Re: XEmacs/GTK does menus...
25 years, 5 months
William M. Perry
William M. Perry
Michael Harnois <mdharnois(a)home.com> writes:
> So, is this at a point where we should send you lisp and/or C
> backtraces when things upshit and die?
Definitely.
-bp
- 1 participants
- 0 comments
Re: XEmacs/GTK does menus...
25 years, 5 months
William M. Perry
William M. Perry
karlheg(a)bittersweet.inetarena.com (Karl M. Hegbloom) writes:
> Are you using `glade' to build the interface, Bill?
Nope, but I plan on allowing you to call libglade from elisp, eventually.
The generic representation of the UI that I plan on working on next week
will be really close to a parsed glade file.
> `apt-get install glade'
:) That doesn't work on FreeBSD
-bp
- 1 participants
- 0 comments
fatal error handling
25 years, 5 months
Ben Wing
Ben Wing
jan, this should supersede your changes.
? fixup-427-5.diff
? lib-src/etags.pdb
? lib-src/hexl.pdb
? lib-src/i.pdb
? lib-src/make-docfile.pdb
? lib-src/mmencode.pdb
? lib-src/movemail.pdb
? lib-src/run.res
? lib-src/sorted-doc.pdb
? lib-src/vc50.pdb
? lib-src/wakeup.pdb
? lisp/process2.el
? nt/obj
? src/NEEDTODUMP
? src/runxemacs.pdb
? src/temacs.bsc
? src/temacs.map
? src/temacs.pdb
? src/xemacs.opt
? src/Xpm.dll
cvs server: Diffing .
cvs server: Diffing dynodump
cvs server: Diffing dynodump/i386
cvs server: Diffing dynodump/ppc
cvs server: Diffing dynodump/sparc
cvs server: Diffing etc
cvs server: Diffing etc/custom
cvs server: Diffing etc/custom/example-themes
cvs server: Diffing etc/eos
cvs server: Diffing etc/idd
cvs server: Diffing etc/photos
cvs server: Diffing etc/sparcworks
cvs server: Diffing etc/tests
cvs server: Diffing etc/tests/external-widget
cvs server: Diffing etc/toolbar
cvs server: Diffing info
cvs server: Diffing lib-src
cvs server: Diffing lisp
cvs server: Diffing lisp/mule
cvs server: Diffing lisp/term
cvs server: Diffing lock
cvs server: Diffing lwlib
cvs server: Diffing man
cvs server: Diffing man/internals
cvs server: Diffing man/lispref
cvs server: Diffing man/new-users-guide
cvs server: Diffing man/xemacs
cvs server: Diffing modules
cvs server: Diffing modules/base64
cvs server: Diffing modules/ldap
cvs server: Diffing modules/sample
cvs server: Diffing modules/zlib
cvs server: Diffing nt
cvs server: Diffing nt/inc
cvs server: Diffing nt/inc/arpa
cvs server: Diffing nt/inc/netinet
cvs server: Diffing nt/inc/sys
cvs server: Diffing nt/installer
cvs server: Diffing nt/installer/Wise
cvs server: Diffing src
Index: src/ChangeLog
===================================================================
RCS file: /usr/CVSroot/XEmacs/xemacs/src/ChangeLog,v
retrieving revision 1.154.2.589
diff -u -r1.154.2.589 ChangeLog
--- ChangeLog 2000/05/11 10:29:32 1.154.2.589
+++ ChangeLog 2000/05/12 09:03:44
＠＠ -1,3 +1,34 ＠＠
+2000-05-12 Ben Wing <ben(a)xemacs.org>
+
+ * emacs.c:
+ * emacs.c (ensure_no_quitting_from_now_on):
+ * emacs.c (fatal_error_signal):
+ * emacs.c (mswindows_handle_hardware_exceptions):
+ * emacs.c (main):
+ * emacs.c (Fkill_emacs):
+ * emacs.c (shut_down_emacs):
+ * emacs.c (assert_failed):
+ various improvements in fatal error handling.
+
+ * eval.c:
+ move preparing_for_armageddon to emacs.c.
+
+ * lisp.h:
+ declare fatal_error_in_progress.
+
+ * print.c:
+ * print.c (std_handle_out_external):
+ * print.c (std_handle_out_va):
+ * print.c (stderr_out):
+ * print.c (stdout_out):
+ use console under mswin when no standard output.
+ don't do code conversion during fatal error.
+
+ * scrollbar.c (Fscrollbar_page_up):
+ * scrollbar.c (Fscrollbar_page_down):
+ fix missing else. reindent.
+
+
2000-05-11 Ben Wing <ben(a)xemacs.org>
* cmdloop.c (Freally_early_error_handler):
Index: src/emacs.c
===================================================================
RCS file: /usr/CVSroot/XEmacs/xemacs/src/emacs.c,v
retrieving revision 1.82.2.63
diff -u -r1.82.2.63 emacs.c
--- emacs.c 2000/05/11 05:40:14 1.82.2.63
+++ emacs.c 2000/05/12 09:03:45
＠＠ -2,6 +2,7 ＠＠
Copyright (C) 1985, 1986, 1987, 1992, 1993, 1994
Free Software Foundation, Inc.
Copyright (C) 1995 Sun Microsystems, Inc.
+ Copyright (C) 2000 Ben Wing.
This file is part of XEmacs.
＠＠ -305,8 +306,10 ＠＠
priority. */
int emacs_priority;
-/* If non-zero a filter or a sentinel is running. Tested to save the match
- data on the first attempt to change it inside asynchronous code. */
+/* Some FSF junk with running_asynch_code, to preserve the match
+ data. Not necessary because we don't call process filters
+ asynchronously (i.e. from within QUIT). */
+/* #### Delete this when merging the rest of my code */
int running_asynch_code;
/* If non-zero, a window-system was specified on the command line. */
＠＠ -376,12 +379,64 ＠＠
extern Lisp_Object Vlisp_EXEC_SUFFIXES;
-/* Signal code for the fatal signal that was received */
-static int fatal_error_code;
-/* Nonzero if handling a fatal error already */
-static int fatal_error_in_progress;
+/* Ben's capsule summary about expected and unexpected exits from XEmacs.
+ Expected exits occur when the user directs XEmacs to exit, for example
+ by pressing the close button on the only frame in XEmacs, or by typing
+ C-x C-c. This runs `save-buffers-kill-emacs', which saves any necessary
+ buffers, and then exits using the primitive `kill-emacs'.
+
+ However, unexpected exits occur in a few different ways:
+
+ -- a memory access violation or other hardware-generated exception
+ occurs. This is the worst possible problem to deal with, because
+ the fault can occur while XEmacs is in any state whatsoever, even
+ quite unstable ones. As a result, we need to be *extremely* careful
+ what we do.
+ -- we are using one X display (or if we've used more, we've closed the
+ others already), and some hardware or other problem happens and
+ suddenly we've lost our connection to the display. In this situation,
+ things are not so dire as in the last one; our code itself isn't
+ trashed, so we can continue execution as normal, after having set
+ things up so that we can exit at the appropriate time. Our exit
+ still needs to be of the emergency nature; we have no displays, so
+ any attempts to use them will fail. We simply want to auto-save
+ (the single most important thing to do during shut-down), do minimal
+ cleanup of stuff that has an independent existence outside of XEmacs,
+ and exit.
+
+ Currently, both unexpected exit scenarios described above set
+ preparing_for_armageddon to indicate that nonessential and possibly
+ dangerous things should not be done, specifically:
+
+ -- no garbage collection.
+ -- no hooks are run.
+ -- no messages of any sort from autosaving.
+ -- autosaving tries harder, ignoring certain failures.
+ -- existing frames are not deleted.
+
+ (Also, all places that set preparing_for_armageddon also
+ set dont_check_for_quit. This happens separately because it's
+ also necessary to set other variables to make absolutely sure
+ no quitting happens.)
+
+ In the first scenario above (the access violation), we also set
+ fatal_error_in_progress. This causes more things to not happen:
+
+ -- assertion failures do not abort.
+ -- printing code does not do code conversion or gettext when
+ printing to stdout/stderr.
+*/
+
+/* Nonzero if handling a fatal error already. */
+int fatal_error_in_progress;
+
+/* Non-nil means we're going down, so we better not run any hooks
+ or do other non-essential stuff. */
+int preparing_for_armageddon;
+
+
static JMP_BUF run_temacs_catch;
static int run_temacs_argc;
＠＠ -390,23 +445,54 ＠＠
static size_t run_temacs_argv_size;
static size_t run_temacs_args_size;
-static void shut_down_emacs (int sig, Lisp_Object stuff);
+static void shut_down_emacs (int sig, Lisp_Object stuff, int no_auto_save);
+static void
+ensure_no_quitting_from_now_on (void)
+{
+ /* make sure no quitting from now on!! */
+ dont_check_for_quit = 1;
+ Vinhibit_quit = Qt;
+ Vquit_flag = Qnil;
+}
+
/* Handle bus errors, illegal instruction, etc. */
SIGTYPE
fatal_error_signal (int sig)
{
- fatal_error_code = sig;
- signal (sig, SIG_DFL);
+ fatal_error_in_progress++;
+ preparing_for_armageddon = 1;
+
+ ensure_no_quitting_from_now_on ();
+
/* Unblock the signal so that if the same signal gets sent in the
code below, we avoid a deadlock. */
- EMACS_UNBLOCK_SIGNAL (fatal_error_code);
+ EMACS_UNBLOCK_SIGNAL (sig);
+
+ /* Only try auto-saving first time through. If we crash in auto-saving,
+ don't do it again. */
+ if (fatal_error_in_progress == 1)
+ {
+ Fdo_auto_save (Qt, Qnil); /* do this before anything hazardous */
+ /* Do this so that the variable has the same value of 2 regardless of
+ whether we made it through auto-saving correctly. */
+ fatal_error_in_progress++;
+ }
+ else if (fatal_error_in_progress == 2)
+ stderr_out ("WARNING: Unable to auto-save your files properly.\n"
+ "Some or all may in fact have been auto-saved.\n"
+ "\n");
+
+ /* Now, reset our signal handler, so the next time, we just die.
+ Don't do this before auto-saving. */
+ signal (sig, SIG_DFL);
+ /* Keep in mind that there's more than one signal that we can crash
+ on. */
/* If fatal error occurs in code below, avoid infinite recursion. */
- if (! fatal_error_in_progress)
+ if (fatal_error_in_progress <= 2)
{
- fatal_error_in_progress = dont_check_for_quit = 1;
- shut_down_emacs (sig, Qnil);
+ shut_down_emacs (sig, Qnil, 1);
stderr_out ("\nLisp backtrace follows:\n\n");
Fbacktrace (Qexternal_debugging_output, Qt);
# if 0 /* This is evil, rarely useful, and causes grief in some cases. */
＠＠ -425,9 +511,62 ＠＠
# endif
}
/* Signal the same code; this time it will really be fatal. */
- kill (getpid (), fatal_error_code);
+ kill (getpid (), sig);
SIGRETURN;
}
+
+#ifdef _MSC_VER
+
+static DWORD
+mswindows_handle_hardware_exceptions (DWORD code)
+{
+ if (code != STATUS_ACCESS_VIOLATION && code != STATUS_ILLEGAL_INSTRUCTION
+ && code != STATUS_PRIVILEGED_INSTRUCTION
+ && code != STATUS_DATATYPE_MISALIGNMENT)
+ return EXCEPTION_CONTINUE_SEARCH;
+
+ /* I don't know if this filter is still wrapped in the outer __try, but
+ it doesn't hurt to have another one. --ben */
+ __try
+ {
+ fatal_error_in_progress++;
+ preparing_for_armageddon = 1;
+
+ ensure_no_quitting_from_now_on ();
+
+ /* Only try auto-saving first time through. If we crash in auto-saving,
+ don't do it again. */
+ if (fatal_error_in_progress == 1)
+ {
+ Fdo_auto_save (Qt, Qnil); /* do this before anything hazardous */
+ /* Do this so that the variable has the same value of 2 regardless of
+ whether we made it through auto-saving correctly. */
+ fatal_error_in_progress++;
+ }
+ else if (fatal_error_in_progress == 2)
+ stderr_out ("WARNING: Unable to auto-save your files properly.\n"
+ "Some or all may in fact have been auto-saved.\n"
+ "\n");
+
+ /* If fatal error occurs in code below, avoid infinite recursion. */
+ if (fatal_error_in_progress <= 2)
+ {
+ shut_down_emacs (-1, Qnil, 1);
+ stderr_out ("\nLisp backtrace follows:\n\n");
+ Fbacktrace (Qexternal_debugging_output, Qt);
+ }
+ }
+ /* VC++ documentation says that
+ GetExceptionCode() cannot be called inside the filter itself. */
+ __except (mswindows_handle_hardware_exceptions (GetExceptionCode ())) {}
+
+ /* pretend we didn't handle this, so that the debugger is invoked and/or
+ the normal GPF box appears. */
+ return EXCEPTION_CONTINUE_SEARCH;
+}
+
+#endif /* _MSC_VER */
+
#ifdef SIGDANGER
＠＠ -2308,6 +2447,16 ＠＠
int
main (int argc, char **argv, char **envp)
{
+
+#ifdef _MSC_VER
+ /* Under VC++, access violations and the like are not sent through
+ the standard signal() mechanism. Rather, they need to be handled
+ using the Microsoft "structured exception handling" mechanism,
+ which vaguely resembles the C++ mechanisms. */
+ __try
+ {
+#endif
+
int volatile vol_argc = argc;
char ** volatile vol_argv = argv;
char ** volatile vol_envp = envp;
＠＠ -2425,6 +2574,14 ＠＠
run_temacs_argc = -1;
main_1 (vol_argc, vol_argv, vol_envp, restarted);
+
+#ifdef _MSC_VER
+ }
+ /* VC++ documentation says that
+ GetExceptionCode() cannot be called inside the filter itself. */
+ __except (mswindows_handle_hardware_exceptions (GetExceptionCode ())) {}
+#endif
+
return 0; /* unreached */
}
＠＠ -2469,9 +2626,7 ＠＠
if (!preparing_for_armageddon && !noninteractive)
run_hook (Qkill_emacs_hook);
- /* make sure no quitting from now on!! */
- dont_check_for_quit = 1;
- Vinhibit_quit = Qt;
+ ensure_no_quitting_from_now_on ();
if (!preparing_for_armageddon)
{
＠＠ -2499,7 +2654,7 ＠＠
UNGCPRO;
- shut_down_emacs (0, STRINGP (arg) ? arg : Qnil);
+ shut_down_emacs (0, STRINGP (arg) ? arg : Qnil, 0);
#if defined(GNU_MALLOC)
__free_hook =
＠＠ -2522,45 +2677,54 ＠＠
If SIG is a signal number, print a message for it.
- This is called by fatal signal handlers, X protocol error handlers,
- and Fkill_emacs. */
+ This is called by fatal signal handlers and Fkill_emacs. It used to
+ be called by X protocol error handlers, but instead they now call
+ Fkill_emacs. */
static void
-shut_down_emacs (int sig, Lisp_Object stuff)
+shut_down_emacs (int sig, Lisp_Object stuff, int no_auto_save)
{
/* This function can GC */
/* Prevent running of hooks and other non-essential stuff
from now on. */
preparing_for_armageddon = 1;
- /* In case frames or windows are screwed up, avoid assertion
- failures here */
- Vinhibit_quit = Qt;
+ ensure_no_quitting_from_now_on ();
#ifdef QUANTIFY
quantify_stop_recording_data ();
#endif /* QUANTIFY */
-#if 0
/* This is absolutely the most important thing to do, so make sure
we do it now, before anything else. We might have crashed and
be in a weird inconsistent state, and potentially anything could
set off another protection fault and cause us to bail out
immediately. */
- /* I'm not removing the code entirely, yet. We have run up against
+ /* Steve writes the following:
+
+ [[I'm not removing the code entirely, yet. We have run up against
a spate of problems in diagnosing crashes due to crashes within
crashes. It has very definitely been determined that code called
during auto-saving cannot work if XEmacs crashed inside of GC.
We already auto-save on an itimer so there cannot be too much
unsaved stuff around, and if we get better crash reports we might
- be able to get more problems fixed so I'm disabling this. -slb */
- Fdo_auto_save (Qt, Qnil); /* do this before anything hazardous */
-#endif
+ be able to get more problems fixed so I'm disabling this. -slb]]
+ and DISABLES AUTO-SAVING ENTIRELY during crashes! Way way bad idea.
+
+ Instead let's just be more intelligent about avoiding crashing
+ when possible, esp. nested crashes.
+ */
+ if (!no_auto_save)
+ Fdo_auto_save (Qt, Qnil); /* do this before anything hazardous */
+
fflush (stdout);
reset_all_consoles ();
if (sig && sig != SIGTERM)
{
- stderr_out ("\nFatal error (%d).\n", sig);
+ if (sig == -1)
+ stderr_out ("\nFatal error.\n\n");
+ else
+ stderr_out ("\nFatal error (%d).\n\n", sig);
stderr_out
("Your files have been auto-saved.\n"
"Use `M-x recover-session' to recover them.\n"
＠＠ -2579,6 +2743,7 ＠＠
#endif
"*MAKE SURE* to include the information in the command\n"
"M-x describe-installation.\n"
+#ifndef _MSC_VER
"\n"
"If at all possible, *please* try to obtain a C stack backtrace;\n"
"it will help us immensely in determining what went wrong.\n"
＠＠ -2587,7 +2752,10 ＠＠
"directory in which you started the editor, or maybe in your home\n"
"directory), and type\n"
"\n"
- " gdb ");
+ " gdb "
+#endif
+ );
+#ifndef _MSC_VER
{
const char *name;
char *dir = 0;
＠＠ -2614,6 +2782,7 ＠＠
"(If you don't have GDB on your system, you might have DBX,\n"
"or XDB, or SDB. A similar procedure should work for all of\n"
"these. Ask your system administrator if you need more help.)\n");
+#endif /* _MSC_VER */
}
stuff_buffered_input (stuff);
＠＠ -2877,17 +3046,75 ＠＠
#ifdef USE_ASSERTIONS
/* This highly dubious kludge ... shut up Jamie, I'm tired of your slagging. */
-DOESNT_RETURN
-assert_failed (const char *file, int line, const char *expr)
-{
- stderr_out ("Fatal error: assertion failed, file %s, line %d, %s\n",
- file, line, expr);
+static int in_assert_failed;
+static const char *assert_failed_file;
+static int assert_failed_line;
+static const char *assert_failed_expr;
+
+#ifdef fprintf
+#undef fprintf
+#endif
+
#undef abort /* avoid infinite #define loop... */
+
#if defined (WINDOWSNT) && defined (DEBUG_XEMACS)
- DebugBreak ();
-#elif !defined (ASSERTIONS_DONT_ABORT)
+#define enter_debugger() DebugBreak ()
+#else
+#define enter_debugger()
+#endif
+
+void
+assert_failed (const char *file, int line, const char *expr)
+{
+ /* If we're already crashing, let's not crash again. This might be
+ critical to getting auto-saving working properly. */
+ if (fatal_error_in_progress)
+ return;
+
+ /* We are extremely paranoid so we sensibly deal with recursive
+ assertion failures. */
+ in_assert_failed++;
+
+ if (in_assert_failed >= 4)
+ _exit (-1);
+ else if (in_assert_failed == 3)
+ {
+ enter_debugger ();
+ _exit (-1);
+ }
+ else if (in_assert_failed == 2)
+ {
+ /* Not stderr_out(), which does additional things and may trigger
+ a recursive assertion failure. fprintf was undeffed above, in
+ case it was encapsulated. */
+ fprintf (stderr,
+ "Fatal error: recursive assertion failure, "
+ "file %s, line %d, %s\n",
+ file, line, expr);
+ fprintf (stderr,
+ "Original assertion failure: file %s, line %d, %s\n",
+ assert_failed_file, assert_failed_line, assert_failed_expr);
+ }
+ else
+ {
+ assert_failed_file = file;
+ assert_failed_line = line;
+ assert_failed_expr = expr;
+
+ if (!initialized)
+ fprintf (stderr,
+ "Fatal error: assertion failed, file %s, line %d, %s\n",
+ file, line, expr);
+ else
+ stderr_out ("Fatal error: assertion failed, file %s, line %d, %s\n",
+ file, line, expr);
+ }
+
+ enter_debugger ();
+#if !defined (ASSERTIONS_DONT_ABORT)
abort ();
#endif
+ in_assert_failed = 0;
}
#endif /* USE_ASSERTIONS */
Index: src/eval.c
===================================================================
RCS file: /usr/CVSroot/XEmacs/xemacs/src/eval.c,v
retrieving revision 1.27.2.22
diff -u -r1.27.2.22 eval.c
--- eval.c 2000/04/06 21:57:49 1.27.2.22
+++ eval.c 2000/05/12 09:03:47
＠＠ -143,10 +143,6 ＠＠
/* Special catch tag used in call_with_suspended_errors(). */
Lisp_Object Qunbound_suspended_errors_tag;
-/* Non-nil means we're going down, so we better not run any hooks
- or do other non-essential stuff. */
-int preparing_for_armageddon;
-
/* Non-nil means record all fset's and provide's, to be undone
if the file being autoloaded is not fully loaded.
They are recorded by being consed onto the front of Vautoload_queue:
Index: src/lisp.h
===================================================================
RCS file: /usr/CVSroot/XEmacs/xemacs/src/lisp.h,v
retrieving revision 1.38.2.61
diff -u -r1.38.2.61 lisp.h
--- lisp.h 2000/05/11 10:29:36 1.38.2.61
+++ lisp.h 2000/05/12 09:03:48
＠＠ -2130,6 +2130,7 ＠＠
Lisp_Object decode_path (const char *);
/* Nonzero means don't do interactive redisplay and don't change tty modes */
extern int noninteractive, noninteractive1;
+extern int fatal_error_in_progress;
extern int preparing_for_armageddon;
extern int emacs_priority;
extern int running_asynch_code;
Index: src/print.c
===================================================================
RCS file: /usr/CVSroot/XEmacs/xemacs/src/print.c,v
retrieving revision 1.27.2.17
diff -u -r1.27.2.17 print.c
--- print.c 2000/03/13 10:04:29 1.27.2.17
+++ print.c 2000/05/12 09:03:49
＠＠ -106,6 +106,10 ＠＠
int stdout_needs_newline;
+#ifdef WINDOWSNT
+static int no_useful_stderr;
+#endif
+
static void
std_handle_out_external (FILE *stream, Lisp_Object lstream,
const Extbyte *extptr, Extcount extlen,
＠＠ -117,9 +121,12 ＠＠
if (stream)
{
#ifdef WINDOWSNT
+ if (!no_useful_stderr)
+ no_useful_stderr = GetStdHandle (STD_ERROR_HANDLE) == 0 ? 1 : -1;
+
/* we typically have no useful stdout/stderr under windows if we're
being invoked graphically. */
- if (!noninteractive)
+ if (!noninteractive || no_useful_stderr > 0)
msw_output_console_string (extptr, extlen);
else
#endif
＠＠ -174,9 +181,16 ＠＠
int retval;
retval = vsprintf ((char *) kludge, fmt, args);
- TO_EXTERNAL_FORMAT (DATA, (kludge, strlen ((char *) kludge)),
- ALLOCA, (extptr, extlen),
- Qnative);
+ if (initialized && !fatal_error_in_progress)
+ TO_EXTERNAL_FORMAT (DATA, (kludge, strlen ((char *) kludge)),
+ ALLOCA, (extptr, extlen),
+ Qnative);
+ else
+ {
+ extptr = (Extbyte *) kludge;
+ extlen = (Extcount) strlen ((char *) kludge);
+ }
+
std_handle_out_external (stream, Qnil, extptr, extlen, 1, 1);
return retval;
}
＠＠ -190,7 +204,10 ＠＠
int retval;
va_list args;
va_start (args, fmt);
- retval = std_handle_out_va (stderr, GETTEXT (fmt), args);
+ retval =
+ std_handle_out_va
+ (stderr, initialized && !fatal_error_in_progress ? GETTEXT (fmt) : fmt,
+ args);
va_end (args);
return retval;
}
＠＠ -204,7 +221,10 ＠＠
int retval;
va_list args;
va_start (args, fmt);
- retval = std_handle_out_va (stdout, GETTEXT (fmt), args);
+ retval =
+ std_handle_out_va
+ (stdout, initialized && !fatal_error_in_progress ? GETTEXT (fmt) : fmt,
+ args);
va_end (args);
return retval;
}
Index: src/scrollbar.c
===================================================================
RCS file: /usr/CVSroot/XEmacs/xemacs/src/scrollbar.c,v
retrieving revision 1.17.2.6
diff -u -r1.17.2.6 scrollbar.c
--- scrollbar.c 2000/05/08 12:32:06 1.17.2.6
+++ scrollbar.c 2000/05/12 09:03:49
＠＠ -737,24 +737,24 ＠＠
cursor motion routines is necessary. */
if (NILP (XCDR (object)))
- window_scroll (window, Qnil, -1, ERROR_ME_NOT);
+ window_scroll (window, Qnil, -1, ERROR_ME_NOT);
else
- {
- Bufpos bufpos;
- Lisp_Object value = Fcdr (object);
-
- CHECK_INT (value);
- Fmove_to_window_line (Qzero, window);
- /* can't use Fvertical_motion() because it moves the buffer point
- rather than the window's point.
-
- #### It does? Why does it take a window argument then? */
- bufpos = vmotion (XWINDOW (window), XINT (Fwindow_point (window)),
- XINT (value), 0);
- Fset_window_point (window, make_int (bufpos));
- Fcenter_to_window_line (Qzero, window);
- }
+ {
+ Bufpos bufpos;
+ Lisp_Object value = Fcdr (object);
+ CHECK_INT (value);
+ Fmove_to_window_line (Qzero, window);
+ /* can't use Fvertical_motion() because it moves the buffer point
+ rather than the window's point.
+
+ #### It does? Why does it take a window argument then? */
+ bufpos = vmotion (XWINDOW (window), XINT (Fwindow_point (window)),
+ XINT (value), 0);
+ Fset_window_point (window, make_int (bufpos));
+ Fcenter_to_window_line (Qzero, window);
+ }
+
zmacs_region_stays = 1;
return Qnil;
}
＠＠ -776,14 +776,16 ＠＠
with their standard behaviors. It is not possible to hide the
differences down in lwlib because knowledge of XEmacs buffer and
cursor motion routines is necessary. */
+
if (NILP (XCDR (object)))
- window_scroll (window, Qnil, 1, ERROR_ME_NOT);
- {
- Lisp_Object value = Fcdr (object);
- CHECK_INT (value);
- Fmove_to_window_line (value, window);
- Fcenter_to_window_line (Qzero, window);
- }
+ window_scroll (window, Qnil, 1, ERROR_ME_NOT);
+ else
+ {
+ Lisp_Object value = Fcdr (object);
+ CHECK_INT (value);
+ Fmove_to_window_line (value, window);
+ Fcenter_to_window_line (Qzero, window);
+ }
zmacs_region_stays = 1;
return Qnil;
cvs server: Diffing src/m
cvs server: Diffing src/s
cvs server: Diffing tests
cvs server: Diffing tests/DLL
cvs server: Diffing tests/Dnd
cvs server: Diffing tests/automated
cvs server: Diffing tests/mule
cvs server: Diffing tests/tooltalk
--
Ben
In order to save my hands, I am cutting back on my mail. I also write
as succinctly as possible -- please don't be offended. If you send me
mail, you _will_ get a response, but please be patient, especially for
XEmacs-related mail. If you need an immediate response and it is not
apparent in your message, please say so. Thanks for your understanding.
See also http://www.666.com/ben/chronic-pain/
- 1 participants
- 0 comments
Re: detachable menubars for XEmacs...
25 years, 5 months
Karl M. Hegbloom
Karl M. Hegbloom
Iirc, tearoff toolbars is an option you can set in the gnome control
panel. (the toolbox icon on the default panel, or under Settings on
the main panel menu. So for gnome compliance, I guess you honor
that. It's under the GUI folder there.
--
I am karlheg of deB-ORG. You will be freed.
mailto:karlheg＠debian.org (Karl M. Hegbloom)
Portland, OR USA
Debian GNU Potato Linux 2.2 AMD K6-200(＠233) XEmacs-21.2beta
- 1 participants
- 0 comments
Re: XEmacs/GTK does menus...
25 years, 5 months
Karl M. Hegbloom
Karl M. Hegbloom
>>>>> "William" == William M Perry <wmperry(a)aventail.com> writes:
>> > Well, is the only reason for motion events while over the toolbar
>> > for tooltips? Those are handled natively by the GTK toolbar.
>>
>> Can you embed images in such tooltips?
William> Ick! I should hope not. :)
Ahhhwwww. :( I want a dancing paperclip, Bill!
Hey, can you do the transparent / shaded XEmacs thing? That would be
pretty neat. I think it just looks for a Gnome specific property on
the root window that holds the pixmap.
- 1 participants
- 0 comments
Re: XEmacs/GTK does menus...
25 years, 5 months
Karl M. Hegbloom
Karl M. Hegbloom
Look up `xgdvi' at `www.google.com', and follow the links to the
fontlib. Will that work?
--
I am karlheg of deB-ORG. You will be freed.
mailto:karlheg＠debian.org (Karl M. Hegbloom)
Portland, OR USA
Debian GNU Potato Linux 2.2 AMD K6-200(＠233) XEmacs-21.2beta
- 1 participants
- 0 comments
[crash] Popup menus, lwlib.
25 years, 5 months
Karl M. Hegbloom
Karl M. Hegbloom
Currently, with latest beta and HEAD, if I use the far button to
bring up a popup mode menu, then click outside of the menu, XEmacs
crashes with an assert failure. I don't understand `lwlib' much at
all... I started poking around to see if I could fix it...
At one point a function is called with an XPointer arg of -1, and
that is cast to a Lisp_Object, then when the XCAR of that is taken,
the assertion failure is thrown.
I can try and spend time on learning how that works... but it would
be faster if someone who already understands would do it. I have a
HUGE reading list right now and would prefer to spend my time at that.
--
I am karlheg of deB-ORG. You will be freed.
mailto:karlheg＠debian.org (Karl M. Hegbloom)
Portland, OR USA
Debian GNU Potato Linux 2.2 AMD K6-200(＠233) XEmacs-21.2beta
- 1 participants
- 0 comments
Re: XEmacs/GTK does menus...
25 years, 5 months
Karl M. Hegbloom
Karl M. Hegbloom
Are you using `glade' to build the interface, Bill?
`apt-get install glade'
--
I am karlheg of deB-ORG. You will be freed.
mailto:karlheg＠debian.org (Karl M. Hegbloom)
Portland, OR USA
Debian GNU Potato Linux 2.2 AMD K6-200(＠233) XEmacs-21.2beta
- 1 participants
- 0 comments
Re: XEmacs/GTK does menus...
25 years, 5 months
Michael Harnois
Michael Harnois
So, is this at a point where we should send you lisp and/or C
backtraces when things upshit and die?
--
Michael D. Harnois, Redeemer Lutheran Church, Washburn, IA
mdharnois(a)home.com aa0bt(a)aa0bt.ampr.org
It is better to die excommunicated than to live
and violate the conscience. -- Thomas Aquinas
- 1 participants
- 0 comments
Re: XEmacs/GTK does menus...
25 years, 5 months
William M. Perry
William M. Perry
Michael Harnois <mdharnois(a)home.com> writes:
> On 11 May 2000 10:10:13 -0500, wmperry(a)aventail.com (William M. Perry) said:
>
> > ./configure --with-menubars=no --with-toolbars=no
>
> Gee, you've forgotten already that menubars work now?
Hehehehe... even better, toolbars work now. :) Have to update the web page
tonight.
-bp
- 1 participants
- 0 comments