nnrss strips Unicode

nnrss strips Unicode

[jidanni]

In http://pinyin.info/news/2008/critique-of-proposed-guidelines-for-writing-taiwan-place-names/#comment-378523
I complain that Unicode #8217 got stripped.
And what stripped it was nnrss.el, as you can see if you read the
article via RSS via http://pinyin.info/news/feed/ .

Re: nnrss strips Unicode

[Katsumi Yamaoka]

>>>>> jidanni@jidanni.org wrote:

> In http://pinyin.info/news/2008/critique-of-proposed-guidelines-for-writing-taiwan-place-names/#comment-378523
> I complain that Unicode #8217 got stripped.
> And what stripped it was nnrss.el, as you can see if you read the
> article via RSS via http://pinyin.info/news/feed/ .

You seem to use Emacs 22.x, or the old Emacs CVS that has not
merged the Unicode-2 branch yet.  If so, the problem is due to
the function `mm-char-or-char-int-p' that the function
`mm-url-decode-entities' uses.  This form returns SPC in your
system, doesn't it?

(with-temp-buffer
  (insert "’")
  (mm-url-decode-entities)
  (buffer-string))

There seems to be no easy solution for Emacs 22 but I can offer
a workaround:

(eval-after-load "mm-url"
  '(progn
     (require 'w3m)
     (defalias 'mm-url-decode-entities 'w3m-decode-entities)))

Regards,

Re: nnrss strips Unicode

[jidanni]

KY> You seem to use Emacs 22
Debian sid 22.1+1-2.3
KY> This form returns SPC in your system, doesn't it?
Yes.
KY> There seems to be no easy solution for Emacs 22 but I can offer
KY> a workaround:
That didn't fix it. The problem is here, not in any .html file:
$ GET http://pinyin.info/news/feed/|grep Da\&
<td>Da&#8217;an</td>
I hope you can see it in nnrss in your current emacs. OK, never mind
for my old emacs. I will just live with it. Thanks.