Join

[meitata]

I tried to joining my file with 166.723 row

but the my output (Left union Join) have 166.761 (Left 142.001 and Join 24.760). How to fix it and get 166.723 row as output?

[DataNath · Accepted solution]

Are you able to expand? What are you joining on? Are they exactly the same?

All we can tell currently is that whatever you're joining on, there are 142,001 records where the field you have chosen can't find a match across the 2 inputs, and so you'll need to fix this.

[MichaelThomsen94 · Accepted solution]

Hey Meitata.
To get all rows as of both files as an output, it is necessary that the specific fields you are joining on are represented in all of the rows. So make sure to that you join on the correct field. Otherwise you can join by the Record position

[messi007 · Accepted solution]

@meitata,

Please check the key used on the Join.

Make sure your keys are unique in both side left and right.

Hope this helps!

Regards,

[meitata]

Both of them hv 1 column as config column, and I know that thing should not happen but that's happened and I kinda confused with that....

[IraWatt]

The reason you get more records from the (Left union Join) is because your join condition is not one to one. Your join must be many to one or many to many meaning the values your joining on are not unique.

[MichaelThomsen94]

Can you attach your file/Workflow?

[DataNath]

@meitata are you able to share extra screenshots that show the fields etc? Or perhaps the workflow? Otherwise it's kind of impossible to diagnose the issue. The only way this will be happening is if there aren't matching values in the two fields you've selected as the join condition. 

[Ladarthure]

@meitata if you want to check that the field you use in your join is unique, you can do this with a summarize tool before your join on each files by doing a count and a count distinct on your join field, if you have more values in the count than in the count distinct, it means you have duplicate rows on this key. if it is the same, it is fine.

You could also do this after your join in the J output but first you can group by your join key and you will see the rows that have more than one occurence, by filtering the rows where the count is different of 1.