How to remove all characters except alpha numeric in text column

[rajeshchawla]

The column is populated with data that contains quite a few Unicode characters. The rows of interest to me are the ones where the characters are only in the range of a-z (upper or lower case) or 0-9. To me, the replace functionality was not enough as there are too many characters to replace.

I did find a potential solution using regular expressions & R in the post: http://radacad.com/power-bi-and-regular-expressions, however, I was hoping for an alternative that did not use R.

Pointers / Suggestions appreciated.

Regards,

Rajesh

[OwenAuger · Accepted solution]

@rajeshchawla

Here is one way of doing it. The below code filters rows to only those where Text is equal to just the alphanumeric characters from Text.

This assumes the last step of the query is PreviousStep and the column of interest is named Text.

= Table.SelectRows( PreviousStep, each [Text] = Text.Select([Text],{"a".."z","A".."Z","0".."9"} ) )

Regards,

Owen

[rajeshchawla]

Thank you!

[GarrettU]

Posted to unsolved new post

https://community.powerbi.com/t5/Desktop/Changes-to-duplicated-columns-also-affect-the-original-colu...

[Anonymous]

Could you provide more information regarding your requirement? Do you want identify if the column contains any invalid character, such as !@#?

[rajeshchawla]

Added more detail to original question.