Ar an cúigiú lá de mí Lúnasa, scríobh Stephen J. Turnbull:
Aidan Kehoe writes:
> > > * mule/mule-cmds.el (finish-set-language-environment):
> > > Implement error-sequence-coding-system.
> >
> > Please call this "invalid-character-coding-system" or something
like
> > that.
>
> Why do you find that clearer?
Several reasons. "Error" is a signal in Lisp, but no error will be
signaled here. "Error sequence" could be almost anything unexpected;
people encountering these names will not necessarily have the context
to realize that they refer to invalid encoding sequences. Even in
that context, errors will by definition not be coding systematic, so
it's sort of an oxymoron. In fact, my original thought was that
e-s-c-s was the coding system that threw the error in the first place.
OK. I prefer “invalid-sequence” over “invalid-character”, though; each
invalid UTF-8 octet becomes a single valid XEmacs character, and a
collection of invalid octets may well correspond to a single incorrectly
encoded character. E.g. the output of
(decode-coding-string "\xC0\x8a" 'utf-8)
is a string with two XEmacs characters, but the source string is an overlong
encoding of a single Unicode character.
--
¿Dónde estará ahora mi sobrino Yoghurtu Nghé, que tuvo que huir
precipitadamente de la aldea por culpa de la escasez de rinocerontes?
_______________________________________________
XEmacs-Patches mailing list
XEmacs-Patches(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-patches