OK. The separation chaacter string used in the code to build the arrays of matching text is the '^^^' on the assumption that this is unlikely to be found within the content itself.
The pipe character '|' is still used to specify replacement pairs as this is independent of the source text extraction and used only to define find and replace patterns in order to clean the output as necessary.
I hope that this clarifies and if you have any ideas for improvement then we can look to implement.
