string-match in unicode-2 (Re: Calc broken in Emacs unicode-2 branch)

string-match in unicode-2 (Re: Calc broken in Emacs unicode-2 branch)

[Jay Belanger]

Kenichi Handa <handa@m17n.org> writes:
..
> I'm sorry for the late response.

And I'm sorry for mine.  (I responded to Leo offlist, but I should
have done it on the list.)

> Jay, it seems that you are the maintainer of this package.
> Could you please work on it with emacs-unicode-2 branch?

Yes.  The error seems related to this:
  (string-match "⁸" "x")
(The first string is a superscript "8".)  
It returns nil in the normal cvs emacs, but 0 in the unicode-2
branch.  I don't know why it doesn't return nil in the unicode-2
branch; perhaps someone can enlighten me.
For what it's worth, C-x= on the characters returns
in cvs
Char: ⁸ (342520, #o1234770, #x539f8, file ...) point=17 of 24 (67%) column=15
Char: x (120, #o170, #x78) point=21 of 24 (83%) column=19
in unicode-2
Char: ⁸ (8312, #o20170, #x2078, file ...) point=17 of 24 (67%) column=15
Char: x (120, #o170, #x78) point=21 of 24 (83%) column=19

Jay

Re: string-match in unicode-2 (Re: Calc broken in Emacs unicode-2 branch)

[Kenichi Handa]

In article <87ejq0rpxo.fsf_-_@vh213602.truman.edu>, Jay Belanger <belanger@truman.edu> writes:

> Yes.  The error seems related to this:
>   (string-match "⁸" "x")
> (The first string is a superscript "8".)  
> It returns nil in the normal cvs emacs, but 0 in the unicode-2
> branch.

Thank you for tracking down the bug to that code.  As I
found a bug of multibyteness handling in regex code, I'm now
working on fixing it.

---
Kenichi Handa
handa@m17n.org

Re: string-match in unicode-2 (Re: Calc broken in Emacs unicode-2 branch)

[Jay Belanger]

Kenichi Handa <handa@m17n.org> writes:
...
> Thank you for tracking down the bug to that code.  As I
> found a bug of multibyteness handling in regex code, I'm now
> working on fixing it.

Great.  Meanwhile, as a temporary fix for Calc, adding 

 (add-hook 'calc-ext-load-hook
           (lambda ()
             (defun math-read-preprocess-string (str)
               str)))

to the user's emacs init file should get Calc working again.

Jay

Re: string-match in unicode-2 (Re: Calc broken in Emacs unicode-2 branch)

[Kenichi Handa]

In article <87y7o5z7rv.fsf@vh213602.truman.edu>, Jay Belanger <belanger@truman.edu> writes:

> Kenichi Handa <handa@m17n.org> writes:
...
> > Thank you for tracking down the bug to that code.  As I
> > found a bug of multibyteness handling in regex code, I'm now
> > working on fixing it.

I've just installed a fix for emacs-unicode-2 branch.

---
Kenichi Handa
handa@m17n.org

Re: string-match in unicode-2 (Re: Calc broken in Emacs unicode-2 branch)

[Juanma Barranquero]

[-- Attachment #1: Type: text/plain, Size: 1076 bytes --]

On 2/15/07, Kenichi Handa <handa@m17n.org> wrote:

> I've just installed a fix for emacs-unicode-2 branch.

The unicode-2 branch fails to compile.  I think the attached patch is
needed, but even with it, I'm getting this error:

In toplevel form:
ja-dic/ja-dic.el:46:1:Error: Args out of range: "ã??s æ??", 1663, 1663
make[2]: *** [ja-dic/ja-dic.elc] Error 1

             Juanma


Index: src/regex.c
===================================================================
RCS file: /sources/emacs/emacs/src/regex.c,v
retrieving revision 1.186.4.20
diff -u -2 -r1.186.4.20 regex.c
--- src/regex.c	15 Feb 2007 11:23:52 -0000	1.186.4.20
+++ src/regex.c	15 Feb 2007 12:22:50 -0000
@@ -312,5 +312,5 @@
 # define CHAR_STRING(c, s) (*(s) = (c), 1)
 # define STRING_CHAR_AND_LENGTH(p, s, actual_len) ((actual_len) = 1, *(p))
-# define RE_STRING_CHAR_AND_LENGTH(p, s, multibyte)
STRING_CHAR_AND_LENGTH ((p), (s))
+# define RE_STRING_CHAR_AND_LENGTH(p, s, len, multibyte)
STRING_CHAR_AND_LENGTH ((p), (s), (len))
 # define RE_CHAR_TO_MULTIBYTE(c) (c)
 # define RE_CHAR_TO_UNIBYTE(c) (c)

[-- Attachment #2: Type: text/plain, Size: 142 bytes --]

_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-devel

Re: string-match in unicode-2 (Re: Calc broken in Emacs unicode-2 branch)

[Leo]

On 2007-02-15, Kenichi Handa said:

>> > Thank you for tracking down the bug to that code.  As I found a
>> > bug of multibyteness handling in regex code, I'm now working on
>> > fixing it.
>
> I've just installed a fix for emacs-unicode-2 branch.

I have come back to emacs-unicode-2 branch. Indeed, it is fixed and so
is the calc bug.

-- 
Leo <sdl.web AT gmail.com>                         (GPG Key: 9283AA3F)

Re: string-match in unicode-2 (Re: Calc broken in Emacs unicode-2 branch)

[Kenichi Handa]

In article <m2fy7pat4e.fsf@sl392.st-edmunds.cam.ac.uk>, Leo <sdl.web@gmail.com> writes:

> On 2007-02-15, Kenichi Handa said:
>>> > Thank you for tracking down the bug to that code.  As I found a
>>> > bug of multibyteness handling in regex code, I'm now working on
>>> > fixing it.
> >
> > I've just installed a fix for emacs-unicode-2 branch.

> I have come back to emacs-unicode-2 branch. Indeed, it is fixed and so
> is the calc bug.

??? As far as I remember, the problem of calc was because of
the bug of regex.

Do you mean that calc still has a problem even after the bug
of regex was fixed?

---
Kenichi Handa
handa@m17n.org

Re: string-match in unicode-2 (Re: Calc broken in Emacs unicode-2 branch)

[Leo]

Hello, Kenichi!

On 2007-03-29, Kenichi Handa said:
>>>> > Thank you for tracking down the bug to that code.  As I found a
>>>> > bug of multibyteness handling in regex code, I'm now working on
>>>> > fixing it.
>> >
>> > I've just installed a fix for emacs-unicode-2 branch.
>
>> I have come back to emacs-unicode-2 branch. Indeed, it is fixed and so
>> is the calc bug.
>
> ??? As far as I remember, the problem of calc was because of
> the bug of regex.
>
> Do you mean that calc still has a problem even after the bug
> of regex was fixed?

It runs smoothly now ;) No problem so far.

regards,
-- 
Leo <sdl.web AT gmail.com>                         (GPG Key: 9283AA3F)