Better Parsing of Numbers from Text Files.

User Story:

As a data analyst that gets data from all sorts of places, I'd like Dataiku DSS to do more of the heavy lifting when it comes to parsing number columns so that I don't have to take a lot of time to figure out how to do the parsing on my own.

Example:

  • One "accounting oriented data set I was looking at from a CSV file was using the accounting style for negative numbers with the ( ) before and after the number.  See here for a more complete discussion.
  • I suspect that others have other examples.  Please jump in with more examples.

Possible solutions:

  • It would be nice if the import parser dealt with these common-ish cases.
  • Or It would be nice if the Extract Number Visual Recipe Step could handle more cases
--Tom
1 Comment

Thanks to @Ignacio_Toledo there is a viable workaround for the accounting habit of using (100) to mean -100.  See here for details.

That said better parsing would always be welcomed.

--Tom

Thanks to @Ignacio_Toledo there is a viable workaround for the accounting habit of using (100) to mean -100.  See here for details.

That said better parsing would always be welcomed.