Hi, After a fresh checkout, I got the following error: xemacs$ ./configure ./configure: line 2343: syntax error near unexpected token `)' ./configure: line 2343: ` , )' I think it comes from the usage of XE_COMPLEX_ARG for xft options. -- Jérôme Marant

Mac OSX 10.4.3 system with gcc-4.01 Darwin 8.3.0 Darwin Kernel Version 8.3.0: Mon Oct 3 20:04:04 PDT 2005; root:xnu-792.6.22.obj~2/RELEASE_PPC Power Macintosh powerpc XEmacs 21.5-b23 "daikon" (+CVS-20051125) configured for `powerpc-apple-darwin8.3.0'. Note, I removed the ", )" at line 2343 of configure just to get the configure to work. Compiling with --with-xft=none still works. Compiling with --with-xft=all, --with-xft=emacs, or without xft (i.e. no --with-xft in configure command) fails. If I compile configured --enable-union-type, I get a number of warnings early on about pointers differing in signedness (in Xft-related files) and then a failure with an error in src/objects-x.c (note the same warnings at first). If I compile without the union-type for lisp objects, I do not get the warnings, but I get the same error (note that the initiallizations that do not produce a warning have NULL as the value for the stuct members language and rfc3066). The variable 'name' is not declared. It is declared at line 1248 of objects-x.c (as a FcChar8 type), but note that the scope for that variable ends at line 1287, before the variable is referenced. name is freed immediately before the end of that block, so moving the closing bracket seems not the correct solution. I cannot figure why it would compile for any other compiler/system. gcc -c -Wall -Wno-switch -Wundef -Wsign-compare -Wno-char-subscripts -Wpacked -Wpointer-arith -Wunused-parameter -g -Os -pipe -mcpu=7450 -mtune=7450 -mpim-altivec -Os -pipe -mcpu=7450 -mtune=7450 -mpim-altivec -Demacs -I. -I/Users/royar/src/xemacs/src -DHAVE_CONFIG_H -I/usr/X11R6/include/freetype2 -I/opt/local/include -I/usr/X11R6/include objects-x.c objects-x.c: In function 'x_initialize_font_instance': objects-x.c:349: warning: pointer targets in passing argument 3 of 'XftTextExtents8' differ in signedness objects-x.c: In function 'x_font_instance_truename': objects-x.c:810: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness objects-x.c: At top level: objects-x.c:1127: warning: pointer targets in initialization differ in signedness objects-x.c:1128: warning: pointer targets in initialization differ in signedness objects-x.c:1129: warning: pointer targets in initialization differ in signedness objects-x.c:1130: warning: pointer targets in initialization differ in signedness objects-x.c:1131: warning: pointer targets in initialization differ in signedness objects-x.c:1132: warning: pointer targets in initialization differ in signedness objects-x.c:1133: warning: pointer targets in initialization differ in signedness objects-x.c:1136: warning: pointer targets in initialization differ in signedness objects-x.c:1137: warning: pointer targets in initialization differ in signedness objects-x.c:1138: warning: pointer targets in initialization differ in signedness objects-x.c:1139: warning: pointer targets in initialization differ in signedness objects-x.c:1148: warning: pointer targets in initialization differ in signedness objects-x.c:1151: warning: pointer targets in initialization differ in signedness objects-x.c:1152: warning: pointer targets in initialization differ in signedness objects-x.c: In function 'x_find_charset_font': objects-x.c:1213: warning: pointer targets in initialization differ in signedness objects-x.c:1225: warning: pointer targets in passing argument 1 of 'FcNameParse' differ in signedness objects-x.c:1226: warning: pointer targets in initialization differ in signedness objects-x.c:1230: warning: pointer targets in initialization differ in signedness objects-x.c:1232: warning: pointer targets in initialization differ in signedness objects-x.c:1236: warning: pointer targets in initialization differ in signedness objects-x.c:1252: warning: pointer targets in initialization differ in signedness objects-x.c:1269: warning: pointer targets in initialization differ in signedness objects-x.c:1283: warning: pointer targets in initialization differ in signedness objects-x.c:1307: error: 'name' undeclared (first use in this function) objects-x.c:1307: error: (Each undeclared identifier is reported only once objects-x.c:1307: error: for each function it appears in.) objects-x.c:1307: error: parse error before '}' token objects-x.c:1353: error: parse error before '}' token objects-x.c:1367: error: parse error before '}' token objects-x.c:1399: error: parse error before '}' token objects-x.c:1413: error: parse error before '}' token make[1]: *** [objects-x.o] Error 1 make: *** [src] Error 2 -- Dr. Robert Delius Royar Associate Professor of English Morehead State University Morehead, Kentucky Making meaning one message at a time.

APPROVE COMMIT 21.5 robert> Compiling with --with-xft=all, --with-xft=emacs, or robert> without xft (i.e. no --with-xft in configure command) robert> fails. Thanks for that detail! I just overlooked the defaults; fixed. robert> If I compile configured --enable-union-type, I get a robert> number of warnings early on about pointers differing in robert> signedness That may be an issue of GCC 4.0, I guess. I don't see it on Mac OS X 10.3.9 with GCC 3.3. I'll see what happens with GCC 4 tomorrow or so. All the rest of the bugs were related to getting workspaces confused while trying to fix conflicts, fixed in a separate patch. Index: ChangeLog =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/ChangeLog,v retrieving revision 1.475 diff -u -U0 -r1.475 ChangeLog --- ChangeLog 26 Nov 2005 18:24:51 -0000 1.475 +++ ChangeLog 26 Nov 2005 18:35:48 -0000 ＠＠ -0,0 +1,5 ＠＠ +2005-11-27 Stephen J. Turnbull <stephen(a)xemacs.org&gt; + + * configure.ac (XE_COMPLEX_ARG[xft]): Default everything off; + change doc. + Index: configure.ac =================================================================== RCS file: /pack/xemacscvs/XEmacs/xemacs/configure.ac,v retrieving revision 1.24 diff -u -r1.24 configure.ac --- configure.ac 26 Nov 2005 18:24:54 -0000 1.24 +++ configure.ac 26 Nov 2005 18:35:56 -0000 ＠＠ -637,12 +637,12 ＠＠ [Xft], [`emacs' (buffers), `menubars', `tabs', and `gauges'], [X11, Xft, Xrender, freetype, and fontconfig], - [`emacs,menubars,tabs,gauges'])], + [`noemacs, nomenubars, notabs, nogauges'])], [],[], - [XE_COMPLEX_OPTION([emacs],[yes]), - XE_COMPLEX_OPTION([menubars],[yes]), - XE_COMPLEX_OPTION([tabs],[yes]), - XE_COMPLEX_OPTION([gauges],[yes])]) + [XE_COMPLEX_OPTION([emacs],[no]), + XE_COMPLEX_OPTION([menubars],[no]), + XE_COMPLEX_OPTION([tabs],[no]), + XE_COMPLEX_OPTION([gauges],[no])]) XE_MERGED_ARG([gtk], AC_HELP_STRING([--with-gtk],[Support GTK on the X Window System. (EXPERIMENTAL)]), [true], [with_gtk=no]) -- School of Systems and Information Engineering 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.

jmarant> It is not enough. configure --with-xft still fails (well, jmarant> every option using XE_COMPLEX_ARG does). I didn't try to fix that, and there's no extreme rush. My first take is that I don't want "yes" to be equivalent to "all", so I'm going to let Malcolm take a look at it. -- School of Systems and Information Engineering 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.

jmarant> OK, but how should a bare "--with-xft" act? It should give you the defaults. If that is "emacs,tabs,nomenubars,nogauges" it's not the same as "all". Malcolm, do you disagree with that? I agree that "--with-xft" should produce the same effect as "--with-xft=yes"; I think that both should be different from "--with-xft=all". I'll take a look at the code and see what actually happens. -- School of Systems and Information Engineering 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.

jmarant> OK, but how should a bare "--with-xft" act? Stephen> It should give you the defaults. If that is Stephen> "emacs,tabs,nomenubars,nogauges" it's not the same as "all". Stephen> Malcolm, do you disagree with that? On reflection that should be the expected behaviour but it looks like a recent change I made to the complex option support broke it. I'll organise a fix soon. Malcolm -- Malcolm Purvis <malcolmp(a)xemacs.org&gt;