The question was: Is there an easy way to do a delimited search and replace?

In other words, one would like to replace the prefix and suffix of a given pattern while leaving the middle bits alone.

For example, suppose one wanted to replace occurrences of




or vice versa?

Naively one would issue

  M-% filbert[ RET george( RET

and then go back and change the appropriate ]'s to )'s by hand – very error prone and time consuming.

The answer is to use a RegexpGroup? to save the middle bits, viz,

 C-M-% filbert\[\([^]]*\)\] RET george(\1) RET

CategoryRegexp CategorySearchAndReplace