> 2025年6月20日 22:32，Steffen Nurpmeso <[email protected]> 写道： > > > I myself had problems to parse his problem per se.
You mean "[t]his problem"? > > If you, Niu Danny, create your own regex engine, for your very own > work tasks, you can do whatever you want? You mean to use a period or exclamation mark instead of a question mark? > Why do you care whether this is compatible to POSIX? Since we're talking on this mailing list, we're doing standard development activities, which IMHO to some extent, include implementation prototyping and testing activities. As to "why" - because I want to be able to run existing applications, whether scripts or compiled programs. > I think the bytes 0x80-0xFF ... wasn't there an issue that takes > about the (effectively) high bit in the POSIX locale? I don't follow, can you explain? > > Other than that, if you want to, instead, create a regular > expression engine that *really* supports native languages of the > world, you will not be able to do this with POSIX, and also not > ISO C 23 (as far as i have looked). > You will need the ICU library or, you say it, perl is known to > have deeply penetrated that topic. You need to inspect entire > strings, not individual wchar_t's, you need to look around for > context in certain languages and for certain operations. I begin to realize this, sort of. > The latter is plain. I'm not a native English speaker, I don't understand this rhetoric/metaphor.
