Re: windows-XXXX and cpXXXX

[Kenichi Handa <handa@m17n.org>]

In article <ulkxkse69.fsf@gnu.org>, Eli Zaretskii <eliz@gnu.org> writes:

> Why would non-DOS users need cp1252? why cannot they use windows-1252,
> which is the official IANA name of that encoding?

> The fact that people get confused by what the Emacs manual says about
> this (which I believe is what led us to this discussion) can be taken
> care of by fixing the manual.  There's no need to modify code for that.

> In short, cp1252 is something we invented in codepage.el.  So we might
> as well retire those cpXXX symbols for good, on all platforms (except
> DOS, which must use codepage.el).  That is my suggestion: let's remove
> cpXXX and cpXXXX aliases from code-pages.el, and let's start educating
> users to use windows-XXXX instead.

At least cpXXX can't be replaced by windows-XXX.  Do you
suggest to use the official name ibmXXX instead?

But, I've just checked the name cp1252 on google and found this:

iconv supports cpXXXX from fairly long ago.  CPXXXX are used
also in this page:
  <http://www.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS/>
It seems that Java (JVM) also support the name cp1252.

So, I think the name cpXXXXs have already acquired
citizenship, and thus it is a lack of sophistication not to
support the name cpXXXX.

Of course, I agree that it is the right thing to somehow
merge codepage.el and code-pages.el, and provide the same
definition of a coding system to both DOS and non-DOS users.
But we don't have a man power to work on it now.

---
Kenichi Handa
handa@m17n.org