https://bugs.exim.org/show_bug.cgi?id=2305
--- Comment #4 from Philip Hazel <[email protected]> --- I see. Experiments with Perl indicate that the mere presence of \N{U+xxx} turns on Unicode interpretation for the whole pattern. So changing \x{41} into \N{U+41} for example is more than just a change of notation. PCRE2 cannot do this. It operates in one of two modes: EITHER input is UTF and case handling is per Unicode rules, OR input is not UTF and case handling is locale dependent. Whichever is controlled by the PCRE2_UTF flag, set at the start of processing. So, the only thing I can do, and will do, is to make \N{U+xxx} give an error if encountered in non-Unicode (non-UTF) mode. -- You are receiving this mail because: You are on the CC list for the bug. -- ## List details at https://lists.exim.org/mailman/listinfo/pcre-dev
