Changed behavior for EXTRACTDATES() - doesn't extract YYYY-MM anymore.

Today, I realised, that EXTRACTDATES() has changed its behavior in the last few days:

It cannot extract the Date from a simple text string with the Date in the format YYYY-MM.

Example: EXTRACTDATES("Project 🌳 - 2022-12 (Deadline)") returns a blank list as of today whereas it used to return the 1st December 2022 as a single list element with base type Date

Anybody experiencing the same?
Is this now the new, behavior to be expected or can we consider this as an unexpected behavior that will be fixed soon?

Any thoughts are appreciated.

 

Solved Solved
0 4 130
1 ACCEPTED SOLUTION

Yes, there seems to be some issue. If you eliminate certain special characters, it is extracting dates but it is extracting a list with two dates now.

EXTRACTDATES(SUBSTITUTE("Project :deciduous_tree: - 2022-12 (Deadline)"," - "," "))

So you may further need

INDEX(EXTRACTDATES(SUBSTITUTE("Project :deciduous_tree: - 2022-12 (Deadline)"," - "," ")),1)

View solution in original post

4 REPLIES 4

Yes, there seems to be some issue. If you eliminate certain special characters, it is extracting dates but it is extracting a list with two dates now.

EXTRACTDATES(SUBSTITUTE("Project :deciduous_tree: - 2022-12 (Deadline)"," - "," "))

So you may further need

INDEX(EXTRACTDATES(SUBSTITUTE("Project :deciduous_tree: - 2022-12 (Deadline)"," - "," ")),1)

Thanks for your suggestion, Suvrutt.

You are perfectly right! Having removed the dash "-" characters from the string, the date is successfully extracted again, actually solving the issue in my very use case 😀👍

Final formula to extract the date from the calendar event's title:

INDEX( EXTRACTDATES( SUBSTITUTE( [Title], "-", " ") ), 1)

Are there other characters I should pay attention to?

I can confirm that the :deciduous_tree: tree emoji 🌳 doesn't have an impact...

Hi @pawa ,

Nice to know it works the way you want. Regarding other characters, it is difficult to mention right away. I think typically for such functions, we will need to know the expected pattern of text strings with all possible edge cases and come up with an expression  that extracts for all possible data. 

In summary, such expressions will be data pattern specific.

Steve
Platinum 4
Platinum 4

Escalated.

Top Labels in this Space