[Date Prev][Date Next][Thread Prev][][Date Index][Thread Index]

Re: windows-1252



>>>>> In [emacs-w3m : No.08007] David Hansen wrote:

>> However I'm not sure whether it is a right solution (I don't
>> feel like recognizing those extra data as iso-8859-1, though I'm
>> not familiar with the actual situation in the Latin world).  In
>> addition, it is not effective in the Gnus article buffer.

> What's the problem in article buffers?

Because w3m-current-content-charset is nil, w3m-fix-fscked-latin-1
does nothing (I checked it only with sb-zeit-de).  We need a way
to know the contents use Latin-1 and the extra data there.

> Actually i think it is no problem to treat all latin-1 pages as
> windows-1252.  Seems that's the way how Mozilla and related browsers
> handle this problem (at least the pages discussed here all display
> "fine").

> Any sane latin-1 page will not have any control characters within
> the text so i don't see a problem with interpreting them as
> windows-1252.

I think ones who extend the Latin-1 specification is not us even
if Mozilla temporizes to the standard which someone has invented
arbitrarily.  We cannot help to do that if it is the only way,
though.

>> On the other hand, sb-fau.el provides shimbun-fau-coding-system,
>> which will be windows-1252 if it is available.

> Do you know which emacs versions support windows-1252?

Yes.  But there's a way to support windows-1252 even in Emacs 21.

>>>>> In [emacs-w3m : No.08008] ARISAWA Akihiro wrote:

> I created codepage-ex library last year, which provides cp1252
> coding system.  Is this no use?
> http://www.nijino.com/ari/emacs/codepage-ex.html

To use this is different than enhancing the Latin-1
specification arbitrarily.  Even if XEmacs and old Emacsen don't
work with it, it cannot be helped.  It is similar that there's
no way to display images in Emacs 20.

> BTW at least the middle and east European windows charset is not
> "compatible" with any latin-n encoding, so it's likely that this
> problem exists only in the latin-1 world.

It can become a reason why we enhance the Latin-1 specification
in emacs-w3m.  However, I'm not yet convinced which is better,
enhancing Latin-1 or providing windows-1252.