From: Asmus Freytag (email@example.com)
Date: Fri Apr 17 2009 - 13:14:33 CDT
On 4/17/2009 9:14 AM, John W Kennedy wrote:
> On Apr 17, 2009, at 7:32 AM, Sam Mason wrote:
>> On Thu, Apr 16, 2009 at 01:04:30PM -0700, Asmus Freytag wrote:
>>> What should definitely result in an error is to write '\U0000D800'
>>> because the 8-byte form is to be understood as UTF-32, and in that
>>> context there would be an issue.
>> That strikes me at too pedantic; if we did that should we also reject
>> the number one when spelled as '00000000001'?
> Quite a few programming languages will reject '00000000008'.
Note that my example (D800) was an unpaired surrogate.
These are not legal UTF-32, hence "there would be an issue".
I should be more precise in one aspect: This usage should
result in an error, if the string into which it is inserted is
supposed to be valid when mapped to UTF-32.
If not, then the only issue is whether redundant leading 0
are a problem. They are not, as long as the U and not the
u prefix are used.
This archive was generated by hypermail 2.1.5 : Fri Apr 17 2009 - 13:17:01 CDT