[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Denis Barbier: THANK YOU



On Sun, Apr 18, 2004 at 01:43:51PM -0500, Branden Robinson wrote:
> Hi Denis,
> 
> Fabio and I were talking on and IRC and we were both very impressed with
> your hard work on scrubbing out old bugs, especially the XKB problems.
> 
> So please accept this Steel-Toed Boot award for excellence in BTS
> Ass-Kicking.
> 
>     -------
>     |     |
>     |     |
>     |     |
>     |     /
>     |    |
>     |    |
>     |    \
>    /|     -----^\
>   * |           |
>     |___|\______/

Arf, thanks :)
BTW several xlibs bugreports should be reassigned, but I do not know
which packages are the culprits.

Here are some thoughts about remaining bugs:
 * Macintosh layouts
      #234113 #67812 #111905 #121297 #157170 #166478
   I have no access to Apple keyboards, so I won't try to fix them,
   but will be glad to help if some people from debian-powerpc are
   interested.  Upstream now ships symbols/apple which might fixes
   some of these bugs.
 * Hungarian layouts
      #33154 #76439 #86539
   This layout is special, its behavior under hu_HU locale is different
   from other locales.  Fixing these bugs certainly require some help
   from Hungarian folks.
 * altwin:left_meta_win and altwin:meta_win options
      #234081 #241034 #164767
   Explanations on these bugs have been sent to
     http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=241034&msg=18
   Fixing altwin:left_meta_win is not trivial, but please apply the
   patch to fix altwin:meta_win, you could then tell submitters
   to use altwin:meta_win instead of altwin:left_meta_win and close
   these bugs.  (In da.po, "altwin: meta_win" has a spurious space)

One problem with latest XKB bugreports (changes in X 4.3) is that we do
not know if xlibs has to be blamed, or if other apps are modifying XKB
settings.  In order to reproduce the reported bugs, I slightly hacked
setxkbmap.c to ignore current server settings.  If a -noserver
command-line flag is added (patch attached), we could ask submitters to
run a setxkbmap command and determine if this bug belowgs to xlibs.

Denis
Index: setxkbmap.c
===================================================================
RCS file: /cvs/xc/programs/setxkbmap/setxkbmap.c,v
retrieving revision 3.8
diff -u -r3.8 setxkbmap.c
--- setxkbmap.c	17 Nov 2003 22:20:50 -0000	3.8
+++ setxkbmap.c	18 Apr 2004 19:59:22 -0000
@@ -83,6 +83,7 @@
 /***====================================================================***/
 Bool			print= False;
 Bool			synch= False;
+Bool			noserver= False;
 int			verbose= 5;
 
 Display *		dpy;
@@ -210,6 +211,7 @@
     MSG("-keymap <name>      Specifies name of keymap to load\n");
     MSG("-layout <name>      Specifies layout used to choose component names\n");
     MSG("-model <name>       Specifies model used to choose component names\n");
+    MSG("-noserver           Ignores current server settings\n");
     MSG("-option <name>      Adds an option used to choose component names\n");
     MSG("-print              Print a complete xkb_keymap description and exit\n");
     MSG("-rules <name>       Name of rules file to use\n");
@@ -370,6 +372,8 @@
 	    ok= setOptString(&i,argc,argv,SYMBOLS_NDX,FROM_CMD_LINE);
 	else if (streq(argv[i],"-synch"))
 	    synch= True;
+	else if (streq(argv[i],"-noserver"))
+	    noserver= True;
 	else if (streq(argv[i],"-types"))
 	    ok= setOptString(&i,argc,argv,TYPES_NDX,FROM_CMD_LINE);
 	else if (streq(argv[i],"-verbose")||(streq(argv[i],"-v"))) {
@@ -835,9 +839,9 @@
     if ((!parseArgs(argc,argv))||(!getDisplay(argc,argv)))
 	exit(-1);
     svValue[LOCALE_NDX]= setlocale(LC_ALL,svValue[LOCALE_NDX]);
-    svSrc[LOCALE_NDX]= FROM_SERVER;
+    svSrc[LOCALE_NDX]= (noserver?FROM_CMD_LINE:FROM_SERVER);
     VMSG1(7,"locale is %s\n",svValue[LOCALE_NDX]);
-    if (dpy)
+    if (dpy && !noserver)
         getServerValues();
     if (svValue[CONFIG_NDX] && (!applyConfig(svValue[CONFIG_NDX])))
 	exit(-3);

Reply to: