Nigel Swinson wrote:
Folks, we now have a bunch of sieve drafts on the table. Is it time to
start some of them through the standards process? If so, which ones?
I'd like to see the regex draft through to a standard. Very useful and
don't think it's changed very much recently. Ken?
http://search.ietf.org/internet-drafts/draft-murchison-sieve-regex-06.txt
Most of the changes have been editorial. I think in order for it do get
through the standards process we will probably need to do something
about internationalization/localization (unless Ned thinks otherwise):
0.3. Open Issues
The major open issue with this draft is what to do, if anything,
about localization/internationalization. Are [POSIX.2] collating
sequences and character equivalents sufficient?
POSIX regexp is awkward in that it is locale based, and we really don't want
that. OTOH, it is clearly the definition of choice for regexps. I think the way
this is handled in the TCL regexp package is the right way to go: In that case
regexp always operates in the "Unicode" locale. If we want to do fancier things
with normalization and such we should do it via stringprep and tie it in with a
comparator.
Does POSIX.2 actually define additional collating sequences or any
character equivalents? If not, I think simply stating that this has to
operate in the "Unicode" locale is the way to go.
Ned