[PATCH 21.5] shell-command regression
18 years, 12 months
Jerry James
Jerry James
PATCH 21.5
Invoking shell-command with a buffer name or object and an asynchronous
command (i.e., one ending with "&") does NOT put the shell output into
the requested buffer like it did in 21.4. This patch fixes the
regression.
lisp/ChangeLog addition:
2006-06-30 Jerry James <james(a)xemacs.org>
* process.el (shell-command): If a specific output buffer was
requested, use it.
xemacs-21.5 source patch:
Diff command: cvs -q diff -uN
Files affected: lisp/process.el
Index: lisp/process.el
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/lisp/process.el,v
retrieving revision 1.19
diff -d -u -r1.19 process.el
--- lisp/process.el 2005/01/09 09:44:44 1.19
+++ lisp/process.el 2006/06/30 20:49:29
＠＠ -359,8 +359,8 ＠＠
;; Command ending with ampersand means asynchronous.
(progn
(if-fboundp 'background
- (background (substring command 0
- (match-beginning 0)))
+ (background (substring command 0 (match-beginning 0))
+ output-buffer)
(error
'unimplemented
"backgrounding a shell command requires package `background'")))
--
Jerry James, Assistant Professor james(a)xemacs.org
Computer Science Department http://www.cs.usu.edu/~jerry/
Utah State University
- 1 participants
- 1 comments
[AC21.5] Shut up marking and insertion commands in test suite
19 years
Stephen J. Turnbull
Stephen J. Turnbull
APPROVE COMMIT 21.5
Several of the tests call functions that produce messages concerning
marks being set or text inserted. This patch introduces a macro
Silence-Message for use in the test suite to suppress such output, and
uses it in several places.
Index: tests/ChangeLog
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/tests/ChangeLog,v
retrieving revision 1.81
diff -u -r1.81 ChangeLog
--- tests/ChangeLog 24 Jun 2006 13:50:19 -0000 1.81
+++ tests/ChangeLog 24 Jun 2006 13:54:47 -0000
＠＠ -0,0 +1,7 ＠＠
+2006-06-24 Stephen J. Turnbull <stephen(a)xemacs.org>
+
+ * automated/test-harness.el (Silence-Message): New macro.
+ * automated/mule-tests.el: Use it.
+ * automated/region-tests.el: Use it.
+ * automated/tag-tests.el: Use it.
+
Index: tests/automated/mule-tests.el
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/tests/automated/mule-tests.el,v
retrieving revision 1.9
diff -u -r1.9 mule-tests.el
--- tests/automated/mule-tests.el 3 Jun 2006 17:51:06 -0000 1.9
+++ tests/automated/mule-tests.el 24 Jun 2006 13:54:47 -0000
＠＠ -340,18 +340,18 ＠＠
'iso-8859-2))
)
;; This is how you suppress output from `message', called by `write-region'
- (flet ((append-message (&rest args) ()))
- (Assert (not (equal name1 name2)))
- (Assert (not (file-exists-p name1)))
- (write-region (point-min) (point-max) name1)
- (Assert (file-exists-p name1))
- (when (fboundp 'make-symbolic-link)
- (make-symbolic-link name1 name2)
- (Assert (file-exists-p name2))
- (Assert (equal (file-truename name2) name1))
- (Assert (equal (file-truename name1) name1)))
+ (Assert (not (equal name1 name2)))
+ (Assert (not (file-exists-p name1)))
+ (Silence-Message
+ (write-region (point-min) (point-max) name1))
+ (Assert (file-exists-p name1))
+ (when (fboundp 'make-symbolic-link)
+ (make-symbolic-link name1 name2)
+ (Assert (file-exists-p name2))
+ (Assert (equal (file-truename name2) name1))
+ (Assert (equal (file-truename name1) name1)))
- (ignore-file-errors (delete-file name1) (delete-file name2))))
+ (ignore-file-errors (delete-file name1) (delete-file name2)))
;; Add many more file operation tests here...
Index: tests/automated/region-tests.el
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/tests/automated/region-tests.el,v
retrieving revision 1.1
diff -u -r1.1 region-tests.el
--- tests/automated/region-tests.el 24 Feb 2005 23:02:37 -0000 1.1
+++ tests/automated/region-tests.el 24 Jun 2006 13:54:47 -0000
＠＠ -43,10 +43,12 ＠＠
;; Using active regions
(let ((zmacs-regions t)
(first-buffer (current-buffer)))
- (insert (buffer-name))
+ (Silence-Message
+ (insert (buffer-name)))
(Assert (not (region-exists-p)))
(Assert (not (region-active-p)))
- (mark-whole-buffer)
+ (Silence-Message
+ (mark-whole-buffer))
(Assert (region-exists-p))
(Assert (region-active-p))
;; Turn off active regions
＠＠ -62,7 +64,8 ＠＠
;; Region is active again
(Assert (region-active-p))
(with-temp-buffer
- (insert (buffer-name))
+ (Silence-Message
+ (insert (buffer-name)))
;; Region exists in first buffer, not this second one
(Assert (not (region-exists-p)))
;; Region not active in this second temp buffer
＠＠ -70,7 +73,8 ＠＠
;; Region still active in first temp buffer
(Assert (eq (zmacs-region-buffer) first-buffer))
;; Activate region in second temp buffer
- (mark-whole-buffer)
+ (Silence-Message
+ (mark-whole-buffer))
;; Region exists in second temp buffer
(Assert (region-exists-p))
;; Region active in second temp buffer
Index: tests/automated/tag-tests.el
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/tests/automated/tag-tests.el,v
retrieving revision 1.3
diff -u -r1.3 tag-tests.el
--- tests/automated/tag-tests.el 13 May 2004 16:41:27 -0000 1.3
+++ tests/automated/tag-tests.el 24 Jun 2006 13:54:47 -0000
＠＠ -68,20 +68,23 ＠＠
(let ((tags-always-exact t))
;; Search for the tag "mystruct"; this should succeed
- (find-tag "mystruct")
+ (Silence-Message
+ (find-tag "mystruct"))
(Assert (eq (point) 2))
;; Search again. The search should fail, based on the patch that
;; Sven Grundmann submitted for 21.4.16.
(Check-Error-Message error "No more entries matching mystruct"
- (tags-loop-continue)))
+ (Silence-Message
+ (tags-loop-continue))))
(let ((tags-always-exact nil))
;; Search for the definition of "require". Until the etags.el upgrade
;; from 21.5 in 21.4.16, this test would fail.
(condition-case nil
- (find-tag "require")
+ (Silence-Message
+ (find-tag "require"))
(t t))
(Assert (eq (point) 52)))
Index: tests/automated/test-harness.el
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/tests/automated/test-harness.el,v
retrieving revision 1.21
diff -u -r1.21 test-harness.el
--- tests/automated/test-harness.el 24 Jun 2006 13:50:23 -0000 1.21
+++ tests/automated/test-harness.el 24 Jun 2006 13:54:47 -0000
＠＠ -38,6 +38,14 ＠＠
;;; Implementation-Incomplete-Expect-Failure wrapper macros to mark them.
;;; A lot of the tests we run push limits; suppress Ebola message with the
;;; Ignore-Ebola wrapper macro.
+;;; Some noisy code will call `message'. Output from `message' can be
+;;; suppressed with the Silence-Message macro. Functions that are known to
+;;; issue messages include `write-region', `find-tag', `tag-loop-continue',
+;;; `insert', and `mark-whole-buffer'. N.B. The Silence-Message macro
+;;; currently does not suppress the newlines printed by `message'.
+;;; Definitely do not use Silence-Message with Check-Message.
+;;; In general it should probably only be used on code that prepares for a
+;;; test, not on tests.
;;;
;;; You run the tests using M-x test-emacs-test-file,
;;; or $(EMACS) -batch -l .../test-harness.el -f batch-test-emacs file ...
＠＠ -291,7 +299,7 ＠＠
,quoted-body ',expected-error error-info)
(incf wrong-error-failures)))))
-
+ ;; Do not use this with Silence-Message.
(defmacro Check-Message (expected-message-regexp &rest body)
(Skip-Test-Unless (fboundp 'defadvice)
"can't defadvice"
＠＠ -322,6 +330,10 ＠＠
,quoted-body error-info)
(incf other-failures)))
(ad-unadvise 'message)))))
+
+ ;; #### Perhaps this should override `message' itself, too?
+ (defmacro Silence-Message (&rest body)
+ `(flet ((append-message (&rest args) ())) ,＠body))
(defmacro Ignore-Ebola (&rest body)
`(let ((debug-issue-ebola-notices -42)) ,＠body))
--
Graduate School of Systems and Information Engineering University of Tsukuba
http://turnbull.sk.tsukuba.ac.jp/ Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Economics of Information Communication and Computation Systems
Experimental Economics, Microeconomic Theory, Game Theory
- 3 participants
- 2 comments
[AC21.5R21.4] Document X11R7 distros often omit bitmaps
19 years
Stephen J. Turnbull
Stephen J. Turnbull
APPROVE COMMIT 21.5 RECOMMEND 21.4
Index: ChangeLog
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/ChangeLog,v
retrieving revision 1.502
diff -u -r1.502 ChangeLog
--- ChangeLog 16 May 2006 08:22:56 -0000 1.502
+++ ChangeLog 16 May 2006 15:30:22 -0000
＠＠ -0,0 +1,4 ＠＠
+2006-05-17 Stephen J. Turnbull <stephen(a)xemacs.org>
+
+ * PROBLEMS: X11R7 loses x11/bitmaps/gray.
+
Index: man/ChangeLog
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/man/ChangeLog,v
retrieving revision 1.341
diff -u -r1.341 ChangeLog
--- man/ChangeLog 16 May 2006 08:23:49 -0000 1.341
+++ man/ChangeLog 16 May 2006 15:30:23 -0000
＠＠ -0,0 +1,6 ＠＠
+2006-05-17 Stephen J. Turnbull <stephen(a)xemacs.org>
+
+ * xemacs-faq.texi (Q2.2.3): New node.
+ (Q2.2.2, Q2.3.1): Fix navigation references.
+ (Top, Installation): Add to menus.
+
Index: PROBLEMS
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/PROBLEMS,v
retrieving revision 1.71
diff -u -r1.71 PROBLEMS
--- PROBLEMS 16 Nov 2005 12:12:57 -0000 1.71
+++ PROBLEMS 16 May 2006 15:30:22 -0000
＠＠ -44,6 +44,16 ＠＠
Much general information is in INSTALL. If it's covered in
INSTALL, we don't repeat it here.
+*** X11/bitmaps/gray (or other X11-related file) not found.
+
+The X11R6 distribution was monolithic, but the X11R7 distribution is
+much more modular. Many OS distributions omit these bitmaps (assuming
+nobody uses them, evidently). Your OS distribution should have a
+developer's package containing these files, probably with a name
+containing the string "bitmap". Known package names (you may need to
+add an extension such as .deb or .rpm) include x11/xbitmaps (Ubuntu)
+and xorg-x11-xbitmaps (Fedora Core 5).
+
*** How do I configure to get the buffer tabs/progress bars?
These features depend on support for "native widgets". Use the
Index: man/xemacs-faq.texi
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/man/xemacs-faq.texi,v
retrieving revision 1.88
diff -u -r1.88 xemacs-faq.texi
--- man/xemacs-faq.texi 27 Dec 2005 18:51:21 -0000 1.88
+++ man/xemacs-faq.texi 16 May 2006 15:30:24 -0000
＠＠ -293,6 +293,7 ＠＠
2.2: Unix/Mac OS X Installation (Also Relevant to Cygwin, MinGW)
* Q2.2.1:: Libraries in non-standard locations
* Q2.2.2:: Why can't I strip XEmacs?
+* Q2.2.3:: X11/bitmaps/gray (or other X11-related file) not found.
2.3: Windows Installation (Windows, Cygwin, MinGW)
* Q2.3.1:: What exactly are all the different ways to build XEmacs under Windows?
＠＠ -3089,6 +3090,7 ＠＠
2.2: Unix/Mac OS X Installation (Also Relevant to Cygwin, MinGW)
* Q2.2.1:: Libraries in non-standard locations
* Q2.2.2:: Why can't I strip XEmacs?
+* Q2.2.3:: X11/bitmaps/gray (or other X11-related file) not found.
2.3: Windows Installation (Windows, Cygwin, MinGW)
* Q2.3.1:: What exactly are all the different ways to build XEmacs under Windows?
＠＠ -3584,7 +3586,7 ＠＠
--with-site-prefixes=WHATEVER
＠end example
-＠node Q2.2.2, Q2.3.1, Q2.2.1, Installation
+＠node Q2.2.2, Q2.2.3, Q2.2.1, Installation
＠unnumberedsubsec Q2.2.2: Why can't I strip XEmacs?
＠email{cognot＠(a)fronsac.ensg.u-nancy.fr, Richard Cognot} writes:
＠＠ -3639,9 +3641,20 ＠＠
＠end enumerate
＠end quotation
+＠node Q2.2.3, Q2.3.1, Q2.2.2, Installation
+＠unnumberedsubsec Q2.2.3: X11/bitmaps/gray (or other X11-related file) not found.
+
+The X11R6 distribution was monolithic, but the X11R7 distribution is
+much more modular. Many OS distributions omit these bitmaps (assuming
+nobody uses them, evidently). Your OS distribution should have a
+developer's package containing these files, probably with a name
+containing the string "bitmap". Known package names (you may need to
+add an extension such as .deb or .rpm) include x11/xbitmaps (Ubuntu)
+and xorg-x11-xbitmaps (Fedora Core 5).
+
＠unnumberedsec 2.3: Windows Installation (Windows, Cygwin, MinGW)
-＠node Q2.3.1, Q2.3.2, Q2.2.2, Installation
+＠node Q2.3.1, Q2.3.2, Q2.2.3, Installation
＠unnumberedsubsec Q2.3.1: What exactly are all the different ways to build XEmacs under Windows?
XEmacs can be built in several ways in the MS Windows environment.
--
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.
- 2 participants
- 1 comments
[AC21.5R21.4] Improve geometry debugging messages
19 years
Stephen J. Turnbull
Stephen J. Turnbull
APPROVE COMMIT 21.5 RECOMMEND 21.4
Should apply to 21.4 but I haven't actually tested.
This is "safe" in the sense that the builder has to directly specify
the controlling #define, there's no ordinary way to configure it. So
only developers explicitly trying to debug geometry management will
ever run into it.
I haven't found a way to completely test this, as I don't have a
window manager that ever returns XtGeometryAlmost. :-) However, the
Shell dimensions reported are now meaningful and correct.
Based on a suggestion by Francois L'Archeveque.
diff --git a/src/ChangeLog b/src/ChangeLog
index cca5be4..12a7348 100644
--- src/ChangeLog
+++ src/ChangeLog
＠＠ -0,0 +1,6 ＠＠
+2006-05-08 Stephen J. Turnbull <stephen(a)xemacs.org>
+
+ * EmacsShell-sub.c (RootGeometryManager):
+ Change garbage to useful messages under DEBUG_GEOMETRY_MANAGEMENT.
+ Based on a suggestion by Francois L'Archeveque.
+
index ed82214..e335a4a 100644
--- src/EmacsShell-sub.c
+++ src/EmacsShell-sub.c
＠＠ -339,11 +339,16 ＠＠ RootGeometryManager (Widget gw,
result == XtGeometryAlmost ? "XtGeometryAlmost" :
"XtGeometryDone");
if (reply->request_mode & CWWidth)
- printf ("width returned;");
+ printf ("width returned was %d%s",
+ reply->width,
+ reply->request_mode & CWHeight ? "; " : ".\n");
if (reply->request_mode & CWHeight)
- printf ("height returned;");
- printf ("\n");
- printf (" resulting shell size: %d %d\n", reply->width, reply->height);
+ printf ("height returned was %d.\n", reply->height);
+ /* #### does this also need to depend on the result?
+ With XtGeometryYes there doesn't seem to be a useful reply object. */
+ printf (" resulting shell size: %d %d\n",
+ reply->request_mode & CWWidth ? reply->width : w->core.width,
+ reply->request_mode & CWHeight ? reply->height : w->core.height);
printf ("----------\n");
fflush (stdout);
#endif
--
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.
- 3 participants
- 3 comments
[PATCH 21.4] Collected fixes from static checkers
19 years
Jerry James
Jerry James
PATCH 21.4
DO NOT APPLY THIS PATCH: It is causing an assertion failure in the md5
tests. I will track it down and send a superceding version of this
patch. I'm only sending this because I spent so much time writing
it.:-) Please review it anyway. If there is anything controversial,
I'm happy to split it out as a separate patch.
This is the 21.4 version of the patches I've been applying to 21.5
recently, to fix various problems (or potential problems) found by
various code analysis tools. I ran the tools over the 21.4 base and
found a few that don't appear in the 21.5 code base, as well. I won't
repeat the rationale for those that have already been sent in for 21.5,
but will describe the new changes for 21.4.
EmacsShell-sub.c: If the for loop terminates because i becomes too
large, then child will not have been initialized, so dereferencing it is
bad. Let's just dereference it where we set it.
dumper.c: The array 'backtrace' has 65536 elements. Here we are
checking to see if we have overflowed the array, but the check is off by
one. If me == 65536, then we will try to access backtrace[65536], but
that's the first bad element, not the last good one.
editfns.c: The array warnpath is shorter than the array path, but we
were copying from path to warnpath without checking how many bytes were
transferred. I tried to rationalize the code a bit. This code was
rewritten for 21.5.
keymap.c: Actually, this problem appears in 21.5 too. I'll send in a
patch for it later. We are allocating a new array and then throwing it
away. It looks like we are rolling our own realloc() here. I don't
understand that, but we might as well finish the job instead of leaking
memory.
sound.c: We're discussing this one on xemacs-beta right now. This is my
preferred solution.
lwlib/ChangeLog addition:
2006-06-22 Jerry James <james(a)xemacs.org>
* lwlib-Xlw.c (xlw_scrollbar_callback): Do not dereference
instance before checking whether it is NULL.
* xlwmenu.c (xlw_map_menu): Prevent uninitialized access to root
and waste.
src/ChangeLog addition:
2006-06-22 Jerry James <james(a)xemacs.org>
* EmacsShell-sub.c (ChangeManaged):
* device-x.c (x_IO_error_handler): Do not dereference d if it is
NULL.
* dgif_lib.c (DGifCloseFile): Do not dereference GifFile before
checking if it is NULL. Also fix a memory leak.
* dialog-x.c (dbox_selection_callback): Ensure f is non-NULL, then
dereference it, not the other way around.
* dumper.c (pdump_register_object): Fix off-by-one array bounds
overflow check.
* dumper.c (pdump_register_struct): Ditto.
* editfns.c (Ftemp_directory): Don't let a copy from (long) path
to (short) warnpath overflow warnpath.
* extents.c (detach_all_extents): Call extent_list_delete_all with
a non-NULL parameter only.
* glyphs-widget.c (widget_query_geometry): Guard against possibly
NULL width and height.
* input-method-xlib.c (XIM_SetGeometry): Do not dereference f or
xic before checking if they are NULL.
* keymap.c (where_is_recursive_mapper): Don't throw away the newly
allocated array.
* md5.c (Fmd5): Check whether Lstream_read encountered an error.
* nas.c (Err): Fix a memory leak.
* redisplay-gtk.c (gtk_output_display_block): Fix a Dynarr leak.
Don't create the buffer if there is nothing to do.
* redisplay-msw.c (mswindows_output_display_block): Ditto.
* redisplay-output.c (redisplay_output_layout): Ditto.
* redisplay-tty.c (tty_output_display_block): Ditto.
* redisplay-x.c (x_output_display_block): Ditto.
* scrollbar-gtk.c (gtk_free_scrollbar_instance): Do not
dereference instance->scrollbar_data if it is NULL.
* scrollbar-x.c (x_free_scrollbar_instance): Ditto.
* scrollbar-msw.c (mswindows_free_scrollbar_instance): Ditto, but
for sb->scrollbar_data.
* sound.c (init_native_sound): Avoid a buffer overflow by
allocating a buffer with the appropriate size.
xemacs-21.4 source patch:
Diff command: cvs -q diff -uN
Files affected: src/sound.c src/scrollbar-x.c src/scrollbar-msw.c src/scrollbar-gtk.c src/redisplay-x.c src/redisplay-tty.c src/redisplay-output.c src/redisplay-msw.c src/redisplay-gtk.c src/nas.c src/md5.c src/keymap.c src/input-method-xlib.c src/glyphs-widget.c src/extents.c src/editfns.c src/dumper.c src/dialog-x.c src/dgif_lib.c src/device-x.c src/EmacsShell-sub.c lwlib/xlwmenu.c lwlib/lwlib-Xlw.c
Index: lwlib/lwlib-Xlw.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/lwlib/lwlib-Xlw.c,v
retrieving revision 1.8.2.2
diff -d -u -r1.8.2.2 lwlib-Xlw.c
--- lwlib/lwlib-Xlw.c 2005/11/26 03:47:33 1.8.2.2
+++ lwlib/lwlib-Xlw.c 2006/06/22 23:33:13
＠＠ -163,13 +163,13 ＠＠
XlwScrollBarCallbackStruct *data =
(XlwScrollBarCallbackStruct *) call_data;
scroll_event event_data;
- scrollbar_values *val =
- (scrollbar_values *) instance->info->val->scrollbar_data;
+ scrollbar_values *val;
double percent;
if (!instance || widget->core.being_destroyed)
return;
+ val = (scrollbar_values *) instance->info->val->scrollbar_data;
id = instance->info->id;
percent = (double) (data->value - 1) / (double) (INT_MAX - 1);
Index: lwlib/xlwmenu.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/lwlib/xlwmenu.c,v
retrieving revision 1.33.2.2
diff -d -u -r1.33.2.2 xlwmenu.c
--- lwlib/xlwmenu.c 2003/02/15 18:57:19 1.33.2.2
+++ lwlib/xlwmenu.c 2006/06/22 23:33:14
＠＠ -3412,8 +3412,8 ＠＠
if (!mw->menu.pointer_grabbed)
{
XWindowAttributes ret;
- Window parent,root;
- Window *waste;
+ Window parent,root = 0UL;
+ Window *waste = NULL;
unsigned int num_waste;
lw_menu_active = True;
Index: src/EmacsShell-sub.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/EmacsShell-sub.c,v
retrieving revision 1.5.2.2
diff -d -u -r1.5.2.2 EmacsShell-sub.c
--- src/EmacsShell-sub.c 2005/01/31 02:55:01 1.5.2.2
+++ src/EmacsShell-sub.c 2006/06/22 23:33:14
＠＠ -366,11 +366,11 ＠＠
for (i = 0; i < w->composite.num_children; i++) {
if (XtIsManaged(w->composite.children[i])) {
child = w->composite.children[i];
+ update_size_hints_internal (w, child->core.width,
+ child->core.height);
break;
}
}
-
- update_size_hints_internal (w, child->core.width, child->core.height);
}
/* call the real ChangeManaged */
Index: src/device-x.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/device-x.c,v
retrieving revision 1.36.2.3
diff -d -u -r1.36.2.3 device-x.c
--- src/device-x.c 2005/01/31 02:55:08 1.36.2.3
+++ src/device-x.c 2006/06/22 23:33:14
＠＠ -1161,8 +1161,10 ＠＠
Xlib might just decide to exit(). So we mark the offending
console for deletion and throw to top level. */
if (d)
- enqueue_magic_eval_event (io_error_delete_device, dev);
- DEVICE_X_BEING_DELETED (d) = 1;
+ {
+ enqueue_magic_eval_event (io_error_delete_device, dev);
+ DEVICE_X_BEING_DELETED (d) = 1;
+ }
Fthrow (Qtop_level, Qnil);
return 0; /* not reached */
Index: src/dgif_lib.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/dgif_lib.c,v
retrieving revision 1.12.2.1
diff -d -u -r1.12.2.1 dgif_lib.c
--- src/dgif_lib.c 2004/05/06 22:21:43 1.12.2.1
+++ src/dgif_lib.c 2006/06/22 23:33:14
＠＠ -366,10 +366,11 ＠＠
******************************************************************************/
int DGifCloseFile(GifFileType *GifFile)
{
- GifFilePrivateType *Private = (GifFilePrivateType *)GifFile->Private;
+ GifFilePrivateType *Private;
if (GifFile == NULL) return -1;
+ Private = (GifFilePrivateType *)GifFile->Private;
if (!IS_READABLE(Private))
{
/* This file was NOT open for reading: */
＠＠ -930,7 +931,10 ＠＠
Object->Colors = (GifColorType *)calloc(ColorCount, sizeof(GifColorType));
if (Object->Colors == (GifColorType *)NULL)
+ {
+ free (Object);
return((ColorMapObject *)NULL);
+ }
Object->ColorCount = ColorCount;
Object->BitsPerPixel = BitSize(ColorCount);
Index: src/dialog-x.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/dialog-x.c,v
retrieving revision 1.7
diff -d -u -r1.7 dialog-x.c
--- src/dialog-x.c 2001/04/12 18:23:34 1.7
+++ src/dialog-x.c 2006/06/22 23:33:14
＠＠ -105,7 +105,7 ＠＠
ourselves. */
#ifdef EXTERNAL_WIDGET
/* #### Not sure if this special case is necessary. */
- if (!FRAME_X_EXTERNAL_WINDOW_P (f) && f)
+ if (f && !FRAME_X_EXTERNAL_WINDOW_P (f))
#else
if (f)
#endif
Index: src/dumper.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/dumper.c,v
retrieving revision 1.2.2.5
diff -d -u -r1.2.2.5 dumper.c
--- src/dumper.c 2005/01/31 02:55:08 1.2.2.5
+++ src/dumper.c 2006/06/22 23:33:14
＠＠ -566,7 +566,7 ＠＠
if (imp->description)
{
int me = depth++;
- if (me>65536)
+ if (me>=65536)
{
stderr_out ("Backtrace overflow, loop ?\n");
ABORT ();
＠＠ -601,7 +601,7 ＠＠
{
int me = depth++;
int i;
- if (me>65536)
+ if (me>=65536)
{
stderr_out ("Backtrace overflow, loop ?\n");
ABORT ();
Index: src/editfns.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/editfns.c,v
retrieving revision 1.27.2.4
diff -d -u -r1.27.2.4 editfns.c
--- src/editfns.c 2002/08/20 11:36:52 1.27.2.4
+++ src/editfns.c 2006/06/22 23:33:14
＠＠ -644,14 +644,17 ＠＠
}
else
{
- strcpy(path, getenv("HOME")); strncat(path, "/tmp/", _POSIX_PATH_MAX);
+ path[5 + _POSIX_PATH_MAX] = '\0';
+ strncpy(path, getenv("HOME"), 5 + _POSIX_PATH_MAX);
+ strncat(path, "/tmp/", 5 + _POSIX_PATH_MAX);
if (stat(path, &st) < 0 && errno == ENOENT)
{
int fd;
- char warnpath[1+_POSIX_PATH_MAX];
+ char warnpath[6+_POSIX_PATH_MAX];
mkdir(path, 0700); /* ignore retvals */
- strcpy(warnpath, path);
- strncat(warnpath, ".created_by_xemacs", _POSIX_PATH_MAX);
+ warnpath[_POSIX_PATH_MAX] = '\0';
+ strncpy(warnpath, path, 5 + _POSIX_PATH_MAX);
+ strncat(warnpath, ".created_by_xemacs", 5 + _POSIX_PATH_MAX);
if ((fd = open(warnpath, O_WRONLY|O_CREAT, 0644)) > 0)
{
write(fd, "XEmacs created this directory because /tmp/<yourname> was unavailable -- \nPlease check !\n", 89);
Index: src/extents.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/extents.c,v
retrieving revision 1.31.2.4
diff -d -u -r1.31.2.4 extents.c
--- src/extents.c 2005/01/31 02:55:13 1.31.2.4
+++ src/extents.c 2006/06/22 23:33:14
＠＠ -1175,11 +1175,12 ＠＠
set_extent_start (e, -1);
set_extent_end (e, -1);
}
+
+ /* But we need to clear all the lists containing extents or
+ havoc will result. */
+ extent_list_delete_all (data->extents);
}
- /* But we need to clear all the lists containing extents or
- havoc will result. */
- extent_list_delete_all (data->extents);
soe_invalidate (object);
}
}
Index: src/glyphs-widget.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/glyphs-widget.c,v
retrieving revision 1.2.2.2
diff -d -u -r1.2.2.2 glyphs-widget.c
--- src/glyphs-widget.c 2002/11/07 06:04:44 1.2.2.2
+++ src/glyphs-widget.c 2006/06/22 23:33:14
＠＠ -558,20 +558,20 ＠＠
IMAGE_INSTANCE_WIDGET_FACE (ii),
&w, &h, 0, domain);
/* Adjust the size for borders. */
- if (IMAGE_INSTANCE_SUBWINDOW_H_RESIZEP (ii))
+ if (width && IMAGE_INSTANCE_SUBWINDOW_H_RESIZEP (ii))
*width = w + 2 * widget_instance_border_width (ii);
- if (IMAGE_INSTANCE_SUBWINDOW_V_RESIZEP (ii))
+ if (height && IMAGE_INSTANCE_SUBWINDOW_V_RESIZEP (ii))
*height = h + 2 * widget_instance_border_width (ii);
}
}
/* Finish off with dynamic sizing. */
- if (!NILP (IMAGE_INSTANCE_WIDGET_WIDTH_SUBR (ii)))
+ if (width && !NILP (IMAGE_INSTANCE_WIDGET_WIDTH_SUBR (ii)))
{
dynamic_width = Feval (IMAGE_INSTANCE_WIDGET_WIDTH_SUBR (ii));
if (INTP (dynamic_width))
*width = XINT (dynamic_width);
}
- if (!NILP (IMAGE_INSTANCE_WIDGET_HEIGHT_SUBR (ii)))
+ if (height && !NILP (IMAGE_INSTANCE_WIDGET_HEIGHT_SUBR (ii)))
{
dynamic_height = Feval (IMAGE_INSTANCE_WIDGET_HEIGHT_SUBR (ii));
if (INTP (dynamic_height))
Index: src/input-method-xlib.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/input-method-xlib.c,v
retrieving revision 1.12.2.1
diff -d -u -r1.12.2.1 input-method-xlib.c
--- src/input-method-xlib.c 2005/01/31 02:55:20 1.12.2.1
+++ src/input-method-xlib.c 2006/06/22 23:33:14
＠＠ -426,13 +426,18 ＠＠
void
XIM_SetGeometry (struct frame *f)
{
- XIC xic = FRAME_X_XIC (f);
- XIMStyle style = FRAME_X_XIC_STYLE (f);
+ XIC xic;
+ XIMStyle style;
XRectangle area;
- if (!xic || !f)
+ if (!f)
return;
+ xic = FRAME_X_XIC (f);
+ if (!xic)
+ return;
+
+ style = FRAME_X_XIC_STYLE (f);
if (style & XIMStatusArea)
{
/* Place Status Area in bottom right corner */
Index: src/keymap.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/keymap.c,v
retrieving revision 1.32.2.4
diff -d -u -r1.32.2.4 keymap.c
--- src/keymap.c 2005/01/31 02:55:20 1.32.2.4
+++ src/keymap.c 2006/06/22 23:33:15
＠＠ -3663,6 +3663,8 ＠＠
struct key_data *new = xnew_array (struct key_data, size);
memcpy ((void *)new, (const void *)c->keys_so_far,
c->keys_so_far_total_size * sizeof (struct key_data));
+ xfree (c->keys_so_far);
+ c->keys_so_far = new;
}
else
XREALLOC_ARRAY (c->keys_so_far, struct key_data, size);
Index: src/md5.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/md5.c,v
retrieving revision 1.9
diff -d -u -r1.9 md5.c
--- src/md5.c 2001/04/12 18:24:01 1.9
+++ src/md5.c 2006/06/22 23:33:15
＠＠ -581,7 +581,7 ＠＠
Bufbyte tempbuf[1024]; /* some random amount */
Lstream_data_count size_in_bytes =
Lstream_read (XLSTREAM (instream), tempbuf, sizeof (tempbuf));
- if (!size_in_bytes)
+ if (size_in_bytes <= 0)
break;
/* Process the bytes. */
Index: src/nas.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/nas.c,v
retrieving revision 1.7.2.2
diff -d -u -r1.7.2.2 nas.c
--- src/nas.c 2002/10/31 15:08:26 1.7.2.2
+++ src/nas.c 2006/06/22 23:33:15
＠＠ -751,7 +751,7 ＠＠
/* Stuff taken from wave.c from NAS. Just like snd files, NAS can't
read wave data from memory, so these functions do that for us. */
-#define Err() { return NULL; }
+#define Err() { free (wi); return NULL; }
#define readFourcc(_f) dread(_f, sizeof(RIFF_FOURCC), 1)
#define cmpID(_x, _y) \
strncmp((char *) (_x), (char *) (_y), sizeof(RIFF_FOURCC))
Index: src/redisplay-gtk.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/redisplay-gtk.c,v
retrieving revision 1.2.2.3
diff -d -u -r1.2.2.3 redisplay-gtk.c
--- src/redisplay-gtk.c 2005/01/31 02:55:27 1.2.2.3
+++ src/redisplay-gtk.c 2006/06/22 23:33:15
＠＠ -294,7 +294,7 ＠＠
int cursor_width, int cursor_height)
{
struct frame *f = XFRAME (w->frame);
- Emchar_dynarr *buf = Dynarr_new (Emchar);
+ Emchar_dynarr *buf;
Lisp_Object window;
struct display_block *db = Dynarr_atp (dl->display_blocks, block);
＠＠ -326,7 +326,7 ＠＠
if (end < 0)
end = Dynarr_length (rba);
- Dynarr_reset (buf);
+ buf = Dynarr_new (Emchar);
while (elt < end)
{
Index: src/redisplay-msw.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/redisplay-msw.c,v
retrieving revision 1.29.2.2
diff -d -u -r1.29.2.2 redisplay-msw.c
--- src/redisplay-msw.c 2005/01/31 02:55:27 1.29.2.2
+++ src/redisplay-msw.c 2006/06/22 23:33:15
＠＠ -1,4 +1,4 ＠＠
-/* mswindows output and frame manipulation routines.
+* mswindows output and frame manipulation routines.
Copyright (C) 1994, 1995 Board of Trustees, University of Illinois.
Copyright (C) 1994 Lucid, Inc.
Copyright (C) 1995 Sun Microsystems, Inc.
＠＠ -1044,7 +1044,7 ＠＠
int cursor_width, int cursor_height)
{
struct frame *f = XFRAME (w->frame);
- Emchar_dynarr *buf = Dynarr_new (Emchar);
+ Emchar_dynarr *buf;
Lisp_Object window;
struct display_block *db = Dynarr_atp (dl->display_blocks, block);
＠＠ -1071,7 +1071,7 ＠＠
if (end < 0)
end = Dynarr_length (rba);
- Dynarr_reset (buf);
+ buf = Dynarr_new (Emchar);
while (elt < end)
{
Index: src/redisplay-output.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/redisplay-output.c,v
retrieving revision 1.12.2.4
diff -d -u -r1.12.2.4 redisplay-output.c
--- src/redisplay-output.c 2005/01/31 02:55:28 1.12.2.4
+++ src/redisplay-output.c 2006/06/22 23:33:15
＠＠ -1317,7 +1317,7 ＠＠
{
Lisp_Image_Instance *p = XIMAGE_INSTANCE (image_instance);
Lisp_Object rest, window = DOMAIN_WINDOW (domain);
- Emchar_dynarr *buf = Dynarr_new (Emchar);
+ Emchar_dynarr *buf;
struct window *w = XWINDOW (window);
struct device *d = DOMAIN_XDEVICE (domain);
int layout_height, layout_width;
＠＠ -1333,6 +1333,8 ＠＠
/* This makes the glyph area fit into the display area. */
if (!redisplay_normalize_glyph_area (db, dga))
return;
+
+ buf = Dynarr_new (Emchar);
/* Highly dodgy optimization. We want to only output the whole
layout if we really have to. */
Index: src/redisplay-tty.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/redisplay-tty.c,v
retrieving revision 1.16.2.2
diff -d -u -r1.16.2.2 redisplay-tty.c
--- src/redisplay-tty.c 2005/11/28 02:10:20 1.16.2.2
+++ src/redisplay-tty.c 2006/06/22 23:33:15
＠＠ -196,7 +196,7 ＠＠
int cursor_height)
{
struct frame *f = XFRAME (w->frame);
- Emchar_dynarr *buf = Dynarr_new (Emchar);
+ Emchar_dynarr *buf;
struct display_block *db = Dynarr_atp (dl->display_blocks, block);
rune_dynarr *rba = db->runes;
＠＠ -222,7 +222,7 ＠＠
if (end < 0)
end = Dynarr_length (rba);
- Dynarr_reset (buf);
+ buf = Dynarr_new (Emchar);
while (elt < end && Dynarr_atp (rba, elt)->xpos < start_pixpos)
{
Index: src/redisplay-x.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/redisplay-x.c,v
retrieving revision 1.25.2.5
diff -d -u -r1.25.2.5 redisplay-x.c
--- src/redisplay-x.c 2005/01/31 02:55:28 1.25.2.5
+++ src/redisplay-x.c 2006/06/22 23:33:15
＠＠ -318,7 +318,7 ＠＠
int cursor_width, int cursor_height)
{
struct frame *f = XFRAME (w->frame);
- Emchar_dynarr *buf = Dynarr_new (Emchar);
+ Emchar_dynarr *buf;
Lisp_Object window;
struct display_block *db = Dynarr_atp (dl->display_blocks, block);
＠＠ -345,7 +345,7 ＠＠
if (end < 0)
end = Dynarr_length (rba);
- Dynarr_reset (buf);
+ buf = Dynarr_new (Emchar);
while (elt < end)
{
Index: src/scrollbar-gtk.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/scrollbar-gtk.c,v
retrieving revision 1.2.2.4
diff -d -u -r1.2.2.4 scrollbar-gtk.c
--- src/scrollbar-gtk.c 2005/01/31 02:55:30 1.2.2.4
+++ src/scrollbar-gtk.c 2006/06/22 23:33:15
＠＠ -57,14 +57,16 ＠＠
static void
gtk_free_scrollbar_instance (struct scrollbar_instance *instance)
{
- if (SCROLLBAR_GTK_WIDGET (instance))
+ if (instance->scrollbar_data)
{
- gtk_widget_hide_all (SCROLLBAR_GTK_WIDGET (instance));
- gtk_widget_destroy (SCROLLBAR_GTK_WIDGET (instance));
- }
+ if (SCROLLBAR_GTK_WIDGET (instance))
+ {
+ gtk_widget_hide_all (SCROLLBAR_GTK_WIDGET (instance));
+ gtk_widget_destroy (SCROLLBAR_GTK_WIDGET (instance));
+ }
- if (instance->scrollbar_data)
- xfree (instance->scrollbar_data);
+ xfree (instance->scrollbar_data);
+ }
}
/* A device method. */
Index: src/scrollbar-msw.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/scrollbar-msw.c,v
retrieving revision 1.11.2.5
diff -d -u -r1.11.2.5 scrollbar-msw.c
--- src/scrollbar-msw.c 2005/01/31 02:55:31 1.11.2.5
+++ src/scrollbar-msw.c 2006/06/22 23:33:15
＠＠ -83,9 +83,11 ＠＠
static void
mswindows_free_scrollbar_instance (struct scrollbar_instance *sb)
{
- DestroyWindow (SCROLLBAR_MSW_HANDLE (sb));
if (sb->scrollbar_data)
- xfree (sb->scrollbar_data);
+ {
+ DestroyWindow (SCROLLBAR_MSW_HANDLE (sb));
+ xfree (sb->scrollbar_data);
+ }
}
static void
Index: src/scrollbar-x.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/scrollbar-x.c,v
retrieving revision 1.17.2.1
diff -d -u -r1.17.2.1 scrollbar-x.c
--- src/scrollbar-x.c 2005/01/31 02:55:31 1.17.2.1
+++ src/scrollbar-x.c 2006/06/22 23:33:15
＠＠ -71,19 +71,21 ＠＠
static void
x_free_scrollbar_instance (struct scrollbar_instance *instance)
{
- if (SCROLLBAR_X_NAME (instance))
- xfree (SCROLLBAR_X_NAME (instance));
-
- if (SCROLLBAR_X_WIDGET (instance))
+ if (instance->scrollbar_data)
{
- if (XtIsManaged (SCROLLBAR_X_WIDGET (instance)))
- XtUnmanageChild (SCROLLBAR_X_WIDGET (instance));
+ if (SCROLLBAR_X_NAME (instance))
+ xfree (SCROLLBAR_X_NAME (instance));
- lw_destroy_all_widgets (SCROLLBAR_X_ID (instance));
- }
+ if (SCROLLBAR_X_WIDGET (instance))
+ {
+ if (XtIsManaged (SCROLLBAR_X_WIDGET (instance)))
+ XtUnmanageChild (SCROLLBAR_X_WIDGET (instance));
- if (instance->scrollbar_data)
- xfree (instance->scrollbar_data);
+ lw_destroy_all_widgets (SCROLLBAR_X_ID (instance));
+ }
+
+ xfree (instance->scrollbar_data);
+ }
}
/* A device method. */
Index: src/sound.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/sound.c,v
retrieving revision 1.12
diff -d -u -r1.12 sound.c
--- src/sound.c 2001/04/12 18:24:19 1.12
+++ src/sound.c 2006/06/22 23:33:15
＠＠ -517,7 +517,7 ＠＠
DEVICE_ON_CONSOLE_P (d) = 0;
else
{
- char hn [255];
+ char *hn = alloca_array (char, strlen (h->h_name) + 1);
struct hostent *l;
strcpy (hn, h->h_name);
l = gethostbyname (localname);
--
Jerry James, Assistant Professor james(a)xemacs.org
Computer Science Department http://www.cs.usu.edu/~jerry/
Utah State University
- 2 participants
- 2 comments
[PATCH 21.4] Add line-beginning-position, line-end-position
19 years
Ville Skyttä
Ville Skyttä
21.4 (already in 21.5)
This patch adds line-beginning-position and line-end-position to the
21.4 series. We've seen these being erratically used here and there in
packages a few times and lisp things out there outside of the packages
tree uses these functions pretty often, I guess they're thought of as
ubiquitous by many.
The latest problem due to lack of these functions I'm aware of was
reported as https://bugzilla.redhat.com/192481 : the current
haskell-mode's indentation is broken on 21.4.
Jerry, in case you're around, it could be a good idea to fix
haskell-*indent.el to not use line-(beginning|end)-position in the
meantime.
- 4 participants
- 6 comments
[AC21.4] Fix tiny typo in redisplay-msw.c
19 years
Vin Shelton
Vin Shelton
I commited Jerry's dynarr patch for 21.4 and came upon this small
gotcha.
This patch has been committed.
- Vin
src/ChangeLog addition:
2006-07-04 Vin Shelton <acs(a)xemacs.org>
* redisplay-msw.c: Fix tiny, but crucial typo.
21.4 source patch:
Diff command: cvs -q diff -u
Files affected: src/redisplay-msw.c
Index: src/redisplay-msw.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/redisplay-msw.c,v
retrieving revision 1.29.2.3
diff -a -u -u -r1.29.2.3 redisplay-msw.c
--- src/redisplay-msw.c 2006/07/01 05:19:59 1.29.2.3
+++ src/redisplay-msw.c 2006/07/04 13:09:11
＠＠ -1,4 +1,4 ＠＠
-* mswindows output and frame manipulation routines.
+/* mswindows output and frame manipulation routines.
Copyright (C) 1994, 1995 Board of Trustees, University of Illinois.
Copyright (C) 1994 Lucid, Inc.
Copyright (C) 1995 Sun Microsystems, Inc.
- 1 participants
- 0 comments
[21.5] Quiet GCC 4 whining about unused values
19 years
Stephen J. Turnbull
Stephen J. Turnbull
21.5
Marcus, your opinion would be appreciated on this one. Seems
"obviously correct" to me, but ....
GCC 4 whines about unused values, and it's right. I think this patch
actually makes the code clearer, as well.
diff --git a/src/ChangeLog b/src/ChangeLog
index b9f6993..3fa529a 100644
--- a/src/ChangeLog
+++ b/src/ChangeLog
＠＠ -0,0 +1,4 ＠＠
+2006-06-24 Stephen J. Turnbull <stephen(a)xemacs.org>
+
+ * gc.c (gc_mark_root_set): Quiet GCC 4 whining about unused values.
+
diff --git a/src/gc.c b/src/gc.c
index 972821a..18f0944 100644
--- a/src/gc.c
+++ b/src/gc.c
＠＠ -1623,29 +1623,25 ＠＠ gc_mark_root_set (
{ /* staticpro() */
Lisp_Object **p = Dynarr_begin (staticpros);
Elemcount count;
- for (count = Dynarr_length (staticpros); count; count--)
+ for (count = Dynarr_length (staticpros); count; count--, p++)
/* Need to check if the pointer in the staticpro array is not
NULL. A gc can occur after variable is added to the staticpro
array and _before_ it is correctly initialized. In this case
its value is NULL, which we have to catch here. */
if (*p)
- mark_object (**p++);
- else
- **p++;
+ mark_object (**p);
}
{ /* staticpro_nodump() */
Lisp_Object **p = Dynarr_begin (staticpros_nodump);
Elemcount count;
- for (count = Dynarr_length (staticpros_nodump); count; count--)
+ for (count = Dynarr_length (staticpros_nodump); count; count--, p++)
/* Need to check if the pointer in the staticpro array is not
NULL. A gc can occur after variable is added to the staticpro
array and _before_ it is correctly initialized. In this case
its value is NULL, which we have to catch here. */
if (*p)
- mark_object (**p++);
- else
- **p++;
+ mark_object (**p);
}
#ifdef NEW_GC
--
Graduate School of Systems and Information Engineering University of Tsukuba
http://turnbull.sk.tsukuba.ac.jp/ Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Economics of Information Communication and Computation Systems
Experimental Economics, Microeconomic Theory, Game Theory
- 2 participants
- 1 comments