RegEx using Unicode Blocks Inquiry

[IvanaF]

While trying to use Unicode Blocks either in RegEx tool  or in RegexCountMatch formula, I am not able to achieve correct results?

The ultimate goal would be to get characters count using the \p{IsLatin} block or variations of other blocks.

Could anybody be so kind to help/share experience in this area, please? Or confirm this syntax character classes are not supported in Alteryx? 

Main resources I have used so far:

• Unicode Blocks: http://www.regular-expressions.info/unicode.html
• Alteryx RegEx help + RegEx Perl Syntax Guide (Alteryx Knowlege Base), Boost-Supported RegEx Character Classes

Issues using RegEx tool (using only InLatin_Basic for simplicity):

1. [[:InLatin_Basic:]] -  error: RegEx (#): RegEx: An invalid character class name was specified in a [[:name:]] block at character 3
2. [:InLatin_Basic:] - no error but incorrect result - i.e. 'e' or 'm'  were not matched

As a workaround, using the unicode range seems to work ok ([\U+0000-\U+007F]) - but this is a bit cumbersome (especially when trying to work with multiple blocks).

Testing workflow is attached.

Many thanks in advance.

[jrgo]

Hi @IvanaF

Not sure if I fully understand what your end result should be, but would this expression work within the RegEx tool?

[^[:unicode:]]

[jdunkerley79 · Accepted solution]

I don't believe the Boost library (which Alteryx uses for its Regex functionality) supports Unicode blocks at present.

The recommended workaround I know is to do what you were with Unicode ranges.

[IvanaF]

Hi @jrgo, many thanks for the suggestion; it helped me to the find additional resources in Alteryx help and can definitely be used in certain use cases. Thanks again for your time on this.

[IvanaF]

Hi @jdunkerley79, many thanks too for your time and advice. Cheers!