5:04 p.m.
This should resume normal service. I have also fixed the elhash stuff.
andy
2000-07-31 Andy Piper <andy(a)xemacs.org>
* buffer.c (Frecord_buffer): make absolutely sure that redisplay
will acknowledge the change.
* glyphs.h: declare tab_control_order_only_changed.
* glyphs-x.c (x_tab_control_redisplay): use
tab_control_order_only_changed.
* glyphs-widget.c (tab_control_order_only_changed): new function.
* glyphs-msw.c (mswindows_tab_control_redisplay): use
tab_control_order_only_changed.
* gui.c (gui_item_equal_sans_selected): new function.
(gui_item_equal): use it.
* glyphs-msw.c (mswindows_combo_box_instantiate): deprecate
:properties in favor of :items..
* glyphs-widget.c (check_valid_item_list): rename from
check_valid_item_list_1.
(check_valid_item_list_1): renamed.
(combo_box_validate): deprecate :properties in favor of :items.
(widget_instantiate): ditto.
(tab_control_update): ditto.
(image_instantiator_combo_box): ditto.
(image_instantiator_tree_view): ditto.
(image_instantiator_tab_control): ditto.
(layout_post_instantiate): remove dead code.
* print.c (debug_print_no_newline): only write to debugger if in
WIN32_NATIVE.
* elhash.c (Fmake_hash_table): update doc string.
* event-msw.c (mswindows_wnd_proc): don't allow processing of
messages whilst in GC. This at least stops XEmacs crashing but has
the potential for wierd behaviour.
2000-07-31 Andy Piper <andy(a)xemacs.org>
* gutter-items.el (update-tab-in-gutter): deprecate :properties.
2000-07-27 Andy Piper <andy(a)xemacs.org>
* lispref/hash-tables.texi: add new hash table type.
--------------------------------------------------------------
Dr Andy Piper
Principal Consultant, BEA Systems Ltd
2:27 a.m.
i still think the correct solution is just to postpone the calls to
DestroyWindow until after the finalize method.
Index: src/event-msw.c
===================================================================
RCS file: /usr/CVSroot/XEmacs/xemacs/src/event-msw.c,v
retrieving revision 1.38.2.54
diff -u -r1.38.2.54 event-msw.c
--- src/event-msw.c 2000/07/21 10:16:02 1.38.2.54
+++ src/event-msw.c 2000/07/31 15:58:47
＠＠ -2023,6 +2023,11 ＠＠
struct frame *frame;
struct mswindows_frame* msframe;
+ /* Not perfect but avoids crashes. There is potential for wierd
+ behaviour here. */
+ if (gc_in_progress)
+ goto defproc;
+
assert (!GetWindowLong (hwnd, GWL_USERDATA));
switch (message_)
{
Andy Piper wrote:
This should resume normal service. I have also fixed the elhash stuff.
andy
2000-07-31 Andy Piper <andy(a)xemacs.org>
* buffer.c (Frecord_buffer): make absolutely sure that redisplay
will acknowledge the change.
* glyphs.h: declare tab_control_order_only_changed.
* glyphs-x.c (x_tab_control_redisplay): use
tab_control_order_only_changed.
* glyphs-widget.c (tab_control_order_only_changed): new function.
* glyphs-msw.c (mswindows_tab_control_redisplay): use
tab_control_order_only_changed.
* gui.c (gui_item_equal_sans_selected): new function.
(gui_item_equal): use it.
* glyphs-msw.c (mswindows_combo_box_instantiate): deprecate
:properties in favor of :items..
* glyphs-widget.c (check_valid_item_list): rename from
check_valid_item_list_1.
(check_valid_item_list_1): renamed.
(combo_box_validate): deprecate :properties in favor of :items.
(widget_instantiate): ditto.
(tab_control_update): ditto.
(image_instantiator_combo_box): ditto.
(image_instantiator_tree_view): ditto.
(image_instantiator_tab_control): ditto.
(layout_post_instantiate): remove dead code.
* print.c (debug_print_no_newline): only write to debugger if in
WIN32_NATIVE.
* elhash.c (Fmake_hash_table): update doc string.
* event-msw.c (mswindows_wnd_proc): don't allow processing of
messages whilst in GC. This at least stops XEmacs crashing but has
the potential for wierd behaviour.
2000-07-31 Andy Piper <andy(a)xemacs.org>
* gutter-items.el (update-tab-in-gutter): deprecate :properties.
2000-07-27 Andy Piper <andy(a)xemacs.org>
* lispref/hash-tables.texi: add new hash table type.
--------------------------------------------------------------------------------
Name: tabfix.patch
tabfix.patch Type: Plain Text (text/plain)
Encoding: quoted-printable
--------------------------------------------------------------------------------
--------------------------------------------------------------
Dr Andy Piper
Principal Consultant, BEA Systems Ltd
--
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/
8:51 a.m.
At 06:27 PM 7/31/00 -0700, Ben Wing wrote:
i still think the correct solution is just to postpone the calls to
DestroyWindow until after the finalize method.
You are absolutely correct, but I wanted to get something in there that
will at least guard against crashes, postponing the calls is more
complicated and there's no telling that that won't be the only thing that
needs postponing.
andy
--------------------------------------------------------------
Dr Andy Piper
Principal Consultant, BEA Systems Ltd
4:22 p.m.
-----Original Message-----
From: owner-xemacs-patches(a)xemacs.org
[mailto:owner-xemacs-patches＠xemacs.org]On Behalf Of Ben Wing
Sent: 01 August 2000 02:27
To: Andy Piper
Cc: xemacs-patches(a)xemacs.org; xemacs-beta(a)xemacs.org
Subject: Re: Fix tab lossage
i still think the correct solution is just to postpone the calls to
DestroyWindow until after the finalize method.
Me too. Though perhaps we should drop an assert into the
window procedure so it dies if the wndproc is running during
GC, then we could detect any other situations that could
be problematic.
Kind regards,
Alastair.
____________________________________________________________
Alastair Houghton ajhoughton(a)lineone.net
8662
days inactive
8663
days old
3 comments
3 participants
participants (3)
-
Alastair Houghton -
Andy Piper -
Ben Wing