Description:
Since regexp are already supported well enough to use in functions like COUNTIFS, why plain regexp match couldn't be used as a function usable in a formula?
This would make extraction of a desirable substring simple in most situations (even with positive/negative lookahead/lookbehind regexps), without either requiring more custom functions or unnecessarily complication.

Steps to Reproduce:
I want to parse a value out of text — e.g. the text in column $I may contain "SomeTrait[123]" and if so, my goal is to extract value 123. I need a formula to  fill a long column.

Actual Results:
I'll end up with something like
=IFERROR( FIND("SomeTrait[",$I2),0)
in one column ($Z), and then
=IF($Z2>0,VALUE( MID($I2,$Z2+LEN("SomeTrait["),FIND("]",$I2, $Z2)-LEN("SomeTrait[")-$Z2) ),0)
in another, and still have to pre- sanitize text so it won't fail due to containing "Trait [5]" instead, or make it even more complicated.
Every time a string needs parsing, this involves unnecessarily complex constructs made of LEFT()/MID()/RIGHT(), FIND/SEARCH() and LEN() that have to either be repeated with the same arguments or consume half a dozen columns as temporary variables, just to parse this one value (and there are more of them).

Expected Results:
=VALUE(REGEXPMATCH($I2, "(?<=SomeTrait\[).*(?=\])"))
matches "123". Just find the right cell, extract desired text from it, process it. Two functions, fairly transparent… regexps may be not always quite obvious, but at least short enough to analyze as a whole, and often easy to manually test (if something works in search, it will work in a formula too).


Reproducible: Always


User Profile Reset: No



Additional Info: