Column data types after ingestion

Hi Druid community, I have a question about determining the data type of a column after it has been batch ingested from S3. When looking at the table in Druid on the query tab all columns have the letter “A” beside them, except the __time column (see Druid_Column_Type.png). Does that mean these are string columns? Is there a way to determine what the column type is?

When looking at a table in AWS Athena that are based off the same parquet files that I ingested it shows different column types (see Athena_Column_Type.png).

I know there are different columns in the 2 pictures but Druid has A (which makes me think string) beside all columns and Athena has different types (which makes it clearer what the type is).

Is there somewhere in the Druid Docs that explains this?

My problem is that when I try to map some data out in Superset, connected to my Druid datasource, using latitude & longitude it is getting messed up because it is expecting a number, and is getting a string.

Cheers,
Donovan

Hi Donovan -

Those do mean string types. Druid supports a few data types - string, long, float, and double (see https://druid.apache.org/docs/latest/tutorials/tutorial-ingestion-spec.html#column-types ). You should be able to edit your ingestion from S3 to specify, eg, long for latitudes and longitudes, if that’s appropriate.