Regular Expression vs Locale Aware Regular Expression

[Jean_Claude]

Can someone explain in «digestible language» what is the differences between Regular Expression vs Locale Aware Regular Expression we can use in the find/Replace with field. When one should be use and what benefits a user can have to use the other?

Thanks!

 

[garrettm30]

This is what was said in the beta where the feature was introduced:

Quote

-- 'Locale Aware Regular Expression' - Find / Replace will be performed using the locale inferred from the text being searched and locale aware collation is implied
-- Please note that a 'Locale Aware Regular Expression' has to treat each block of text with a different language setting in isolation

Source Thread

In this mode, somehow regex takes into account the language encoding that the searched text is in. I am struggling to think of any examples of when or how to use it, so I am eager to see what the others say.

[walt.farrell]

The collation sequence is certainly part of it. For example, consider the regular expression

[a-z]

and the text string

aàáâãäå

A Find using a Regular Expression in Publisher will match only the first character ("a") but using a locale-aware regular expression will match each of them (one at a time, of course).

On the other hand \w or [\w] will match each of those letters whether you use the locale-aware option or not, at least for English text.

The comment in the Help about treating each block of text in a different language in isolation is also important. Given a text string of "aaaabbbb" where "aaaa" has one language specified, and bbbb has a different language specified, the locale-aware regular expression \w+ will match "aaaa" and "bbbb" separately. The non-locale-aware version will match the entire string "aaaabbbb".

[garrettm30]

That is a real advantage indeed for foreign languages! In the past I have resorted to tricks like [a-zéèêëïîôàùç].

I am so glad to understand this distinction. Thank you.

[rcheetah]

Oh, now I understand. Very well done!