[Top] [All Lists]

Re: FW: "Unicode" vs "ISO 10646"

2005-11-03 19:37:08

On Thu, 2005-11-03 at 09:26 -0800, Philip Guenther wrote:
Kjetil Torgrim Homme <kjetilho(_at_)ifi(_dot_)uio(_dot_)no> writes:
I believe it would be very useful to define a stringprep profile for
Sieve.  the profile may be defined for more general usage (comparators
too, others?), but something [SIEVE] can refer to would make things a
whole lot clearer.  I don't suggest to make it mandatory for Sieve, but
*either* you support US-ASCII only (and use the identity mapping for
other code points) *or* you support the profile.

Could you be more specific about what this profile would be used for in
[SIEVE]?  If it's 'just' a comparator, then it belongs in the collation
doc, but that already defines defines a *nameprep* based collation
(i;nameprep;v=1;uv=3.2) as well as collations using the Unicode
Collation Algorithm, so I suspect you have more in mind that just using
it as a comparator.

well, by using a nameprep profile, quite a few normal characters are
prohibited, in particular U+00A0 (&nbsp;) is way too common in e-mail to
be disallowed, IMHO.  but I'm not quite clear what prohibiting a
character really means for Sieve.

I just want something which makes sense for matching in base Sieve, but
I also want a :lower and :upper which works in "variables".  using the
nameprep profile is certainly better than locking it to US-ASCII,
Kjetil T.

<Prev in Thread] Current Thread [Next in Thread>