0

When I attempt to rerun a recipe with a different data set (CSV) it gives me a column mismatch error even though both columns are strings. I tried adding a step in the recipe to explicitly make both columns stings, but I am still getting this error.

I am not able to delete and remake the BQ table. However, if I do make a new BQ table and let Dataprep make the schema for the new table, the schema matches the old table.

Is this a bug? Is there a workaround?

enter image description here

enter image description here

PeterH
  • 858
  • 1
  • 6
  • 15
  • 1
    Could you please provide the full error message? You can also check the error in the Dataflow job that was spawned by Dataprep, the dataflow job name should contain the Dataprep job ID. Open the Dataflow job associated with the Dataprep job then see if there's any error. – Christopher May 21 '19 at 08:37
  • So that is one of the issues. I can't even get to the step where I am running a Dataflow job. The error is on the Dataprep `publishing actions` page where you set the bigquery table to load the data into. I added a new picture to give you a bigger picture. – PeterH May 22 '19 at 03:13
  • Dataprep has some data types converstions when the data is loaded into BigQuery (documented in [BigQuery Data Type Conversions](https://cloud.google.com/dataprep/docs/html/BigQuery-Data-Type-Conversions_102563896)), but you already have Strings. What's the information you received when you click on "Show columns that don't match"? Could you confirm the data types directly in BigQuery? – rsantiago Sep 26 '19 at 01:38

0 Answers0