Re^5: Extract sequence of UC words?

[monarch]

The issue I have with your examples, BrowserUk, is that you are mandating at least 2 upper case letters. My regexp permits a single capital letter.

I think it is important to have the optional section, because the desired expression is "one or more upper case letters" optionally followed by any number of "spaces followed by upper case letters".

[BrowserUk]

I upvoted your post above, but still your regex m/(\b(?:[A-Z]+(?:\s+[A-Z]+)*)+\b)/g made me squirm. Whenever I see sequences of nested quantifiers like that:+)*)+ I get uncomfortable, remembering various pathelogical cases I've constructed in the past.

To that end, I thunk again, and came up with this which I believe meets the 'spec', whilst avoiding the nested quantifiers;

m[
 (
    \b
    [A-Z]
    (?: [A-Z\s]* [A-Z] )? 
    \b
 )
]gx