2 p.m.
Hello
I am using to a large extend the wonderful babel.el which accesses
translation service such as translate.google.com
However recently translate.google.com changed its syntax and so had
babel.el.
In turns out that Xemacs (21.4.21. Or 21.5.18) cannot deal correctly
with the ' any more. (While GNU emacs can!)
Look at the following example:
Ich war heute in Paris
--->
J'étais aujourd'hui à Paris
instead of
J'étais aujourd'hui à Paris
The relevant code line is
(decode-coding-region (point-min) (point-max) 'utf-8)
Does anybody have an idea?
Uwe Brauer
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
4:46 p.m.
Uwe Brauer wrote:
In turns out that Xemacs (21.4.21. Or 21.5.18) cannot deal
correctly
with the ' any more. (While GNU emacs can!)
Look at the following example:
Ich war heute in Paris
--->
J'étais aujourd'hui à Paris
instead of
J'étais aujourd'hui à Paris
The relevant code line is
(decode-coding-region (point-min) (point-max) 'utf-8)
Does anybody have an idea?
Uwe Brauer
The single quote is
being rendered as an HTML character entity. This must be an encoding
that XEmacs knows about. However, looking over (coding-system-list),
I can't identify it.
--
Rodney Sparapani Center for Patient Care & Outcomes Research (PCOR)
Sr. Biostatistician http://www.mcw.edu/pcor
4 wheels good, 2 wheels better! Medical College of Wisconsin (MCW)
WWLD?: What Would Lombardi Do? Milwaukee, WI, USA
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
7:45 p.m.
Rodney Sparapani writes:
> J'étais aujourd'hui à Paris
The single quote is being rendered as an HTML character entity.
That's right.
This must be an encoding that XEmacs knows about.
No, at least not the base XEmacs. It's an SGML thingee, and is needed
for the purpose of including literal SGML tags in SGML documents, and
has nothing to do with coding systems per se.
However, looking over (coding-system-list), I can't identify it.
I suggested w3-parse.el to Uwe. There may also be some support in
psgml, and nxml-mode (which is not yet provided by XEmacs) as well,
but they're more likely to have functions to *produce* entities than
to parse them.
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
7:39 p.m.
Uwe Brauer writes:
In turns out that Xemacs (21.4.21. Or 21.5.18) cannot deal
correctly
with the ' any more. (While GNU emacs can!)
Look at the following example:
Ich war heute in Paris
--->
J'étais aujourd'hui à Paris
instead of
J'étais aujourd'hui à Paris
The relevant code line is
(decode-coding-region (point-min) (point-max) 'utf-8)
That has nothing to do with it.
Those are HTML character entities,
they are composed of perfectly valid ASCII characters,
Google put them there, not XEmacs, and
the utf-8 coding system must leave them alone.
The problem is that babel.el isn't doing anything about translating
them to characters. Try adding
(require 'w3-parse)
(goto-char (point-min))
(while (progn
(skip-chars-forward "^&")
(not (eobp)))
(w3-expand-entity-at-point-maybe))
somewhere appropriate in babel.el. Right after the
decode-coding-region form might be a good guess.
You'll need the w3 package installed.
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
9:54 a.m.
>>>> "Stephen" == Stephen J Turnbull
<stephen(a)xemacs.org> writes:
Uwe Brauer writes:
> In turns out that Xemacs (21.4.21. Or 21.5.18) cannot deal correctly
> with the ' any more. (While GNU emacs can!)
>
> Look at the following example:
> Ich war heute in Paris
--->
> J'étais aujourd'hui à Paris
> instead of
> J'étais aujourd'hui à Paris
>
> The relevant code line is
> (decode-coding-region (point-min) (point-max) 'utf-8)
That has nothing to do with it.
Those are HTML character entities,
they are composed of perfectly valid ASCII characters,
Google put them there, not XEmacs, and
the utf-8 coding system must leave them alone.
The problem is that babel.el isn't doing anything about
translating
them to characters. Try adding
(require 'w3-parse)
(goto-char (point-min))
(while (progn
(skip-chars-forward "^&")
(not (eobp)))
(w3-expand-entity-at-point-maybe))
Thanks but which version of w3 are you using? I have 1.33
In mine (w3-expand-entity-at-point-maybe) is not a function
but from w3-parse
(defsubst w3-expand-entity-at-point-maybe ()
and
therefore when running your code I obtain
Debugger entered--Lisp error: (void-function w3-expand-entity-at-point-maybe)
w3-expand-entity-at-point-maybe()
babel-url-retrieve("http://translate.google.com/translate_t")
babel-google-fetch("Ich war heute in Paris " "de" "fr")
etc etc
Uwe
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
11:19 a.m.
Uwe Brauer writes:
> The problem is that babel.el isn't doing anything about
translating
> them to characters. Try adding
> (require 'w3-parse)
> (goto-char (point-min))
> (while (progn
> (skip-chars-forward "^&")
> (not (eobp)))
> (w3-expand-entity-at-point-maybe))
Thanks but which version of w3 are you using? I have 1.33
In mine (w3-expand-entity-at-point-maybe) is not a function
but from w3-parse
(defsubst w3-expand-entity-at-point-maybe ()
A defsubst is a function definition; it's just advice to the compiler
to inline it if the definition is already know at the time of
compilation. Hm...
Bill Perry is an idiot. It's wrapped in "eval-when-compile", I guess
in an attempt to save five or six bytes in w3-parse.elc. Try
compiling babel.el. If that doesn't work, unwrap the form (delete the
line "(eval-when-compile" and one parenthesis at the end.
Debugger entered--Lisp error: (void-function
w3-expand-entity-at-point-maybe)
I'm not sure what edebug is supposed to do about defsubsts. However,
unwrapping the definition should do the trick.
HTH
Steve
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
3:43 p.m.
>>>> "Stephen" == Stephen J Turnbull
<stephen(a)xemacs.org> writes:
Uwe Brauer writes:
>
> Thanks but which version of w3 are you using? I have 1.33
> In mine (w3-expand-entity-at-point-maybe) is not a function
> but from w3-parse
> (defsubst w3-expand-entity-at-point-maybe ()
A defsubst is a function definition; it's just advice to the
compiler
to inline it if the definition is already know at the time of
compilation. Hm...
Thanks for you comments.
Bill Perry is an idiot. It's wrapped in
"eval-when-compile", I guess
in an attempt to save five or six bytes in w3-parse.elc. Try
compiling babel.el. If that doesn't work, unwrap the form (delete the
line "(eval-when-compile" and one parenthesis at the end.
Compiling babel
does not help.
You meant delete that line in w3-parse.el not in babel? Anyway deleting
that line in w3-parse help somehow but now another error occurred.
Debugger entered--Lisp error: (wrong-type-argument char-or-string-p
8212)
The full trace is below:
That looks like a coding problem to me now
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
3:43 p.m.
>>>> "Stephen" == Stephen J Turnbull
<stephen(a)xemacs.org> writes:
Uwe Brauer writes:
>
> Thanks but which version of w3 are you using? I have 1.33
> In mine (w3-expand-entity-at-point-maybe) is not a function
> but from w3-parse
> (defsubst w3-expand-entity-at-point-maybe ()
A defsubst is a function definition; it's just advice to the
compiler
to inline it if the definition is already know at the time of
compilation. Hm...
Thanks for you comments.
Bill Perry is an idiot. It's wrapped in
"eval-when-compile", I guess
in an attempt to save five or six bytes in w3-parse.elc. Try
compiling babel.el. If that doesn't work, unwrap the form (delete the
line "(eval-when-compile" and one parenthesis at the end.
Compiling babel
does not help.
You meant delete that line in w3-parse.el not in babel? Anyway deleting
that line in w3-parse help somehow but now another error occurred.
Debugger entered--Lisp error: (wrong-type-argument char-or-string-p
8212)
The full trace is below:
That looks like a coding problem to me now
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
2:39 p.m.
New subject: primitive hack (was: problems with (decode-coding-region 'utf-8: ' gives ')
>>>> "Stephen" == Stephen J Turnbull
<stephen(a)xemacs.org> writes:
Uwe Brauer writes:
> > The problem is that babel.el isn't doing anything about translating
> > them to characters. Try adding
> Debugger entered--Lisp error: (void-function
w3-expand-entity-at-point-maybe)
I'm not sure what edebug is supposed to do about defsubsts.
However,
unwrapping the definition should do the trick.
So the only working solution I found is:
(decode-coding-region (point-min) (point-max) 'utf-8)
(goto-char (point-min))
(replace-string "'" "'" nil)
(copy-to-buffer current (point-min) (point-max))
A shame but the w3-expand-entity-at-point-maybe seems not to work.
Uwe
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
5675
days inactive
5679
days old
8 comments
3 participants
participants (3)
-
Rodney Sparapani -
Stephen J. Turnbull -
Uwe Brauer