Discover this year's submissions to the Dataiku Frontrunner Awards and give kudos to your favorite use cases and success stories!READ MORE

`Fill empty cells` step in Preparation generates bad BigQuery SQL

UgoD
Level 2
`Fill empty cells` step in Preparation generates bad BigQuery SQL

Hi,

I'm using Dataiku 10.0.5, Python 3.6 and BigQuery.

 

I'm trying to fill empty cells of all my columns with 0, but it's not working correctly for some columns.

The first column it's not working with is a boolean column and the others are int (for both database and dataiku types)

When I look the SQL generated, it look like that:

 

CASE 
  WHEN (`my_column` IS NULL OR (`my_column` = '')) THEN CAST('0' AS STRING)
  ELSE `my_column`

 

 

and the error I get is

  com.dataiku.dss.shadelib.com.google.cloud.bigquery.BigQueryException: No matching signature for operator = for argument types: INT64, STRING. Supported signature: ANY = ANY 

for the INT column (replace INT64 by BOOLEAN for the other error)

 

Seems to be normal, but I don't understand why all my INT64 typed columns don't get this error.

 

Do you have an way to avoid this error ? Am I doing smtg wrong ?

Best regards

Ugo


Operating system used: Windows 10


Operating system used: Windows 10

0 Kudos
0 Replies