On Saturday, 28 December 2019 18:25:09 -03 Tom Honermann wrote:
> That would be a significant break to existing code, particularly for
> non-ASCII based implementations.

I understand the boat has sailed. Breaking compatibility with existing code is
a no-go.

My point was that supporting other encodings besides UTF-8, UTF-16 and UTF-32
should never have been supported. No code could be using a class that didn't
yet exist. Existing codebases could have convert text to UTF-8 in order to use
the new classes.

An implementation could provide further encodings as an implementation
extension, so long as they didn't break compatibility with standards-
conforming code.

Oh well, the boat has sailed.

Anyway, as far as I understand from Ville, std::regex has serious flaws. Qt 6
briefly considered using it but decided to stay with PCRE2.

-- 
Thiago Macieira - thiago (AT) macieira.info - thiago (AT) kde.org
   Software Architect - Intel System Software Products