With "Use wildcards" turned on, enter the search term
([! ]@) ([!^13]@^13)
and the replacement term
\1^t\2
In the search term, there is a space character after the first exclamation mark, and another after the first closing parenthesis.
Explanation: The expression in the first pair of parentheses of the search term means "one or more characters that are not spaces", and that corresponds to the \1 in the replacement term.
The expression in the second pair of parentheses means "one or more characters that are not paragraph marks, followed by one paragraph mark", and that becomes the \2 in the replacement.
The space character between the two sets of parentheses is replaced by the ^t (a tab character).
That worked perfectly. I would have never figured that one out.
I ended up deciding to replace the first 2 spaces with tabs. I started
to see if I could write the expression to do it, when I realized that
I could simply run this one twice.
I'm guessing that a single expression to do that would involve
inserting one more "term" in between the ones you have now, maybe
something like this:
([! ]@) ([! ]@) ([!^13]@^13)
and the replacement term would be
\1^t\2^t\3
Is that correct?
Is there any limit to the number of terms?
Thank you very much.