GREP: Apply to first instance only

Question

Hi,
Another GREP question for you if you can help.
Situation is trying to apply a character style before a specific word, i.e. Applied Character Style "four".......
There are 4 words (lets call them: one, two, three, four) which are a common occurence and will always appear after the text that needs to have the applied character style, but my issue crops up when one of the four words appears twice or more times in one line and the GREP applies it to the 2nd or more instance of this word, i.e. Applied Character Style "one" random text "one"
I am using these greps in the same paragraph style:
^.*(?= one)
^.*(?= two)
^.*(?= three)
^.*(?= four)
I only want to have the GREP to apply to the first instance only - hope this stuff above makes sense?
Thanks

Answer 1

It's because GREP is greedy, by default.
The expression
^.*(something more)
will always grab as much characters as possible, before processing "something more": effectively, it will match anything up to the very last occurrence of that text.
You can instruct GREP to not match greedy by adding a '?' after the repeat command code:
^.*?(?=\b(one|two|three|four)\b)
-- the question mark can be added to each of the "repeat-this" codes: +, *, and ? ("!??" would mean: "match '!' zero or once, but uses the shortest match possible).
I condensed your list of four commands into one, because it seems to me that all of them would get matched in turn, leading to problems when you have both "one" and "two" in the same sentence. With your GREPs, first all up to "one" would get marked, then all up to "two". Mine uses 'whatever comes first'.
(I also added 'word boundaries' \b before and after, because you wouldn't want to match the 'one' in "Mr. Jones".)