Data Cleanse Pro vs Data Cleansing (a bug?)

I found strange behaviour of the Data Cleanse Pro tool, which I don't understand and it looks to me as a bug.
I have a large set of data with poorly written names (like: NameSurname, Name Surname, Name Surname, Name Surname, ....)
I wanted to clean that so I used Data Cleansing Pro and set to remove Leading and Trailing Whitespaces and Tabs, Line Breaks, and Duplicate Whitespaces.

To my surprise, after such cleaning I ended up with more name variations than before the cleaning. But when I used the older Data Cleansing tool (with the same settings) the result was as expected - after the cleaning, the number of variations for the same names significantly decreased. When I compared the output of Data Cleanse Pro and Data Cleansing tools I found that the Data Cleanse Pro is not removing all whitespaces as expected, and in certain cases it is even adding (!) spaces to the values.
It seems that it happens when we have more than two spaces in a row and for the cases when we have more than one place where duplication happen

(i.e.: First Name Second Name, Surname - in such case in one place additional space may be removed, in other it can be added).

Did anyone experience such behaviour? Is it a bug or am I missing something?

EDIT: as I just checked, using Data Cleanse Pro with "Tabs, Line Breaks, and Duplicate whitespace" on the value which have 7 words (only one space between each word) crated over 2500 versions with various number of spaces out of only one 😱

By the way, the "remove all Whitespace" option works just fine.

Bug

Accepted answers

NeoInfiniTech

Hi @kwieto,

This is a known defect, and it has been fixed in the latest patch of 2025.1 (Patch 5, 2025.1.2.5.239) and 2025.2 initial release also contains the fix for this bug. I can confirm upon testing that this behavior exists in 2025.1.2.142, which is a version that doesn't contain the fix, whereas running the workflow in version 2025.2.1.117 produces only one row without any whitespace variations. You can either upgrade your Alteryx Designer 2025.1 version by installing the latest patch or install Alteryx Designer 2025.2 to fix the issue.

All comments

alexnajm

Can you show us with some sample data and a workflow? Curious to see more

kwieto

Sure.

The bug (now I am sure it is a bug) can be recreated on large sets of data - in my case it was over 5 millions, but as I played with the tool it 200k records is enough to expect strange behaviour.

I'm attaching a sample workflow and a screenshot - in the workflow, there is only one record with a value "This is a test". It is then multiplied to 1 000 000 by the generate rows tool, and then aggregated in three ways - without any intervention (correct), sending through the old Data Cleansing tool set to remove leading/trailing space and all duplicated whitespaces (also correct) and through the new Data Cleanse Tool (32 different variations generated).
Also, each time you may expect different number of values created, I run the workflow again and got 42 random values instead of 32. Next run and I got 27

Should I report it somehow?

Data cleanse pro bug.jpg

Data Cleanse Pro.yxmd

NeoInfiniTech

Hi @kwieto,

jrlindem

Agree with @NeoInfiniTech . The Patch file for 2025.1 should solve this for you. It will appear as it's own download file in your license portal. If you are installing 2025.1 fresh, the main designer file already has the patch applied so you won't need to do both. The patch is only for existing installs. Cheers, -Jay

kwieto

Thanks for the reply. I use 2025.1.2.120 Patch: 1 version and can't upgrade due to the company policies, so for the time being I'll need to stick to the older Data Cleansing tool.

alexnajm

Thanks everyone for the information - good to know!

Quick Links

This months top contributors

atcodedog05 19458

Qiu 15866

binu_acs 15708

MarqueeCrew 13708

apathetichell 13703