialogs=TYPE Use TYPE dialog boxes (lucid, motif, athena, or no). ＠＠ -159,6 +158,14 ＠＠ unpredictable. --with-xmu=no (*) For those unfortunates whose vendors don't ship Xmu. --external-widget Compile with external widget support. +--with-xft=TYPE[,TYPE[,...]] Compile with support for Xft. Valid types are + emacs (ie, in the text buffer), menubars, tabs, and + gauge (currently unsupported), and all and none. + *WARNING* This interface is UNSTABLE. `all' and + `none' should continue to be accepted + indefinitely, but the other types may + change WITHOUT NOTICE. + Non-nil values are incompatible with --with-xfs. TTY (character terminal) options: ＠＠ -166,29 +173,23 ＠＠ --with-tty=no Don't support ttys. --with-ncurses (*) Use the ncurses library for tty support. ---with-gpm (*) Compile in GPM mouse support for ttys. +--with-gpm (*) Support mouse pointer on ttys. Requires libgpm. Image options: -------------- ---with-xpm (*) Compile with support for XPM images. PRACTICALLY - REQUIRED. Although this library is nonstandard and - a real hassle to build, many basic things (e.g. - toolbars) depend on it, and you will run into - many problems without it. ---with-png (*) Compile with support for PNG images. Recommended - because the images on the About page are not viewable - without it. ---with-jpeg (*) Compile with support for JPEG images. Useful if - you are using a mail, news reader, or web browser - in XEmacs, so that JPEG images can be displayed. ---with-tiff (*) Compile with support for TIFF images. Possibly - useful, for the same reason as JPEG images. ---with-xface (*) Compile with support for X-Face mail headers. - Requires the compface package. Of doubtful - usefulness. ---with-gif=no Compile without the (builtin) support for GIF images. +--with-xpm (*) Support display of XPM images. PRACTICALLY REQUIRED. + Many basic things (e.g. toolbars) depend on it, and + you will run into many problems without it. + Requires libXpm, not standard in older X11s. +--with-png (*) Support display of PNG images. Recommended; + the About page uses PNG images. Requires libpng. +--with-jpeg (*) Support display of JPEG images. Requires libjpeg. +--with-tiff (*) Support display of TIFF images. Requires libtiff. +--with-xface (*) Support display of X-Face mail headers. + Requires compface. +--with-gif=no Disable builtin support for GIF images. Sound options: ＠＠ -202,36 +203,34 ＠＠ Later options override earlier ones for the same TYPE. The default is to autodetect all sound support except for ESD which defaults to off. ---native-sound-lib=LIB Native sound support library. Needed on Suns - with `--with-sound=native,nas' because both sound - libraries are called libaudio. +--native-sound-lib=LIB Native sound support library. For Suns with + `--with-sound=native,nas' because both sound libraries + are called libaudio. Internationalization options: ----------------------------- ---with-mule Compile with Mule (MUlti-Lingual Emacs) support, - needed to support non-Latin-1 (including Asian) - languages. ---with-xim=xlib Compile with support for X input methods, ---with-xim=motif (*) Used in conjunction with Mule support. - Use either raw Xlib to provide XIM support, or - the Motif XmIm* routines (when available). +--with-mule Enable MUlti-Lingual Extensions to Emacs. + Supported interface for input and display of languages + not supportable with ISO 8859/1. +--with-xim=xlib Support the X Input Method protocol with Mule. +--with-xim=motif (*) `xlib' uses raw Xlib to provide XIM support. `motif' + uses the Motif XmIm* routines (when available). NOTE: On some systems bugs in X11's XIM support will cause XEmacs to crash, so by default, no XIM support is compiled in, unless running on Solaris and the XmIm* routines are detected. ---with-canna (*) Compile with support for Canna (a Japanese input method - used in conjunction with Mule support). ---with-wnn (*) Compile with support for Wnn (a multi-language input - method used in conjunction with Mule support). ---with-wnn6 (*) Compile with support for the commercial package Wnn6. +--with-canna (*) Enable "native" support of Japanese input via Canna. + Requires libcanna. +--with-wnn (*) Enable "native" support of Japanese input via Wnn. + Requires libwnn (from the FreeWnn project). +--with-wnn6 (*) Enable "native" support of Japanese input via Wnn6. + Requires libwnn6, a commercial package from Omron. --with-i18n3 Compile with I18N level 3 (support for message - translation). This doesn't currently work. ---with-xfs Compile with XFontSet support for internationalized - menubar. Incompatible with `--with-xim=motif'. - `--with-menubars=lucid' (the default) is desirable. - + translation). CURRENTLY BROKEN. +--with-xfs Enable XFontSet support for localized menubar. + Incompatible with `--with-xim=motif' and `--with-xft'. File-related options: --------------------- ＠＠ -281,6 +280,11 ＠＠ Memory allocation options: -------------------------- +--pdump Portable, relocatable dumper. Now stable, and + recommended for use on all platforms. It is the + default on most platforms now. +--with-modules (*) Compile in experimental support for dynamically + loaded libraries (Dynamic Shared Objects). --rel-alloc Use the relocating allocator (default for this option is system-dependent). --with-dlmalloc Control usage of Doug Lea malloc on systems that have ＠＠ -290,12 +294,16 ＠＠ --with-debug-malloc Use the debugging malloc package. -Emacs Lisp options: -------------------- +Emacs LISP extension options: +----------------------------- --use-number-lib=TYPE Compile in support for bignums, ratios, or bigfloats using library support. TYPE must be one of "gmp" (for GNU MP), "mp" (for BSD MP), or "no" (disabled). + GNU MP is recommended over BSD MP. + *WARNING* The meaning of existing valid programs will + change on some data. Almost always this is + desirable, but there are a few subtle cases. Debugging options: ＠＠ -317,25 +325,23 ＠＠ XEmacs will still work (more or less) without them. Potentially dangerous if you don't know what you're doing. This (1) doesn't garbage-collect after loading - each file during dumping, (2) doesn't - automatically rebuild the DOC file (remove it by hand - to get it rebuilt), (3) Removes config.h, lisp.h and - associated files from the dependency lists, so changes - to these files don't automatically cause all .c files - to be rebuilt. + each file during dumping, (2) doesn't automatically + rebuild the DOC file (remove it by hand to get it + rebuilt), (3) Removes config.h, lisp.h and associated + files from the dependency lists, so changes to these + files will not force all .c files to be rebuilt. --use-union-type Enable or disable use of a union, instead of an int, for the fundamental Lisp_Object type; this - provides stricter type-checking. WARNING: This has - a tendency to trigger compiler bugs, especially in + provides stricter type-checking. *WARNING* This has + a history of triggering compiler bugs, especially in combination with other features that increase the complexity of expressions, for example `--with-mule' and `--error-checking=all'. Crashes have been - observed with union type in combination with the - two options just mentioned under various versions of - GCC as well as MSVC++ 6.0. Furthermore, many - debuggers have problems (i.e. bugs) dealing with - unions, and even for those that don't, debugging can - be inconvenient because of no syntax for entering a + observed with union type in combination with the two + options just mentioned under various versions of GCC + as well as MSVC++ 6.0. Furthermore, many debuggers + handle unions incorrectly, and those that do display + them correctly may lack any syntax for entering a union. We recommend `--use-union-type' *ONLY* for testing purposes, not for production builds. If you are using `-use-union-type' and get some weird ＠＠ -349,15 +355,7 ＠＠ --with-workshop Support the Sun WorkShop (formerly Sparcworks) development environment. ---pdump New, portable, relocatable dumper. Currently works - quite well, somewhere in beta-to-late-beta, we - might say. (Infamous for being the former - "experimental, don't-sue-me-if-your-house-collapses- - and-your-wife-leaves-you" portable dumper.) --use-kkcc Enable the use of new GC algorithms. (EXPERIMENTAL) ---with-modules (*) Compile in experimental support for dynamically - loaded libraries (Dynamic Shared Objects). ---with-xft Antialiased Fonts via Xft If successful, configure leaves its status in config.status. If Index: lwlib/lwlib-Xaw.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/lwlib/lwlib-Xaw.c,v retrieving revision 1.11 diff -u -r1.11 lwlib-Xaw.c --- lwlib/lwlib-Xaw.c 20 Sep 2004 19:19:16 -0000 1.11 +++ lwlib/lwlib-Xaw.c 10 Dec 2004 03:34:50 -0000 ＠＠ -52,6 +52,7 ＠＠ static void xaw_generic_callback (Widget, XtPointer, XtPointer); +extern int debug_xft; Boolean lw_xaw_widget_p (Widget widget) ＠＠ -738,7 +739,7 ＠＠ lw_add_value_args_to_args (val, al, &ac); if (!val->call_data) - button = XtCreateManagedWidget (val->name, labelWidgetClass, + button = XtCreateWidget (val->name, labelWidgetClass, instance->parent, al, ac); else ＠＠ -746,20 +747,21 ＠＠ if (val->type == TOGGLE_TYPE || val->type == RADIO_TYPE) { XtSetArg (al [ac], XtNstate, val->selected); ac++; - button = XtCreateManagedWidget + button = XtCreateWidget (val->name, val->type == TOGGLE_TYPE ? checkboxWidgetClass : radioWidgetClass, instance->parent, al, ac); } else { - button = XtCreateManagedWidget (val->name, commandWidgetClass, + button = XtCreateWidget (val->name, commandWidgetClass, instance->parent, al, ac); } XtRemoveAllCallbacks (button, XtNcallback); XtAddCallback (button, XtNcallback, xaw_generic_callback, (XtPointer)instance); } + /* #### this maybe can be folded into the XtCreateWidget calls above */ XtManageChild (button); return button; ＠＠ -788,7 +790,8 ＠＠ label = XtCreateManagedWidget (val->name, labelWidgetClass, parent, al, ac); - /* Do it again for arguments that have no effect until the widget is realized. */ + /* Do it again for arguments that have no effect until the widget is realized. + #### Uh, but the widget isn't realized until later? Do we mean "created"? */ ac = 0; lw_add_value_args_to_args (val, al, &ac); if (ac > 20) ＠＠ -798,6 +801,60 ＠＠ return label; } +static int debug_gauge = 1; + +static void +lw_debug_print_xt_arglist (ArgList al, int ac) +{ + int i; + for (i = 0; i < ac; i++) + fprintf (stderr, "Widget has arg of type %s with value %lu.\n", + al[i].name, (unsigned long) al[i].value); +} + +static void +lw_debug_print_class_resources (WidgetClass class_) +{ + Cardinal i; + do { + Cardinal m, n = class_->core_class.num_resources; + XtResourceList rl; + fprintf (stderr, "Class is %s (%p/%p) with %d resources.\n", + class_->core_class.class_name, class_, &(class_->core_class), n); + fprintf (stderr, " Class's resources are at %p. Converting...\n", + class_->core_class.resources); + /* resources may be compiled to an internal format */ + XtGetResourceList (class_, &rl, &m); + for (i = 0; i < m; i++) + fprintf (stderr, + " Class has a %s resource of type %s initialized from %s.\n", + rl[i].resource_class, rl[i].resource_type, rl[i].default_type); + /* special cases for commonly problematic resources */ + for (i = 0; i < m; i++) + { + if (!strcmp (rl[i].resource_class, "Font")) + { + fprintf (stderr, " Class has a Font resource.\n"); + fprintf (stderr, " Font resource is %s.\n", + (char *) rl[i].default_addr); + } + if (!strcmp (rl[i].resource_class, "FontSet")) + { + fprintf (stderr, " Class has a FontSet resource.\n"); + fprintf (stderr, " FontSet resource is %s.\n", + (char *) rl[i].default_addr); + } + if (!strcmp (rl[i].resource_class, "International")) + { + fprintf (stderr, " Class has an International resource.\n"); + fprintf (stderr, " International resource is %d.\n", + (int) rl[i].default_addr); + } + } + class_ = class_->core_class.superclass; + } while (class_ != NULL); +} + static Widget xaw_create_progress (widget_instance *instance) { ＠＠ -826,12 +883,20 ＠＠ /* add any args the user supplied for creation time */ lw_add_value_args_to_args (val, al, &ac); - scale = XtCreateManagedWidget (val->name, gaugeWidgetClass, - instance->parent, al, ac); + if (debug_gauge > 1) + lw_debug_print_class_resources (gaugeWidgetClass); + if (debug_gauge > 0) + lw_debug_print_xt_arglist (al, ac); + + scale = XtCreateWidget (val->name, gaugeWidgetClass, + instance->parent, al, ac); + /* add the callback */ if (val->call_data) - XtAddCallback (scale, XtNgetValue, xaw_generic_callback, (XtPointer)instance); + XtAddCallback (scale, XtNgetValue, xaw_generic_callback, + (XtPointer) instance); + /* #### this maybe can be folded into the XtCreateWidget call above */ XtManageChild (scale); return scale; ＠＠ -864,7 +929,7 ＠＠ /* add any args the user supplied for creation time */ lw_add_value_args_to_args (val, al, &ac); - text = XtCreateManagedWidget (val->name, asciiTextWidgetClass, + text = XtCreateWidget (val->name, asciiTextWidgetClass, instance->parent, al, ac); /* add the callback */ Index: lwlib/lwlib-colors.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/lwlib/Attic/lwlib-colors.c,v retrieving revision 1.1.2.1 diff -u -r1.1.2.1 lwlib-colors.c --- lwlib/lwlib-colors.c 25 Nov 2004 12:44:04 -0000 1.1.2.1 +++ lwlib/lwlib-colors.c 10 Dec 2004 03:34:50 -0000 ＠＠ -33,7 +33,7 ＠＠ #include <X11/ShellP.h> /* for ShellWidget */ #include "lwlib-colors.h" -extern int debug_xft; +static int debug_colors = 1; #ifdef __cplusplus #define X_CLASSFIELD c_class ＠＠ -64,14 +64,15 ＠＠ } if (*visual == CopyFromParent || !*visual) { - if (debug_xft) + if (debug_colors > 1) fprintf (stderr, "\nvisual_info_from_widget:" " failed, using DefaultVisualOfScreen"); *visual = DefaultVisualOfScreen (XtScreen (widget)); /* #### Why doesn't DefaultDepth0fScreen work here? */ - *depth = DefaultDepth (XtDisplay (widget), 0); + /* *depth = DefaultDepth (XtDisplay (widget), 0); */ + *depth = DefaultDepthOfScreen (XtScreen (widget)); } - else if (debug_xft) + else if (debug_colors > 1) fprintf (stderr, "\nvisual_info_from_widget: succeeded"); } Index: lwlib/xlwmenu.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/lwlib/xlwmenu.c,v retrieving revision 1.37.2.1 diff -u -r1.37.2.1 xlwmenu.c --- lwlib/xlwmenu.c 25 Nov 2004 12:44:06 -0000 1.37.2.1 +++ lwlib/xlwmenu.c 10 Dec 2004 03:34:51 -0000 ＠＠ -90,13 +90,12 ＠＠ *fontList resource set, or at least know how to deal with this. */ XtRString, (XtPointer) "-*-helvetica-bold-r-*-*-*-120-*-*-*-*-iso8859-1"}, #else -#ifdef USE_XFT_MENUBARS - {XtNfont, XtCFont, XtRString, sizeof (String), - offset(menu.renderFontSpec), - XtRString, (XtPointer) "-*-helvetica-bold-r-*-*-*-120-*-*-*-*-iso8859-1"}, -#else {XtNfont, XtCFont, XtRFontStruct, sizeof(XFontStruct *), offset(menu.font), XtRString, (XtPointer) "XtDefaultFont"}, +#ifdef USE_XFT_MENUBARS + {XtNxftFont, XtCXftFont, XtRString, sizeof (String), + offset(menu.renderFontSpec), + XtRString, (XtPointer) "Helvetica-12:bold"}, #endif # ifdef USE_XFONTSET /* #### Consider using the same method as for Motif; see the comment in Index: lwlib/xlwmenuP.h =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/lwlib/xlwmenuP.h,v retrieving revision 1.4.20.1 diff -u -r1.4.20.1 xlwmenuP.h --- lwlib/xlwmenuP.h 25 Nov 2004 12:44:06 -0000 1.4.20.1 +++ lwlib/xlwmenuP.h 10 Dec 2004 03:34:51 -0000 ＠＠ -32,11 +32,11 ＠＠ XmFontList font_list_2; XmFontList fallback_font_list; #else + XFontStruct * font; #ifdef USE_XFT_MENUBARS + /* #### Fix naming convention here */ String renderFontSpec; XftFont *renderFont; -#else - XFontStruct * font; #endif # ifdef USE_XFONTSET XFontSet font_set; Index: lwlib/xlwtabs.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/lwlib/xlwtabs.c,v retrieving revision 1.4.2.2 diff -u -r1.4.2.2 xlwtabs.c --- lwlib/xlwtabs.c 3 Dec 2004 07:40:58 -0000 1.4.2.2 +++ lwlib/xlwtabs.c 10 Dec 2004 03:34:52 -0000 ＠＠ -148,7 +148,7 ＠＠ offset(selectInsensitive), XtRImmediate, (XtPointer) True}, {XtNfont, XtCFont, XtRFontStruct, sizeof(XFontStruct *), offset(font), XtRString, (XtPointer) XtDefaultFont}, -#ifdef USE_XFT +#ifdef USE_XFT_TABS /* #### Maybe use "-*-helvetica-bold-r-*-*-*-120-*-*-*-*-iso8859-1" here? */ {XtNxftFont, XtCXftFont, XtRString, sizeof (String), offset(renderFontSpec), XtRString, ＠＠ -416,7 +416,7 ＠＠ ((TabsConstraints)((tab)->core.constraints))->tabs.visible)  -#ifdef USE_XFT +#ifdef USE_XFT_TABS extern int debug_xft; /* from lwlib-face.c */ /* #### duplicated from xlwmenu.c -- CLEAN THIS SHIT UP! ＠＠ -472,7 +472,7 ＠＠ */ newTw->tabs.tab_height = 2 * newTw->tabs.internalHeight + SHADWID ; -#ifdef USE_XFT +#ifdef USE_XFT_TABS /* must get font here to do this right, we should add a new Xt Resource type + conversion function ＠＠ -487,11 +487,11 ＠＠ newTw->tabs.tab_height += newTw->tabs.renderFont->ascent + newTw->tabs.renderFont->descent; #endif /* XFT_USE_HEIGHT_NOT_ASCENT_DESCENT */ -#else /* ! USE_XFT */ +#else /* ! USE_XFT_TABS */ if (newTw->tabs.font != NULL) newTw->tabs.tab_height += newTw->tabs.font->max_bounds.ascent + newTw->tabs.font->max_bounds.descent; -#endif /* ! USE_XFT */ +#endif /* ! USE_XFT_TABS */ /* if size not explicitly set, set it to our preferred size now. */ ＠＠ -666,7 +666,7 ＠＠ int i ; if( -#ifdef USE_XFT +#ifdef USE_XFT_TABS tw->tabs.renderFont != curtw->tabs.renderFont || #else tw->tabs.font != curtw->tabs.font || ＠＠ -676,7 +676,7 ＠＠ { tw->tabs.tab_height = 2 * tw->tabs.internalHeight + SHADWID; -#ifdef USE_XFT +#ifdef USE_XFT_TABS if (tw->tabs.renderFont != NULL) #if XFT_USE_HEIGHT_NOT_ASCENT_DESCENT tw->tabs.tab_height += tw->tabs.renderFont->height; ＠＠ -684,11 +684,11 ＠＠ tw->tabs.tab_height += tw->tabs.renderFont->ascent + tw->tabs.renderFont->descent; #endif /* XFT_USE_HEIGHT_NOT_ASCENT_DESCENT */ -#else /* ! USE_XFT */ +#else /* ! USE_XFT_TABS */ if (tw->tabs.font != NULL) tw->tabs.tab_height += tw->tabs.font->max_bounds.ascent + tw->tabs.font->max_bounds.descent; -#endif /* ! USE_XFT */ +#endif /* ! USE_XFT_TABS */ /* Tab size has changed. Resize all tabs and request a new size */ for(i=0, childP=tw->composite.children; ＠＠ -705,7 +705,7 ＠＠ if (tw->core.background_pixel != curtw->core.background_pixel || tw->core.background_pixmap != curtw->core.background_pixmap || -#ifdef USE_XFT +#ifdef USE_XFT_TABS tw->tabs.renderFont != curtw->tabs.renderFont #else tw->tabs.font != curtw->tabs.font ＠＠ -1579,8 +1579,8 ＠＠ GC gc ; int x,y ; -#ifdef USE_XFT - if (debug_xft) fprintf (stderr, "\nDrawTab called.\n"); +#ifdef USE_XFT_TABS + if (debug_xft > 0) fprintf (stderr, "DrawTab called.\n"); #endif if( !XtIsRealized((Widget)tw)) ＠＠ -1595,7 +1595,7 ＠＠ Window win = XtWindow((Widget)tw) ; String lbl = tab->tabs.label != NULL ? tab->tabs.label : XtName(child) ; -#ifdef USE_XFT +#ifdef USE_XFT_TABS XftColor color; XftColor colorBG; Colormap cmap = tw->core.colormap; ＠＠ -1606,17 +1606,14 ＠＠ colorBG = xft_convert_color (dpy, cmap, visual, tw->core.background_pixel, 0); - if (debug_xft) - { - fprintf (stderr, "\n(Re)drawing labels."); - fprintf (stderr, "\nlbl: %s.", lbl); - } + if (debug_xft > 0) + fprintf (stderr, "(Re)drawing labels.\n"); #endif if (XtIsSensitive(child)) { gc = tw->tabs.foregroundGC; -#ifdef USE_XFT +#ifdef USE_XFT_TABS color = xft_convert_color (dpy, cmap, visual, tab->tabs.foreground, 0); #else ＠＠ -1638,7 +1635,7 ＠＠ tab->tabs.greyAlloc = True; } gc = tw->tabs.greyGC; -#ifdef USE_XFT +#ifdef USE_XFT_TABS color = xft_convert_color (dpy, cmap, visual, tab->tabs.grey, 0); #else XSetForeground(dpy, gc, tab->tabs.grey); ＠＠ -1663,14 +1660,14 ＠＠ } if (lbl != NULL && -#ifdef USE_XFT +#ifdef USE_XFT_TABS tw->tabs.renderFont != NULL #else tw->tabs.font != NULL #endif ) { -#ifdef USE_XFT +#ifdef USE_XFT_TABS XftDraw *xftDraw = XftDrawCreate (dpy, win, visual, cmap); XftFont *renderFont = tw->tabs.renderFont; /* [[ #### FMH!! something is tromping on this string! ]] ＠＠ -1685,9 +1682,9 ＠＠ /* #### unnecessary? */ /* draw background rect */ #if 1 - if (debug_xft) - fprintf (stderr, "\nColor: pixel=%08lx, red=%04x," - " green=%04x, blue=%04x, alpha=%04x.", + if (debug_xft > 0) + fprintf (stderr, "background color: pixel=%08lx, r=%04x," + " g=%04x, b=%04x, alpha=%04x.\n", colorDBG.pixel, colorDBG.color.red, colorDBG.color.green, colorDBG.color.blue, colorDBG.color.alpha); ＠＠ -1698,7 +1695,7 ＠＠ glyphinfo.width, glyphinfo.height); #endif /* draw text */ - if (debug_xft) + if (debug_xft > 0) { /* [[ #### URK! fontconfig frees size!! ]] That has got to be an artifact of something writing ＠＠ -1707,17 +1704,17 ＠＠ FcValue size; FcPatternGet (renderFont->pattern, FC_FAMILY, 0, &name); FcPatternGet (renderFont->pattern, FC_SIZE, 0, &size); - fprintf (stderr, "\nlbl: %s.", lbl); - fprintf (stderr, "\nColor: pixel=%08lx, red=%04x," - " green=%04x, blue=%04x, alpha=%04x.", + fprintf (stderr, "label: %s.", lbl); + fprintf (stderr, "foreground color: pixel=%08lx, r=%04x," + " g=%04x, b=%04x, alpha=%04x.\n", color.pixel, color.color.red, color.color.green, color.color.blue, color.color.alpha); - fprintf (stderr, "\nExtent: x=%d, y=%d, xOffset=%d," - " yOffset=%d, height=%d, width=%d", + fprintf (stderr, "extent: x=%d, y=%d, xOffset=%d," + " yOffset=%d, height=%d, width=%d.\n", glyphinfo.x, glyphinfo.y, glyphinfo.xOff, glyphinfo.yOff, glyphinfo.height, glyphinfo.width); - fprintf (stderr, "\nFont: name=%s-%.1f, height=%d," - " ascent=%d, descent=%d", + fprintf (stderr, "font: name=%s-%.1f," + " height=%d, ascent=%d, descent=%d.\n", name.u.s, size.u.d, renderFont->height, renderFont->ascent, renderFont->descent); } ＠＠ -1735,9 +1732,6 ＠＠ if (child == tw->tabs.hilight) DrawHighlight(tw, child, False); -#ifdef USE_XFT - if (debug_xft) fprintf (stderr, "\n"); -#endif } ＠＠ -1939,7 +1933,7 ＠＠ TabsWidget tw = (TabsWidget)XtParent(w) ; String lbl = tab->tabs.label != NULL ? tab->tabs.label : XtName(w); -#ifdef USE_XFT +#ifdef USE_XFT_TABS XftFont *font = tw->tabs.renderFont; #else XFontStruct *font = tw->tabs.font; ＠＠ -1958,7 +1952,7 ＠＠ if( lbl != NULL && font != NULL ) { -#ifdef USE_XFT +#ifdef USE_XFT_TABS tab->tabs.width += x_xft_text_width (XtDisplay(tw), font, lbl, (int)strlen(lbl)) + iw; tab->tabs.l_y = (tw->tabs.tab_height ＠＠ -1967,12 +1961,12 ＠＠ - tw->tabs.renderFont->descent)/2; if (debug_xft) { - fprintf (stderr, "\nTab: height=%d, width=%d, baseline=%d", - tw->tabs.tab_height, tab->tabs.width, tab->tabs.l_y); - fprintf (stderr, "\nFont: height=%d, ascent=%d, descent=%d\n", - tw->tabs.renderFont->height, - tw->tabs.renderFont->ascent, - tw->tabs.renderFont->descent); + fprintf (stderr, "tab: height=%d, width=%d, baseline=%d.\n", + tw->tabs.tab_height, tab->tabs.width, tab->tabs.l_y); + fprintf (stderr, "font: height=%d, ascent=%d, descent=%d.\n", + tw->tabs.renderFont->height, + tw->tabs.renderFont->ascent, + tw->tabs.renderFont->descent); } #else tab->tabs.width += XTextWidth (font, lbl, (int)strlen(lbl)) + iw; ＠＠ -2364,7 +2358,7 ＠＠ values.background = tw->core.background_pixel; values.font = -#ifdef USE_XFT +#ifdef USE_XFT_TABS None; #else tw->tabs.font->fid; ＠＠ -2374,13 +2368,13 ＠＠ tw->tabs.foregroundGC = XtAllocateGC(w, w->core.depth, -#ifndef USE_XFT +#ifndef USE_XFT_TABS GCFont| #endif GCBackground|GCLineStyle, &values, GCForeground, -#ifdef USE_XFT +#ifdef USE_XFT_TABS GCFont| #endif GCSubwindowMode|GCGraphicsExposures|GCDashOffset| ＠＠ -2395,7 +2389,7 ＠＠ values.background = tw->core.background_pixel; values.font = -#ifdef USE_XFT +#ifdef USE_XFT_TABS None; #else tw->tabs.font->fid; ＠＠ -2409,12 +2403,12 ＠＠ tw->tabs.greyGC = XtAllocateGC(w, w->core.depth, -#ifndef USE_XFT +#ifndef USE_XFT_TABS GCFont| #endif GCBackground|GCStipple|GCFillStyle, &values, GCForeground, -#ifdef USE_XFT +#ifdef USE_XFT_TABS GCFont| #endif GCSubwindowMode|GCGraphicsExposures|GCDashOffset| ＠＠ -2425,14 +2419,14 ＠＠ { tw->tabs.greyGC = XtAllocateGC(w, w->core.depth, -#ifdef USE_XFT +#ifdef USE_XFT_TABS 0L, #else GCFont, #endif &values, GCForeground, -#ifdef USE_XFT +#ifdef USE_XFT_TABS GCFont| #endif GCBackground|GCSubwindowMode|GCGraphicsExposures|GCDashOffset| Index: lwlib/xlwtabsP.h =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/lwlib/xlwtabsP.h,v retrieving revision 1.2.20.1 diff -u -r1.2.20.1 xlwtabsP.h --- lwlib/xlwtabsP.h 25 Nov 2004 12:44:07 -0000 1.2.20.1 +++ lwlib/xlwtabsP.h 10 Dec 2004 03:34:52 -0000 ＠＠ -41,7 +41,7 ＠＠ #endif #include "xlwtabs.h" -#ifdef USE_XFT +#ifdef USE_XFT_TABS #include <X11/Xft/Xft.h> #endif ＠＠ -73,10 +73,10 ＠＠ typedef struct { /* resources */ XFontStruct *font ; -#ifdef USE_XFT_MENUBARS +#ifdef USE_XFT_TABS String renderFontSpec; XftFont *renderFont; #endif Index: man/internals/internals.texi =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/man/internals/internals.texi,v retrieving revision 1.52 diff -u -r1.52 internals.texi --- man/internals/internals.texi 16 Nov 2004 07:37:31 -0000 1.52 +++ man/internals/internals.texi 10 Dec 2004 03:35:05 -0000 ＠＠ -638,6 +638,7 ＠＠ * Future Work -- Display Tables:: * Future Work -- Making Elisp Function Calls Faster:: * Future Work -- Lisp Engine Replacement:: +* Future Work -- Better Rendering Support:: Future Work -- Toolbars ＠＠ -17523,6 +17524,7 ＠＠ * Future Work -- Display Tables:: * Future Work -- Making Elisp Function Calls Faster:: * Future Work -- Lisp Engine Replacement:: +* Future Work -- Better Rendering Support:: ＠end menu ＠node Future Work -- General Suggestions, Future Work -- Elisp Compatibility Package, Future Work, Future Work ＠＠ -22362,7 +22364,7 ＠＠ ＠end itemize -＠node Future Work -- Lisp Engine Replacement, , Future Work -- Making Elisp Function Calls Faster, Future Work +＠node Future Work -- Lisp Engine Replacement, Future Work -- Better Rendering Support, Future Work -- Making Elisp Function Calls Faster, Future Work ＠section Future Work -- Lisp Engine Replacement ＠cindex future work, lisp engine replacement ＠cindex lisp engine replacement, future work ＠＠ -22972,6 +22974,332 ＠＠ ad-hocked since kingdom come, and it's high time that we make this work properly so that it could be relied upon, and a lot of things could "just work". + + +＠node Future Work -- Better Rendering Support, , Future Work -- Lisp Engine Replacement, Future Work +＠section Future Work -- Better Rendering Support +＠cindex future work, better rendering support +＠cindex better rendering support, future work + +This section was written by Stephen Turnbull <stephen＠(a)xemacs.org&gt;, so +don't blame Ben (or Eric and Matthias, for that matter). Feel free to +add, edit, and share the blame, guys! + +＠c #### make these ＠urlref's!! +As of late November 2004, this principally means adding support for the +＠file{Xft} library, which provides a more robust ＠emph{font +configuration} mechanism via Keith Packard's ＠file{fontconfig} library +improved glyph rendering, including antialiasing, via the +＠file{freetype} library, and client-side rendering (saving bandwidth and +server memory) via the ＠file{XRender extension}. In fact, patches which +provide Xft support have been available for several years, but the +authors have been unwilling to deal with several important issues which +block integration. These are ＠emph{Mule}, and more generally, +＠emph{face} support; ＠emph{widget} support (including the toolbar and +menubar); and ＠emph{redisplay refactoring}. + +＠c Describe Alexey Gladkov and Yury Konovalov's work. + +However, in late 2003 Eric Knauel <knauel＠(a)informatik.uni-tuebingen.de&gt; +and Matthias Neubauer <neubauer＠(a)informatik.uni-freiburg.de&gt; put forward +a relatively complete patch which was robust to daily use in ISO 8859-1 +locales, and Stephen Turnbull began work on the integration issues. At +this point a (private) CVS branch is available for Stephen's patch +(branch point tag ＠samp{sjt-xft-bp}, branch tag ＠samp{sjt-xft}), and +one may be made available for the Knauel-Matthias patch soon. + +＠menu +* Better Rendering Support -- Issues:: +* Better Rendering Support -- Implementation:: +* Better Rendering Support -- Current Status:: +* Better Rendering Support -- Configuration with the Interim Patches:: +＠end menu + + +＠node Better Rendering Support -- Issues, Better Rendering Support -- Implementation, , Future Work -- Better Rendering Support +＠subsection Better Rendering Support -- Issues +＠cindex better rendering support, issues +＠cindex issues, better rendering support + +Of course it's ``unfair'' to demand that the implementers of a nice +feature like anti-aliasing support deal with accumulated cruft of the +last few years, but somebody must, sometime soon. Even core developers +are complaining about how slow XEmacs is in some applications, and there +is reason to believe that some of the problem is in redisplay. +Adding more ＠emph{ad hoc} features to redisplay will make the whole +module more complex and unintelligible. Even if it doesn't inherently +further detract from efficiency, it will surely make reform and +refactoring harder. + +Similar considerations apply to Mule support. If Xft support is not +carefully designed, or implemented with Mule support soon, it will +undoubtedly make later Mule implementation far more difficult than it +needs to be, and require redundant work be done (＠emph{e.g.}, on +＠samp{Options} menu support). + +Besides the design issue---and many users are requesting more +flexibility, primarily face support, from the widgets---with widget +support there is also an aesthetic issue. It is horribly unimpressive to +have clunky bitmapped fonts on the decorations when pleasant antialiased +fonts are available in the buffer. + +Finally, these issues interact. Widgets and faces are inherently +heavyweight objects, requiring orders of magnitude more computation than +simply displaying a string in a fixed font. This will have an +efficiency impact, of course. And they interact with each other; Mule +was designed for use in buffers and display in Emacs windows---but a +widget's content is usually not a buffer, and widgets need not be +displayed in a window, but may appear in other contexts, especially in +the gutters. So specifiers will probably have to be reworked, in order +to properly support display of different faces in non-buffer, non-window +contexts. + +＠node Better Rendering Support -- Implementation, Better Rendering Support -- Current Status, Better Rendering Support -- Issues, Future Work -- Better Rendering Support +＠subsection Better Rendering Support -- Implementation +＠cindex better rendering support, implementation +＠cindex implementation, better rendering support + +Stephen is thinking in terms of the following components of a +comprehensive proposal. + +＠table ＠strong +＠item Font configuration +In XEmacs, font configuration is handled via ＠emph{faces}. Currently +XEmacs uses a special type of ＠emph{font specifier} to map XEmacs +locales to font names. Especially under X11, this can cause annoying +problems because of the unreliability of X servers' mappings from +＠samp{XLFD} names to X11 fonts, over which XEmacs has no influence +whatsoever. However, the ＠file{fontconfig} library which is used with +＠file{Xft} provides much more reliable mapping, along with a more +reliably parsable naming scheme similar to that used by TrueType fonts +on MS Windows and the Macintosh. Since the capabilities of font +specifiers and ＠file{fontconfig} overlap, we should consider using +＠file{fontconfig} instead of ＠samp{XLFD} names. This implies that use +of ＠file{Xft}'s rendering functionality should be separated from use of +＠file{fontconfig}. + +＠item Font menus +The ＠samp{Options->Font} and ＠samp{Options->Font Sizes} menus are +broken, by design, not just by ＠file{Xft}. Although they work better +in Eric and Matthias's patch than in Stephen's, even their version has +the problem that many fonts are unavailable because they don't match the +current size---which is very strange, since ＠file{Xft} fonts are of +course scalable. But the whole idea of requiring that the font match +the size is strange. And the ＠samp{Options->Font Weights} menu is just +disabled, and has been for eons. + +＠item X resources +Currently in Stephen's patch there are ＠emph{four} treatments of font +resources. There are the ＠samp{XEmacs.(a)var{face}.attributeFont} +resources used to set a single global font specification. In the +widgets, some (still) have a ＠samp{font} resource using the automatic +＠file{Xt} resource conversion to ＠samp{FontStruct}, some have separate +＠samp{font} and ＠samp{xftFont} resources with the former automatically +converted to ＠samp{FontStruct} by ＠file{Xt} and the latter left as a +string, to be converted by ＠samp{XftParseName} later, and some have a +single ＠samp{font} resource which is converted to ＠samp{FontStruct} by +＠file{Xt} or the latter left as a string, depending on whether +＠file{Xft} was enabled by ＠samp{configure} or not. + +Stephen thinks that all of these should be converted to use the face +approach, perhaps with some way to set specifications for individual +widgets, frames, or buffers. This will require some careful design work +to incorporate face support in the widgets. We should just accept any +or all of ＠samp{font}, ＠samp{fontSet}, and ＠samp{fontList} resources, +treat them all as lists of font names, either ＠samp{XLFD}- or +＠file{fontconfig}-style, parse them ourselves (ie, not use the ＠file{Xt} +resource manager), and add them to font specifiers as appropriate. But +this will require a bit of thought to obey POLA vis-a-vis usual ＠file{Xt} +conventions. + +＠item Rendering engine objects +With the introduction of the ``Xft patch,'' the X11, Macintosh, and MS +Windows platforms are all able to support multiple font rendering +engines in the same binary. Generically, there are several tasks that +must be accomplished to render text on the display. In both cases the +code is rather disorganized, with substantial cross-platform duplication +of similar routines. While it may not be worthwhile to go the whole way +to ＠samp{RENDERER_HAS_METHOD} and ＠samp{MAYBE_RENDMETH}, refactoring +these modules around the notion of interfacing a ``generic rendering +engine interface'' to ``text'' seems like a plausible way to focus this +work. + +＠item Colors, fonts, and faces +Besides the rendering engine itself, the XEmacs implementations of these +objects are poorly supported by current widget implementations, +including the traditional menubar and toolbar, as well as the more +recent button, tab control, and progress bar widgets. The refactoring +suggested under ``Rendering engine objects'' should be conducted with an +eye to making these widgets support faces, perhaps even to the extent of +allowing rendering to X pixmaps (which some Athena widgets support, +although they will not support rendering via Xft directly). Especially +with ＠samp{XRender} technology this should not be horribly inefficient. + +＠item Specifiers, charsets, and languages +Traditionally Mule uses a rather rigid and low-level abstraction, the +＠emph{charset}, to characterize font repertoires. Unfortunately, +support for a given charset is generally neither necessary nor +sufficient to support a language. Worse, although X11's only means for +indicating font repertoires is the font's ＠emph{registry}, the actual +repertoire of many fonts is either deficient or font-dependent. The +only convenience is that the registry maps directly to a Mule charset in +most cases, and vice versa. + +To date, XEmacs Mule has supported identification of appropriate fonts +to support a language's repertoire of characters by identifying the +repertoire as a subset of a union of charsets. To each charset there is +a regular expression matching the registry portion of a font name. Then +instantiation of a font proceeds by identifying the specifier domain, +and then walking down the list of specifications, matching the regexp +against font names until a match is found. That font is requested from +the system, and if not found, the process continues similarly until a +font that can be loaded is found. + +This has several problems. First, there's no guarantee that the union +will be disjoint. This problem manifests both in the case of display of +Unicode representations of text in the ＠samp{POSIX} default locale, +where glyphs are typically drawn from several inappropriate fonts. A +similar problem often occurs, though for a different reason, in +multilingual messages composed using ＠file{Gnus}'s ＠samp{message-mode} +and MIME support. This problem ＠emph{cannot} be avoided with the +current design; it is quite possible that a font desired in one context +will be shadowed by a font intended to get higher priority in a +semantically different but syntactically similar (as far as Mule can +tell) context. (Of course, one could attach a different face as a text +property, but that requires programming support; it can't be done by +user configuration.) The problem is only exacerbated as more and more +Unicode fonts, supporting large repertoires with substantial overlap +across fonts, are designed and published. + +A second problem is that registry names are often inaccurate. For +example, the Japanese JIS X 0208 standard was first published in 1978 +(as a relabelling of an older standard). It was then revised in 1983, +again in 1990, and once again in 2000, with slight changes to the +repertoire and mapping in each revision. Technically, these standards +can be distinguished in properly named fonts as ＠samp{jisx0208.1978}, +(a)samp{jisx0208.1983}, ＠samp{jisx0208.1990}, ＠samp{jisx0208.2000}, but +all of them are commonly simply labelled ＠samp{jisx0208}, and Western +distributors, of course, generally lack the expertise to correctly +relabel them. + +A third problem is that you generally can't tell if there are ``holes'' +in the repertoire until you try to display the glyph. + +All of this tends to break standard idioms for handling Mule fonts in +＠file{init} files because they depend on charsets being disjoint +repertoires. + +The TrueType fonts (and the later OpenType standard) provides for a +proper character set query (as a Boolean vector indexed by Unicode code +points), as well as providing a list of supported languages. + +I propose that we take advantage of these latter facilities by allowing +a font to be specified either as a string (a font name), or as a list +whose head is the font name and whose tail is a list of languages and +Mule charsets (for backward compatibility) that user intends to use the +font to display. This will probably require a change to the specifier +code. + +As mentioned above, specifiers will probably also have to be enhanced to +recognize ＠samp{widget} locales and domains, instead of the current hack +where special ＠samp{widget} and ＠samp{gui-element} faces are created. + +＠item Customize + +Customize needs to deal with all this stuff!! +＠end table + + +＠node Better Rendering Support -- Current Status, Better Rendering Support -- Configuration with the Interim Patches, Better Rendering Support -- Implementation, Future Work -- Better Rendering Support +＠subsection Better Rendering Support -- Current Status +＠cindex better rendering support, current status +＠cindex current status, better rendering support + +Stephen has a branch containing his stuff in XEmacs CVS. The branch +point tag is ＠samp{sjt-xft-bp}, roughly corresponding to XEmacs 21.5.18, +and branch tag is ＠samp{sjt-xft}. + +＠subsubsection Bugs Reported in sjt-xft + +＠table ＠code +＠item ChangeLogs +A lot of these, especially for Eric and Matthias's work, are missing. +Mea culpa. + +＠item Options->Font +＠itemx Options->Font Size +These menus don't work. All fonts are greyed out. All sizes are +available, but many (most?) faces don't change size, in particular, +＠samp{default} does not. +＠end table + + +＠node Better Rendering Support -- Configuration with the Interim Patches, , Better Rendering Support -- Current Status, Future Work -- Better Rendering Support +＠subsection Better Rendering Support -- Configuration with the Interim Patches +＠cindex better rendering support, configuration with the interim patches +＠cindex configuration with the interim patches, better rendering support + +For Stephen's ＠samp{sjt-xft} branch, you should keep the following in +mind when configuring: + +＠itemize +＠item +The only way to configure widget fonts at the present time is to use X +resources (or hack the source and rebuild). Currently supported widgets +are +＠itemize +＠item +menubars +＠item +tab controls +＠end itemize + +Here are the resources I use. ＠strong{Warning:} ＠emph{This interface +will change.} The tab control has separate Font and XftFont resources, +and uses the X resource manager to instantiate a FontStruct from the +Font resource. The menubar on the other hand uses Font for both, but +converts to FontStruct if Xft is not compiled in, and leaves it as a +string if Xft is compiled in. There is no equivalent facility for +XftFont yet, and creating one that handles both FontStruct and XftFont +depending on XEmacs's configuration and the font name seems error-prone +at best. Probably we will revert to a simple string representation for +this resource, and convert to a face in XEmacs rather than a font in +Xt/Xft. +＠example +XEmacs*Tabs.xftFont: Bitstream Vera Sans-18 +XEmacs*menubar.font: Bitstream Vera Sans-18 +XEmacs.modeline.attributeFont: Bitstream Charter-26 +XEmacs.default.attributeFont: Bitstream Vera Sans Mono-24 +＠end example +No, I don't understand why I need to use different point sizes to get +what looks like good balance to my eyes. I do highly recommend use of a +proportional font in the modeline because it allows a lot more text to +fit there. + +＠item +Because fonts supporting other languages tend to support English as +well, if you want to use one font for English and another for the other +language, you must use the ＠code{append} method when adding font +specifications for the other language. + +However, this leaves you with a problem if you want to change the other +language's font: you have to remove the existing specification so it +won't shadow the new one when you append. + +I use ＠code{define-specifier-tag} like this: + +＠example +(define-specifier-tag 'lang-ja) +;; No, I don't try to do real work with this font! But it makes it +;; obvious that I got the requested font. :-) +(set-face-font 'default "AirCut-14") +(set-face-font 'default "Kochi Mincho-14" nil '(lang-ja) 'append) +;; Oops, too sober. Try something to match AirCut. +(set-face-font 'default "Mikachan-14" + nil '(lang-ja) 'remove-tag-set-append) +＠end example +＠end itemize + ＠node Future Work Discussion, Old Future Work, Future Work, Top ＠chapter Future Work Discussion Index: src/compiler.h =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/compiler.h,v retrieving revision 1.8 diff -u -r1.8 compiler.h --- src/compiler.h 4 Nov 2004 23:06:17 -0000 1.8 +++ src/compiler.h 10 Dec 2004 03:35:20 -0000 ＠＠ -215,6 +215,11 ＠＠ # else # define USED_IF_MULE_OR_CHECK_TEXT(decl) UNUSED (decl) # endif +# ifdef USE_XFT +# define USED_IF_XFT(decl) decl +# else +# define USED_IF_XFT(decl) UNUSED (decl) +# endif #endif /* UNUSED */ #ifdef DEBUG_XEMACS Index: src/config.h.in =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/config.h.in,v retrieving revision 1.95.2.1 diff -u -r1.95.2.1 config.h.in --- src/config.h.in 25 Nov 2004 12:44:11 -0000 1.95.2.1 +++ src/config.h.in 10 Dec 2004 03:35:20 -0000 ＠＠ -207,9 +207,10 ＠＠ /* Compile with support for Xft? */ #undef USE_XFT - -/* Compile with support for Xft menubars? */ +/* Per-widget stuff will go away? */ #undef USE_XFT_MENUBARS +#undef USE_XFT_TABS +#undef USE_XFT_GAUGE /* Defines for building X applications */ #ifdef HAVE_X_WINDOWS Index: src/emacs.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/emacs.c,v retrieving revision 1.150.2.1 diff -u -r1.150.2.1 emacs.c --- src/emacs.c 25 Nov 2004 12:44:12 -0000 1.150.2.1 +++ src/emacs.c 10 Dec 2004 03:35:21 -0000 ＠＠ -954,7 +954,9 ＠＠ if (argmatch (argv, argc, "-si", "--show-inline-info", 0, NULL, &skip_args)) { #if defined (PDUMP) && !defined (WIN32_NATIVE) - printf ("%u %u\n", dumped_data_max_size (), dumped_data_align_offset ()); + /* #### We really should check for sizeof (size_t) > sizeof (long) */ + printf ("%lu %lu\n", (unsigned long) dumped_data_max_size (), + (unsigned long) dumped_data_align_offset ()); #else printf ("Portable dumper not configured or windows native; -si just forces exit.\n"); #endif Index: src/glyphs-widget.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/glyphs-widget.c,v retrieving revision 1.17 diff -u -r1.17 glyphs-widget.c --- src/glyphs-widget.c 4 Nov 2004 23:06:33 -0000 1.17 +++ src/glyphs-widget.c 10 Dec 2004 03:35:22 -0000 ＠＠ -1819,7 +1819,7 ＠＠ IIFORMAT_VALID_KEYWORD (scrollbar, Q_face, check_valid_face); } -static void image_instantiator_progress_guage (void) +static void image_instantiator_progress_gauge (void) { INITIALIZE_IMAGE_INSTANTIATOR_FORMAT (progress_gauge, "progress-gauge"); IIFORMAT_HAS_SHARED_METHOD (progress_gauge, validate, widget); ＠＠ -1927,7 +1927,7 ＠＠ image_instantiator_edit_fields(); image_instantiator_combo_box(); image_instantiator_scrollbar(); - image_instantiator_progress_guage(); + image_instantiator_progress_gauge(); image_instantiator_tree_view(); image_instantiator_tab_control(); image_instantiator_labels(); Index: src/glyphs-x.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/glyphs-x.c,v retrieving revision 1.80.2.1 diff -u -r1.80.2.1 glyphs-x.c --- src/glyphs-x.c 25 Nov 2004 12:44:14 -0000 1.80.2.1 +++ src/glyphs-x.c 10 Dec 2004 03:35:23 -0000 ＠＠ -2325,22 +2325,27 ＠＠ bcolor = COLOR_INSTANCE_X_COLOR (XCOLOR_INSTANCE (pixel)); lw_add_widget_value_arg (wv, XtNbackground, bcolor.pixel); - /* #### What to do about Xft? */ + { + /* #### What to do about Xft? */ + Lisp_Object face = IMAGE_INSTANCE_WIDGET_FACE (ii); + XFontStruct *fs = + FONT_INSTANCE_X_FONT (XFONT_INSTANCE (query_string_font + (IMAGE_INSTANCE_WIDGET_TEXT (ii), + face, + domain))); #ifdef LWLIB_WIDGETS_MOTIF - fontList = XmFontListCreate - (FONT_INSTANCE_X_FONT - (XFONT_INSTANCE (query_string_font - (IMAGE_INSTANCE_WIDGET_TEXT (ii), - IMAGE_INSTANCE_WIDGET_FACE (ii), - domain))), XmSTRING_DEFAULT_CHARSET); - lw_add_widget_value_arg (wv, XmNfontList, (XtArgVal)fontList); + fontList = XmFontListCreate (fs, XmSTRING_DEFAULT_CHARSET); + lw_add_widget_value_arg (wv, XmNfontList, (XtArgVal)fontList); #endif - lw_add_widget_value_arg - (wv, XtNfont, (XtArgVal)FONT_INSTANCE_X_FONT - (XFONT_INSTANCE (query_string_font - (IMAGE_INSTANCE_WIDGET_TEXT (ii), - IMAGE_INSTANCE_WIDGET_FACE (ii), - domain)))); + /* #### This is an unacceptable hack. */ + if (fs) + lw_add_widget_value_arg (wv, XtNfont, (XtArgVal) fs); + else + warn_when_safe_lispobj + (intern ("xft"), Qwarning, + Fcons (build_string ("missing font in update_widget_face"), + Fface_name (face))); + } wv->change = VISIBLE_CHANGE; /* #### Megahack - but its just getting too complicated to do this in the right place. */ Index: src/objects-x.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/objects-x.c,v retrieving revision 1.26.2.3 diff -u -r1.26.2.3 objects-x.c --- src/objects-x.c 3 Dec 2004 07:41:10 -0000 1.26.2.3 +++ src/objects-x.c 10 Dec 2004 03:35:24 -0000 ＠＠ -206,16 +210,18 ＠＠ Display *dpy = DEVICE_X_DISPLAY (XDEVICE (device)); Extbyte *extname; XFontStruct *fs = NULL; /* _F_ont _S_truct */ - /* always initialize; then constant folding should optimize away anything - that depends on rf != NULL, and we can eliminate most of the #ifdefs */ +#ifdef USE_XFT XftFont *rf = NULL; /* _R_ender _F_ont (X Render extension) */ +#else +#define rf (0) +#endif LISP_STRING_TO_EXTERNAL (f->name, extname, Qx_font_name_encoding); +#ifdef USE_XFT if (debug_xft > 0) stderr_out ("attempting to initialize font spec %s\n", extname); /* #### serialize (optimize) these later... */ -#ifdef USE_XFT /* #### This function really needs to go away. The problem is that the fontconfig/Xft functions work much too hard to ensure that something is returned; but that something need not be ＠＠ -229,8 +235,6 ＠＠ /* #### should this refer to X and/or Xft? */ maybe_signal_error (Qgui_error, "Couldn't load font", f->name, Qfont, errb); - if (debug_xft > 0) - stderr_out ("failed to initialize font spec %s\n", extname); return 0; } ＠＠ -250,13 +254,11 ＠＠ fear that the finalize method may get fucked. */ f->data = xnew (struct x_font_instance_data); - /* handle Xft fonts */ +#ifdef USE_XFT + FONT_INSTANCE_X_XFTFONT (f) = rf; if (rf) + /* Have an Xft font, initialize font info from it. */ { - if (fs) - XFreeFont (dpy, fs); - FONT_INSTANCE_X_XFTFONT (f) = rf; - FONT_INSTANCE_X_FONT (f) = NULL; if (f->ascent < rf->ascent) f->ascent = rf->ascent; if (f->descent < rf->descent) ＠＠ -271,9 +273,16 ＠＠ if (debug_xft > 0) stderr_out ("initialized Xft font %s\n", extname); } - - else if (fs) - /* Handle a core font. */ + else + { + if (debug_xft > 0) + stderr_out ("couldn't initialize Xft font %s\n", extname); + } +#endif + + FONT_INSTANCE_X_FONT (f) = fs; + if (fs && !rf) /* fs && rf is unlikely, but play it safe */ + /* Have to use a core font, initialize font info from it. */ { FONT_INSTANCE_X_FONT (f) = fs; FONT_INSTANCE_X_XFTFONT (f) = NULL; ＠＠ -335,15 +344,20 ＠＠ f->proportional_p = (fs->min_bounds.width != fs->max_bounds.width || (x_handle_non_fully_specified_fonts && !fs->all_chars_exist)); - if (debug_xft > 0) - stderr_out ("initialized Xft font %s\n", extname); } - /* #### check for weirdness */ - if (f->height < f->width) - if (debug_xft > 0) - stderr_out ("width of font %s is %d, larger than its height %d\n", - extname, f->width, f->height); +#ifdef USE_XFT + if (debug_xft > 0) + { + /* #### check for weirdness */ + if (f->height < f->width) + stderr_out ("width of font %s is %d, larger than its height %d\n", + extname, f->width, f->height); + stderr_out ("initialized Xft font %s\n", extname); + } +#else +#undef rf +#endif return 1; } ＠＠ -368,10 +382,13 ＠＠ x_finalize_font_instance (Lisp_Font_Instance *f) { +#ifdef USE_XFT if (debug_xft > 0) stderr_out ("finalizing %s", (STRINGP (f->name) ? (char *) XSTRING_DATA (f->name) : "no font name")); +#endif + if (f->data) { if (DEVICE_LIVE_P (XDEVICE (f->device))) ＠＠ -686,19 +703,19 ＠＠ { struct device *d = XDEVICE (f->device); Display *dpy = DEVICE_X_DISPLAY (d); -#ifndef USE_XFT Extbyte *nameext; char* xlfd; -#else - FcChar8* xlfd; +#ifdef USE_XFT + FcChar8* fcfont; #endif +#ifdef USE_XFT + /* First, try an Xft font. */ if (NILP (FONT_INSTANCE_TRUENAME (f))) { -#ifdef USE_XFT int core; FcPattern *pat = FONT_INSTANCE_X_XFTFONT (f)->pattern; - FcResult res_xlfd = FcPatternGetString (pat, XFT_XLFD, 0, &xlfd); + FcResult res_xlfd = FcPatternGetString (pat, XFT_XLFD, 0, &fcfont); FcResult res_core = FcPatternGetBool (pat, XFT_CORE, 0, &core); if (res_xlfd == FcResultTypeMismatch || res_core == FcResultTypeMismatch) ＠＠ -730,19 +747,25 ＠＠ /* #### Why isn't this appropriate for core fonts, too? */ FONT_INSTANCE_TRUENAME (f) = build_ext_string (truename_via_XListFonts (dpy, - (Extbyte *) &xlfd[0]), + (Extbyte *) &fcfont[0]), Qx_font_name_encoding); } -#else + } +#endif + + /* OK, fall back to core font. */ + if (NILP (FONT_INSTANCE_TRUENAME (f))) + { nameext = &xlfd[0]; LISP_STRING_TO_EXTERNAL (f->name, nameext, Qx_font_name_encoding); FONT_INSTANCE_TRUENAME (f) = x_font_truename (dpy, nameext, FONT_INSTANCE_X_FONT (f)); -#endif } + if (NILP (FONT_INSTANCE_TRUENAME (f))) { + /* Urk, no luck. Whine about our bad luck and exit. */ Lisp_Object font_instance = wrap_font_instance (f); ＠＠ -752,6 +775,8 ＠＠ (This is only used by Fequal() right now.) */ return f->name; } + + /* Return what we found. */ return FONT_INSTANCE_TRUENAME (f); } ＠＠ -764,7 +789,8 ＠＠ Display *dpy = DEVICE_X_DISPLAY (d); XFontProp *props = NULL; - /* #### really should hack Xft fonts, too */ + /* #### really should hack Xft fonts, too + Strategy: fontconfig must have an iterator for this purpose. */ if (! FONT_INSTANCE_X_FONT (f)) return result; props = FONT_INSTANCE_X_FONT (f)->properties; ＠＠ -847,7 +873,8 ＠＠ #ifdef MULE static int -x_font_spec_matches_charset (struct device *d, Lisp_Object charset, +x_font_spec_matches_charset (struct device * USED_IF_XFT (d), + Lisp_Object charset, const Ibyte *nonreloc, Lisp_Object reloc, Bytecount offset, Bytecount length, int stage) ＠＠ -922,6 +949,7 ＠＠ ERROR_ME, 0) >= 0); } +#ifdef USE_XFT /* #### debug functions: find a better place for us */ const char *FcResultToString (FcResult r); const char * ＠＠ -976,6 +1004,7 ＠＠ return buffer; } } +#endif /* find a font spec that matches font spec FONT and also matches (the registry of) CHARSET. */ ＠＠ -1002,10 +1031,10 ＠＠ /* #### does Xft permit/require a different encoding? */ LISP_STRING_TO_EXTERNAL (font, patternext, Qx_font_name_encoding); - if (debug_xft > 0) - stderr_out ("confirming charset for font instance %s\n", patternext); #ifdef USE_XFT + if (debug_xft > 0) + stderr_out ("confirming charset for font instance %s\n", patternext); /* #### totally lazy, let's just get something out the door */ waste = FcInit (); if (waste) /* I don't think this can fail */ ＠＠ -1090,14 +1119,15 ＠＠ } else if (debug_xft > 0) stderr_out ("Failed fontconfig initialization\n"); + + if (debug_xft > 0) + stderr_out ("shit happens, try X11 charset match for %s\n", patternext); #endif names = XListFonts (DEVICE_X_DISPLAY (XDEVICE (device)), patternext, MAX_FONT_COUNT, &count); /* #### This code seems awfully bogus -- mrb */ /* #### fontconfig does it better -- sjt */ - if (debug_xft > 0) - stderr_out ("shit happens, try X11 charset match for %s\n", patternext); for (i = 0; i < count; i ++) { const Ibyte *intname; ＠＠ -1152,17 +1182,10 ＠＠ CONSOLE_HAS_METHOD (x, valid_color_name_p); CONSOLE_HAS_METHOD (x, initialize_font_instance); - /* #### both of these #ifndef USE_XFT are pretty bogus - they cause "defined but not used warnings" - one or the other must use x_font_truename, which also warns */ -#ifndef USE_XFT CONSOLE_HAS_METHOD (x, print_font_instance); -#endif CONSOLE_HAS_METHOD (x, finalize_font_instance); CONSOLE_HAS_METHOD (x, font_instance_truename); -#ifndef USE_XFT CONSOLE_HAS_METHOD (x, font_instance_properties); -#endif CONSOLE_HAS_METHOD (x, list_fonts); #ifdef MULE CONSOLE_HAS_METHOD (x, find_charset_font); Index: src/redisplay-x.c =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/src/redisplay-x.c,v retrieving revision 1.39.2.1 diff -u -r1.39.2.1 redisplay-x.c --- src/redisplay-x.c 25 Nov 2004 12:44:16 -0000 1.39.2.1 +++ src/redisplay-x.c 10 Dec 2004 03:35:24 -0000 ＠＠ -299,7 +299,8 ＠＠ /****************************************************************************/ static int -x_text_width_single_run (struct frame *f, struct face_cachel *cachel, struct textual_run *run) +x_text_width_single_run (struct frame * USED_IF_XFT (f), + struct face_cachel *cachel, struct textual_run *run) { Lisp_Object font_inst = FACE_CACHEL_FONT (cachel, run->charset); Lisp_Font_Instance *fi = XFONT_INSTANCE (font_ins