Evaluating extended ASCII character with RegEX

[PhotoGraphex]

Hello, 

As I am using extensively the function provided by Mark in the thread: 

Regular Expression evaluation in SuperCard

I am regularly getting issues when the variable inText contains an extended ASCII characters. More specifically with word boundary, for instance with the expression [[:>:]].

For instance, the evaluation of a word like Déclaration would be split in two an see the letter D an a word ending. 

I haven't been able to find a workaround, otherwise when using an online RegEX simulator to flag the inExpr for a full match of unicode with /u.

Would it be possible with GNU bash, version 3.2.57? 

Many thanks