bytes_regex() only allows regexs that match valid UTF-8

[hkratz]

Proptests trying to generate invalid UTF-8 sequences with bytes_regex() result in an error.

Source:

proptest! {
    #[test]
    fn simple_invalid_prop(ref input in bytes_regex(r"(?-u)\xFF").unwrap()) {
    }
}

Error:

---- tests::simple_invalid_prop stdout ----
thread 'tests::simple_invalid_prop' panicked at 'called `Result::unwrap()` on an `Err` value: RegexSyntax(Translate(Error { kind: InvalidUtf8, pattern: "(?-u)\\xFF", span: Span(Position(o: 5, l: 1, c: 6), Position(o: 9, l: 1, c: 10)) }))', src\tests.rs:239:67
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace

This makes them much less useful in practice. Maybe creating a parser with regex_syntax::ParserBuilder::allow_invalid_utf8() set to true is sufficient to fix this.

[zackw]

Whoops, this is exactly the same as #336. I should have looked more carefully at the issue list before filing that one.

[matthew-russo]

closing this as duplicate. #336 will be used for tracking