On Thu, Apr 16, 2009 at 10:54:16AM -0400, Tom Lane wrote:
> Sam Mason <sam@samason.me.uk> writes:
> > I'd never heard of UTF-16 surrogate pairs before this discussion and
> > hence didn't realise that it's valid to have a surrogate pair in place
> > of a single code point. The docs say that <D800 DF02> corresponds to
> > U+10302, Python would appear to follow my intuitions in that:
>
> > ord(u'\uD800\uDF02')
>
> > results in an error instead of giving back 66306, as I'd expect. Is
> > this a bug in Python, my understanding, or something else?
>
> I might be wrong, but I think surrogate pairs are expressly forbidden in
> all representations other than UTF16/UCS2. We definitely forbid them
> when validating UTF-8 strings --- that's per an RFC recommendation.
> It sounds like Python is doing the same.
OK, that's good. I thought I was missing something. A minor point is
that in UCS2 each 16bit value is exactly one character and characters
outside the BMP aren't supported, hence the need for UTF-16.
I've failed to keep up with the discussion so I'm not sure where this
conversation has got to! Is the consensus for 8.4 to enable SQL2003
style U&lit escaped literals if and only if standard_conforming_strings
is set? This seems easiest for client code as it can use this
exclusively for knowing what to do with backslashes.
-- Sam http://samason.me.uk/
