On 10/19/21 11:36, Tomasz Dziendzielski wrote:
I found that for optimization purposes common cases in re_string_peek_byte are missed if pstr->mbs_allocated is true, thus skipping them also for case insensitive.
Unfortunately I've lost context here. Is this patch being proposed to improve performance, or to fix a bug?
I don't see how the patch preserves correctness: if pstr->icase is true, surely in general we need to look at the translated string, not the original.
