[Bese-devel] UTF-8

[Alexey Lebedeff]

I'm using ucw with mod_lisp backend and sbcl.

To make it work with utf-8 I made few small changes to arnesi & ucw.

Changes to arnesi:

- When utf-8 is used, encoding of characters > 127 with '&xXXXX'; is not
  the right thing. So i dissallowed such encoding when sb-unicode feature
  is present. This is not quite correct - not all users of
  unicode-enabled sbcl wants this behavior(that's for sure) - but my
  superficial knowledge of arnesi doesn't allow me to make this change in
  better way.

- nunescape-as-uri is now accepting external-format argument, same as
  unescape-as-uri.

Changes to ucw:

- Allow to specify various external-formats for different purposes:
  unescaping of urls, communication with slime and http.

Accompanied by setting slime-net-coding-system on emacs side, it works
pretty well.


And also unrelated to external formats change: 
There was missing argument in call to ucw.backend.dribble. So, when I was
starting ucw, sbcl complained to me about "format" arguments mismatch,
resulting in impossibilty to start ucw without user intervention.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: arnesi.CHANGESET
Type: text/x-patch
Size: 14060 bytes
Desc: not available
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050827/48400d40/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ucw.CHANGESET
Type: text/x-patch
Size: 15907 bytes
Desc: inline
URL: <https://mailman.common-lisp.net/pipermail/bese-devel/attachments/20050827/48400d40/attachment-0001.bin>
-------------- next part --------------

-- 
[1180'99:11-6a] [BMSTU-AK:5-111]