Sorry, I didn’t see that you had replied.
It might be possible to write several that cover the possibilities, not sure. In the project I was working on, there were several different things breaking up the \wj
For a \wj marker that was broken up by verses, the regular expression I used was
(\\wj [^\\]+)(\r\n\\v \d+(-\d+)? )
Replacement string
\1\\wj*\2\\wj
In the replacement string, there is a final space after the \\wj
Basically, it searches for non marker text, followed by a verse marker, and then closes the \wj before the verse marker, and reopens after.
Also, if I recall correctly, if doing a replace all, you have to rerun this multiple times.
BEH