Mercurial > hg > xemacs-beta
diff src/eval.c @ 2500:3d8143fc88e1
[xemacs-hg @ 2005-01-24 23:33:30 by ben]
get working with VC7
config.inc.samp: Declare OPTIONAL_LIBRARY_DIR as root of library directories.
Redo all graphics library defaults to mirror the versions and
directories in the current binary aux distribution on xemacs
web site. Enable TIFF and COMPFACE by default since you can
now compile with them and binary libs are provided.
xemacs.mak: Put our own directories first in case of conflict (e.g. config.h
in compface).
xemacs.mak: Use MSVCRT to avoid link problems.
s/windowsnt.h:
bytecode.c, print.c: Add casts to avoid warning.
compiler.h: Add MSC_VERSION and include definitions of DOESNT_RETURN and
friends here, like for GCC. Need different definitions for VC7
and VC6.
s/windowsnt.h: Remove stuff moved to compiler.h. Disable warning 4646 ("function
declared with __declspec(noreturn) has non-void return type")
on VC7 since lots of Lisp primitives trigger this and there is
no easy way to kludge around the warning.
glyphs-eimage.c: Some really nasty hacks to allow TIFF and JPEG to both be compiled.
#### The better solution is to move the TIFF and JPEG code to
different files.
glyphs-msw.c: Define __STDC__ to avoid problems with compface.h.
intl-auto-encap-win32.c, intl-auto-encap-win32.h, intl-encap-win32.c, syswindows.h: Those wankers at Microsoft cannot leave well enough alone.
Various functions change parameter types semi-randomly between
VC6 and VC7, so we need to include our own versions that
can handle both kinds with appropriate casting.
EmacsFrame.c, EmacsShell-sub.c, EmacsShell.c, alloc.c, alloca.c, buffer.c, bytecode.c, charset.h, chartab.c, cm.c, console-stream.c, console.c, data.c, debug.h, device-msw.c, device-tty.c, device-x.c, doprnt.c, dumper.c, dynarr.c, elhash.c, emacs.c, eval.c, event-Xt.c, event-gtk.c, event-msw.c, event-stream.c, events.c, extents.c, faces.c, file-coding.c, fileio.c, fns.c, font-lock.c, frame-gtk.c, frame-x.c, frame.c, free-hook.c, gccache-gtk.c, glyphs-eimage.c, glyphs-gtk.c, glyphs-msw.c, glyphs-x.c, glyphs.c, gtk-glue.c, gutter.c, input-method-xlib.c, insdel.c, intl-win32.c, keymap.c, lisp.h, lread.c, lstream.c, macros.c, malloc.c, menubar-gtk.c, menubar-msw.c, menubar-x.c, mule-coding.c, native-gtk-toolbar.c, number.c, objects-msw.c, objects.c, print.c, process-nt.c, process-unix.c, process.c, ralloc.c, rangetab.c, redisplay-gtk.c, redisplay-msw.c, redisplay-output.c, redisplay-tty.c, redisplay-x.c, redisplay.c, regex.c, scrollbar-gtk.c, scrollbar-x.c, search.c, select-x.c, signal.c, specifier.c, specifier.h, strftime.c, sunplay.c, symbols.c, sysdep.c, sysproc.h, text.c, text.h, toolbar-common.c, toolbar-msw.c, toolbar.c, ui-gtk.c, unexnt.c, unicode.c, win32.c, window.c, xgccache.c, s/windowsnt.h: abort() -> ABORT(). Eliminate preprocessor games with abort()
since it creates huge problems in VC7, solvable only by including
massive amounts of files in every compile (and not worth it).
author | ben |
---|---|
date | Mon, 24 Jan 2005 23:34:34 +0000 |
parents | ab71ad6ff3dd |
children | 989a7680c221 |
line wrap: on
line diff
--- a/src/eval.c Mon Jan 24 22:51:18 2005 +0000 +++ b/src/eval.c Mon Jan 24 23:34:34 2005 +0000 @@ -1642,7 +1642,7 @@ #ifdef DEFEND_AGAINST_THROW_RECURSION /* die if we recurse more than is reasonable */ if (++throw_level > 20) - abort (); + ABORT (); #endif #ifdef ERROR_CHECK_TRAPPING_PROBLEMS @@ -2228,7 +2228,7 @@ /* who knows how much has been initialized? Safest bet is just to bomb out immediately. */ stderr_out ("Error before initialization is complete!\n"); - abort (); + ABORT (); } assert (!gc_in_progress); @@ -6544,7 +6544,7 @@ t each time a Control-G is detected, and to `critical' each time a Shift-Control-G is detected. The XEmacs core C code is littered with calls to the QUIT; macro, which check the values of `quit-flag' and -`inhibit-quit' and abort (or more accurately, call (signal 'quit)) if +`inhibit-quit' and ABORT (or more accurately, call (signal 'quit)) if it's correct to do so. */ ); Vquit_flag = Qnil; @@ -6607,7 +6607,7 @@ \(force-debugging-signal t). This is useful because debugging noninteractive runs of XEmacs is often very difficult, since they typically happen as part of sometimes large and complex make suites (e.g. rebuilding -the XEmacs packages). NOTE: This runs abort()!!! (As well as and after +the XEmacs packages). NOTE: This runs ABORT()!!! (As well as and after executing INT 3 under MS Windows, which should invoke a debugger if it's active.) This is guaranteed to kill XEmacs! (But in this situation, XEmacs is about to die anyway, and if no debugger is present, this will usefully