mirrorbird(01:00:27)
as i said: if i knew the string was always gonna be short (like <1 KB) i would just use split. it's 2025, memory is cheap.
mirrorbird(01:00:00)
if you can just read your string left-to-right and set/unset a boolean flag when you see something that isn't A-Z, it may be cheaper and faster.
lenny(00:59:30)
since "regular" search won't be useful to match exactly those two points only
mirrorbird(00:59:26)
lenny: i would consider carefully the need for regex. if you are dealing with 'human book text' (where a word may end on a comma, period, etc.) then maybe best. if it's just a load of words with spaces, god no.
lenny(00:58:37)
it is a really trivial task when you think of it, but not when it comes to use regex to solve it