Regex and UniStr

[bkamins]

Working with regexes on UniStr will be slow as currently all has to be converted to String to work (regex as well as the string in which we look for it).

[ScottPJones]

Yes, that's only for the moment. For ASCIIStr and LatinStr, the same library can be used, and will be faster than using UTF-8, it's a matter of a flag passed to the library. For regex on UCS2Str (and _UCS2Str), and UTF16Str, those will need version of the PCRE library compiled for 2 byte, and UTF32Str will need it compiled for 4 byte.
For the moment, as long as it is working, I'm happy, I'll deal with the performance issues of regexes later, after everything else is working correctly.

[bkamins]

Agreed. I just wanted to leave it open as it is an end-user critical part that will require changes to both: how "Regex" object is constructed and how it is called.

[ScottPJones]

BTW, I really appreciate your thoughtful reviews and ideas!

[ScottPJones]

If you have a chance, please take a look at the Regex support that I just added.
For now, it's just for Str types where the codeunit is UInt8.