10:10 p.m.
Somebody please commit it, I changed ssh key recently (to go from ssh1
to ssh2) and haven't bothered Stephen with it yet.
OG.
2007-03-24 Olivier Galibert <galibert(a)pobox.com>
* chartab.c (Fcopy_char_table): Simplify the mirror table
handling: never copy a mirror table, just make a new one and mark
it dirty. Fixes http://article.gmane.org/gmane.emacs.xemacs.beta/17353
_______________________________________________
XEmacs-Patches mailing list
XEmacs-Patches(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-patches
4:49 p.m.
COMMIT 21.5
Per Olivier's request.
I'm curious: Why does this fix the problem? It's "obviously correct,"
but I would think this would be a lazy equivalent of an explicit copy.
Was the wrong thing getting copied, and then left in place because the
dirty flag wasn't set?
Olivier Galibert writes:
Somebody please commit it, I changed ssh key recently (to go from
ssh1
to ssh2) and haven't bothered Stephen with it yet.
OG.
2007-03-24 Olivier Galibert <galibert(a)pobox.com>
* chartab.c (Fcopy_char_table): Simplify the mirror table
handling: never copy a mirror table, just make a new one and mark
it dirty. Fixes http://article.gmane.org/gmane.emacs.xemacs.beta/17353
_______________________________________________
XEmacs-Patches mailing list
XEmacs-Patches(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-patches
5:04 p.m.
New subject: [AC21.5] Add regression test for: Fix copy-char-table on syntax tables
APPROVE COMMIT 21.5
Add a regression test for
2007-03-24 Olivier Galibert <galibert(a)pobox.com>
* chartab.c (Fcopy_char_table): Simplify the mirror table
handling: never copy a mirror table, just make a new one and mark
it dirty. Fixes http://article.gmane.org/gmane.emacs.xemacs.beta/17353
Index: tests/ChangeLog
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/tests/ChangeLog,v
retrieving revision 1.86
diff -u -r1.86 ChangeLog
--- tests/ChangeLog 25 Nov 2006 22:06:29 -0000 1.86
+++ tests/ChangeLog 25 Mar 2007 15:52:25 -0000
＠＠ -0,0 +1,6 ＠＠
+2007-03-26 Stephen J. Turnbull <stephen(a)xemacs.org>
+
+ * automated/syntax-tests.el: Test for regression of bug fixed by
+ Olivier Galibert <20070324221053.GA48218(a)dspnet.fr.eu.org>. Test
+ by Ralf Angeli http://article.gmane.org/gmane.emacs.xemacs.beta/17353.
+
Index: tests/automated/syntax-tests.el
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/tests/automated/syntax-tests.el,v
retrieving revision 1.6
diff -u -r1.6 syntax-tests.el
--- tests/automated/syntax-tests.el 8 Dec 2005 10:58:59 -0000 1.6
+++ tests/automated/syntax-tests.el 25 Mar 2007 15:52:25 -0000
＠＠ -129,6 +129,22 ＠＠
;; Apply word to punctuation
(test-syntax-table "W." 1 `(,(syntax-string-to-code "w")) 2))
+;; According to Ralf Angeli in
+;; http://article.gmane.org/gmane.emacs.xemacs.beta/17353:
+;; Using a fresh CVS checkout of XEmacs trunk the following snippet
+;; returns "1" when evaluated whereas it returns "5" in GNU Emacs
21.3,
+;; CVS GNU Emacs and XEmacs 21.4.15.
+;; If `set-syntax-table' is used instead of `with-syntax-table', CVS
+;; XEmacs returns "5" as well, so I suppose that there is a problem in
+;; `with-syntax-table' or a function called by it.
+
+;; Fixed 2007-03-25 Olivier Galibert <20070324221053.GA48218(a)dspnet.fr.eu.org>
+(with-temp-buffer
+ (with-syntax-table (make-syntax-table)
+ (insert "foo bar")
+ (backward-sexp 1)
+ (Assert (eql (point) 5))))
+
;; Test forward-comment at buffer boundaries
;; #### The second Assert fails (once interpreted, once compiled) on 21.4.9
;; with sjt's version of Andy's syntax-text-property-killer patch.
_______________________________________________
XEmacs-Patches mailing list
XEmacs-Patches(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-patches
6:12 p.m.
New subject: [AC21.5] Fix fix copy-char-table on syntax tables
APPROVE COMMIT 21.5
Built the wrong workspace. Need to use proper macro to compare Lisp
objects or you get an error in lisp-union builds.
Index: src/ChangeLog
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/ChangeLog,v
retrieving revision 1.1050
diff -u -U0 -r1.1050 ChangeLog
--- src/ChangeLog 25 Mar 2007 15:34:59 -0000 1.1050
+++ src/ChangeLog 25 Mar 2007 16:39:12 -0000
＠＠ -0,0 +1,4 ＠＠
+2007-03-26 Stephen J. Turnbull <stephen(a)xemacs.org>
+
+ * chartab.c (Fcopy_char_table): Use EQ, not ==.
+
Index: src/chartab.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/chartab.c,v
retrieving revision 1.40
diff -u -r1.40 chartab.c
--- src/chartab.c 25 Mar 2007 15:35:06 -0000 1.40
+++ src/chartab.c 25 Mar 2007 16:39:19 -0000
＠＠ -698,7 +698,7 ＠＠
#endif /* MULE */
- if (ct->mirror_table != Qnil)
+ if (!EQ (ct->mirror_table, Qnil))
{
ctnew->mirror_table = Fmake_char_table (Qgeneric);
set_char_table_default (ctnew->mirror_table, make_int (Sword));
_______________________________________________
XEmacs-Patches mailing list
XEmacs-Patches(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-patches
8:14 p.m.
On Mon, Mar 26, 2007 at 12:49:47AM +0900, Stephen J. Turnbull wrote:
COMMIT 21.5
Per Olivier's request.
I'm curious: Why does this fix the problem? It's "obviously correct,"
but I would think this would be a lazy equivalent of an explicit copy.
Was the wrong thing getting copied, and then left in place because the
dirty flag wasn't set?
To be honest, I haven't checked. But I suspect you have it right, the
dirty flag wasn't being copied which was a bug in any case. I first
found out that unconditionally setting the dirty flag on copy killed
the bug. Given that copying the reference to the mirror table in some
cases looked extremely dubious in the first place, I decided to just
simplify it. The complexity came in part from using cct to copy the
mirror table itself, so I nixed that and the result was much nicer.
OG.
_______________________________________________
XEmacs-Patches mailing list
XEmacs-Patches(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-patches
8:16 p.m.
New subject: [AC21.5] Fix fix copy-char-table on syntax tables
On Mon, Mar 26, 2007 at 02:12:00AM +0900, Stephen J. Turnbull wrote:
APPROVE COMMIT 21.5
Built the wrong workspace. Need to use proper macro to compare Lisp
objects or you get an error in lisp-union builds.
Ah damn. Something to remember for next time.
OG.
_______________________________________________
XEmacs-Patches mailing list
XEmacs-Patches(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-patches
6249
days inactive
6250
days old
5 comments
3 participants
participants (3)
-
Olivier Galibert -
Stephen J. Turnbull -
Stephen J. Turnbull