I am trying to load data into S3(parquet files) from mongoDB using AWS Data Migration Service(DMS). One particular collection has a field which has data type Integer in some documents and Double in other documents.
DMS scans the collection and writes data to parquet files.
Then I run a Glue crawler to generate tables for AWS Athena. The table for this collection gets created with field's data type as integer, and has null values in those documents wherever this field's data type is double.
Is there any way to configure DMS/Athena to convert all the integers to doubles?
I see we can write transformation rules(change column type) for the source data in AWS DMS. However that is not yet supported for mongoDB source.
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…