Jerry James writes: Let's autoconfiscate it, for reasons given in my other post. _______________________________________________ XEmacs-Patches mailing list XEmacs-Patches(a)xemacs.org http://lists.xemacs.org/mailman/listinfo/xemacs-patches

Ar an triú lá déag de mí Méan Fómhair, scríobh Jeff Sparkes: This is still broken. It checks HAVE_XKBKEYCODETOKEYSYM at compile time, and then tries to call XkbKeycodeToKeysym at run time, when XkbKeycodeToKeysym may well not be there. This is exactly the situation on Apple’s X11, the XKB library is available at compile time, and as I mentioned, the XKB extension is not available on the server at runtime, so the below code fails. There are other servers in use where this is true, see https://www.google.com/search?q=%22xkb+extension+not+present%22 . If you *really* want to use XkbKeycodeToKeysym, keep around a function pointer in the XEmacs device object, initialise it to XkbKeycodeToKeysym on connection if the XKB extension is available, XKeycodeToKeysym if it isn’t, and call through the function pointer every time. This will still trigger a compile-time warning because you’re referencing the XKeycodeToKeysym symbol, you can use dlsym() to trick this away, but I don’t recommend this. As I say, there is no problem with our current code, it is the warning that is incorrect. --- a/src/event-Xt.c +++ b/src/event-Xt.c ＠＠ -382,7 +386,11 ＠＠ for (column = 0; column < 4; column += 2) { KeyCode code = xd->x_modifier_keymap->modifiermap[modifier_index * mkpm + modifier_key]; +#ifdef HAVE_XKBKEYCODETOKEYSYM + KeySym sym = (code ? XkbKeycodeToKeysym (display, code, column, 0) : 0); +#else KeySym sym = (code ? XKeycodeToKeysym (display, code, column) : 0); +#endif if (sym == last_sym) continue; last_sym = sym; switch (sym) { -- ‘Liston operated so fast that he once accidentally amputated an assistant’s fingers along with a patient’s leg, […] The patient and the assistant both died of sepsis, and a spectator reportedly died of shock, resulting in the only known procedure with a 300% mortality.’ (Atul Gawande, NEJM, 2012) _______________________________________________ XEmacs-Patches mailing list XEmacs-Patches(a)xemacs.org http://lists.xemacs.org/mailman/listinfo/xemacs-patches

Ar an naoiú lá de mí Méan Fómhair, scríobh Jeff Sparkes: We shouldn’t make this change. The XKB *extension* is optional, and of the common X11 implementations these days the OS X (XQuartz) implementation of X11 has it turned off by default, and I believe it is broken when you turn it on. (Am honeymooning, don’t have access to my own Mac to experiment too much, but my lady wife’s MacBook Pro doesn’t have the XKB extension in the output of xdpyinfo.) It’s the compiler warning that is broken, and we should complain to x.org about it. Our code is fine. -- ‘Liston operated so fast that he once accidentally amputated an assistant’s fingers along with a patient’s leg, […] The patient and the assistant both died of sepsis, and a spectator reportedly died of shock, resulting in the only known procedure with a 300% mortality.’ (Atul Gawande, NEJM, 2012) _______________________________________________ XEmacs-Patches mailing list XEmacs-Patches(a)xemacs.org http://lists.xemacs.org/mailman/listinfo/xemacs-patches