Re: Re: Gobo Regexp and Unicode support.

Colin Paul Adams wrote:
>>>>>> "Ted" == Ted  <kkkkg999-/[email protected]> writes:
> 
>     Ted> The result is "False" with patched Gobo Regexp.  But using
>     Ted> UC_STRING, with original Gobo Regexp, the result is "False"
>     Ted> too.
> 
> Then it looks like your patch has something to be said for it.
> 
> But not entirely. I took a look at the code, and there is a hard-code
> 256-bit pair of arrays for case translation, so a case-insensitive
> match will definitely crash on Unicode (in any case, that method of
> doing case-insensitive matches is wrong. For instance, the three strings ESSEN,
> essen and e&#x00DF;en  should all compare equal for case insensitive
> matches.)

Would it be acceptable to use Ted's patch (with some protection when
accessing the 256-bit pair of arrays) and make it clear in  the doc
of the regexp library that for case-insensitivity, \w, etc., only
ASCII characters are taken into account? Then those for whom this
limitation is a problem would have to wait and use your fully
unicode aware regexp library.

-- 
Eric Bezault
mailto:ericb-D6Qt/9opevxWk0Htik3J/[email protected]
http://www.gobosoft.com