10:44 p.m.
On 20 Jan 2007, David Kastrup outgrape:
For all that I know, specifying binary format image file names in an
image descriptor is _still_ blowing up XEmacs 21.4 (and likely 21.5,
too) once dired gets loaded. This bug was reported at least 4 years
ago IIRC. Let me take a look:
Indeed. Here's the patch again, against fairly recent 21.5 CVS.
2005-02-04 Nix <nix(a)esperi.org.uk>
* fileio.c (restore_read_coding_system_unwind): Pop the coding
system.
(Finsert_file_contents_internal): Use the CODESYS even for files
read by a file-name-handler.
Index: src/fileio.c
===================================================================
--- src/fileio.c.orig 2006-12-16 21:20:49.000000000 +0000
+++ src/fileio.c 2006-12-16 21:46:42.000000000 +0000
＠＠ -216,6 +216,15 ＠＠
return Fset_marker (point_marker, Qnil, Qnil);
}
+/* Restore coding-system-for-read. */
+
+static Lisp_Object
+restore_read_coding_system_unwind (Lisp_Object coding_system)
+{
+ Vcoding_system_for_read = coding_system;
+ return Qnil;
+}
+
Lisp_Object Qexpand_file_name;
Lisp_Object Qfile_truename;
＠＠ -3360,6 +3369,9 ＠＠
multiple return points make this a pain in the butt. ]] we do
protect curbuf now. --ben */
+ record_unwind_protect (restore_read_coding_system_unwind, Vcoding_system_for_read);
+ Vcoding_system_for_read = codesys;
+
codesys = get_coding_system_for_text_file (codesys, 0);
if (current_buffer->base_buffer && ! NILP (visit))
--
`The serial comma, however, is correct and proper, and abandoning it will
surely lead to chaos, anarchy, rioting in the streets, the Terrorists
taking over, and possibly the complete collapse of Human Civilization.'
_______________________________________________
XEmacs-Patches mailing list
XEmacs-Patches(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-patches
5:38 p.m.
New subject: [Q] respect CODESYS in `insert-file-contents-internal'.
QUERY
Nix writes:
Indeed. Here's the patch again, against fairly recent 21.5 CVS.
2005-02-04 Nix <nix(a)esperi.org.uk>
* fileio.c (restore_read_coding_system_unwind): Pop the coding
system.
(Finsert_file_contents_internal): Use the CODESYS even for files
read by a file-name-handler.
Eh, I trust you, but could you like provide me a clue? I mean, why do
you need to set Vcoding_system_for_read in a *write* function?
(I hate to tell you this, but I'm willing to bet I took one look at
this magic and decided that Ben really needed to look at it just
before he grabbed the loot and vamoosed to Rio de Janeiro ... and
that's why it never got applied, because it's got that Gorgon look.)
It would also nice to have a simple recipe to reproduce, something
that we could add to the XEmacs-killer regression test. (I'll get
around to working backwards from David's comments eventually, but that
could be a lot of events at current rate of progress....)
cheers
_______________________________________________
XEmacs-Patches mailing list
XEmacs-Patches(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-patches
6 p.m.
New subject: [Q] respect CODESYS in `insert-file-contents-internal'.
On 26 Jan 2007, Stephen J. Turnbull verbalised:
QUERY
Nix writes:
> Indeed. Here's the patch again, against fairly recent 21.5 CVS.
>
> 2005-02-04 Nix <nix(a)esperi.org.uk>
>
> * fileio.c (restore_read_coding_system_unwind): Pop the coding
> system.
> (Finsert_file_contents_internal): Use the CODESYS even for files
> read by a file-name-handler.
Eh, I trust you, but could you like provide me a clue? I mean, why do
you need to set Vcoding_system_for_read in a *write* function?
(Hang on, did the line numbers get frotzed in there?
insert-file-contents-internal isn't a write function!)
... hm. Patch fuzz, most profound apologies: you're right that the patch
as supplied makes no sense... and upon further inspection the respinning
of more of insert_file_contents_internal into Lisp has fixed this
problem as a side-effect. So it's no longer a bug in 21.5. Sorry,
discard the patch by all means.
(The underlying problem in 21.4 was that insert_file_contents didn't
actually *change* the coding system in use to that specified by the
CODESYS, but *did* call the file-name-handlers, so the handlers ended up
applying whatever the default coding system was. For raw binary data,
e.g. that specified in the :data part of an image specifier, this was
often a very bad thing.)
--
`The serial comma, however, is correct and proper, and abandoning it will
surely lead to chaos, anarchy, rioting in the streets, the Terrorists
taking over, and possibly the complete collapse of Human Civilization.'
_______________________________________________
XEmacs-Patches mailing list
XEmacs-Patches(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-patches
7:01 p.m.
New subject: [Q] respect CODESYS in `insert-file-contents-internal'.
Nix writes:
(The underlying problem in 21.4 was that insert_file_contents
didn't
actually *change* the coding system in use to that specified by the
CODESYS, but *did* call the file-name-handlers,
OK. I have doubt that's been fixed in 21.4 yet, but I'm pretty sure I
can find the appropriate place, and probably your old patch in the
archives. I assume your recommendation is to apply a similar patch to
21.4.
I notice in looking at insert-file-contents-internal in 21.4 that if a
handler is called, then the "goto handled" jumps past all of the
unbind_to (speccount, Qnil) calls. That would seem to be a horrible
error, isn't that so? Am I missing something?
_______________________________________________
XEmacs-Patches mailing list
XEmacs-Patches(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-patches
8:12 p.m.
New subject: [Q] respect CODESYS in `insert-file-contents-internal'.
On 27 Jan 2007, Stephen J. Turnbull spake thusly:
Nix writes:
> (The underlying problem in 21.4 was that insert_file_contents didn't
> actually *change* the coding system in use to that specified by the
> CODESYS, but *did* call the file-name-handlers,
OK. I have doubt that's been fixed in 21.4 yet, but I'm pretty sure I
can find the appropriate place, and probably your old patch in the
archives. I assume your recommendation is to apply a similar patch to
21.4.
Yes, that would be a good idea. (The patch applies fine, it just needs
some shifting around. The old patch in the archives should apply without
changes.)
I notice in looking at insert-file-contents-internal in 21.4 that if
a
handler is called, then the "goto handled" jumps past all of the
unbind_to (speccount, Qnil) calls. That would seem to be a horrible
error, isn't that so? Am I missing something?
I don't think you are. I'd say inserting an unbind_to() at the jump site
is probably easiest, given that you can't move the unbind_to() sitting
just before handled: down to below it without fubaring the I/O error
handling.
--
`The serial comma, however, is correct and proper, and abandoning it will
surely lead to chaos, anarchy, rioting in the streets, the Terrorists
taking over, and possibly the complete collapse of Human Civilization.'
_______________________________________________
XEmacs-Patches mailing list
XEmacs-Patches(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-patches
6539
days inactive
6544
days old
4 comments
2 participants
participants (2)
-
Nix -
Stephen J. Turnbull