Re: Help: using chinese-gbk

[Katsumi Yamaoka <yamaoka@xxxxxxx>]

Hi,

>>>>> In [emacs-w3m : No.09362] Jielei Fan wrote:

> I surf internet using the emacs-w3m written by you. But I meet a
> problem, I have already installed
>  mule-gbk package on my emacs 22(I use it on windows xp system), and
> it works well, however it
> does not work in w3m-mode, in which iso-8859-1-dos, gb2312-dos or
> other code system will invoked
> automatically. But if I write these codes

> ;; (setq w3m-bookmark-file-coding-system 'chinese-gbk)
> ;; (setq w3m-coding-system 'chinese-gbk)
> ;; (setq w3m-default-coding-system 'chinese-gbk)
> ;; (setq w3m-file-coding-system 'chinese-gbk)
> ;; (setq w3m-file-name-coding-system 'chinese-gbk)
> ;; (setq w3m-terminal-coding-system 'chinese-gbk)
> ;; (setq w3m-input-coding-system 'chinese-gbk)
> ;; (setq w3m-output-coding-system 'chinese-gbk)

> in my .emacs,

> the website will be emerged mess code.

First of all, you should never have need to modify at least
`w3m-input-coding-system' and `w3m-output-coding-system'.  The
values for those variables should be supported by the external
w3m command, and `utf-8' is a good choice.  If I understand
correctly, GBK is a superset of GB2312 and all characters can be
expressed with Unicode.

Emacs-w3m fetches an html page as binary data, decode it
according to the charset that the page specifies[1], encode it
with a certain coding system[2], and passes it to the external
w3m command.  And then the external w3m processes it, encodes it
with a certain coding system[3], returns it to emacs-w3m, and
finally emacs-w3m decodes it with a certain coding system[3].

[1] The charset is specified in the page header or in the meta
tag which looks like:

<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">

The `=' command shows the page header, and the `\' command shows
raw (but charset-decoded) html contents.

[2] The value of `w3m-input-coding-system'.

[3] The value of `w3m-output-coding-system'.

I think the cause of your problem is that emacs-w3m doesn't know
how to find a suitable coding system for the GBK charset, and it
might be solved by adding a proper rule to the
`w3m-charset-coding-system-alist'.  Could you let me know a
typical web page that uses the GBK charset?

> Another problem is that, when I use command w3m-search(google engine),
> after I input chinese character,
> the content will be taken as ??????? in google website. Could you
> please tell me how to solve it?

Well, it will probably be solved if all GBK pages are displayed
correctly.

Regards,