Great news on the wrap/no wrap!
I suppose I could make a REALLY LONG saved group for the roman numerals....
I was inserting id's to match the heading text in a condensed format. "Ch_1" for "Chapter 1", "Pt_2" for "Part 2", etc. I thought that would provide a more intuitive meaning than a generic sequence number when I review links - especially in other files.
example:
original headings
<h2>Part 1</h2>
<h3>1</h3>
<h3>2</h3>
<h3>3</h3>
<h2>Part 2</h2>
<h3>1</h3>
<h3>2</h3>
<h3>3</h3>
yadda, yadda
Replace Part headers: Find <h2>Part (.*?)</h2> Replace <h2 id="Pt_\1">Part \1</h2>
Replace Chapter headers: Find <h3>(.*?)</h3> Replace <h3 id="Ch_\1">\1</h3>
gives <h2 id="Pt_1">Part 1</h2> <h3 id="Ch_3">Chapter 3</h3>
This results in duplicate Chapter id's so I go to the beginning of each part and replace to the end (without wrap):
Find id="Ch_(.*?)"> Replace "Ch_1-\1">
/* which changes ALL the chapter headings but still have duplicates.
Then go to part 2
Find id="Ch_1-(.*?)"> Replace "Ch_2-\1">
/* which changes ALL the chapter remaining headings EXCEPT part 1
Then go to part 3
Find id="Ch_2-(.*?)"> Replace "Ch_3-\1">
/* which changes ALL the chapter remaining headings EXCEPT part 1 and part 2
etc, until the end of the file