I'm trying to copy data from excel/csv to a Snowflake database. Within my files are datetime columns that are being stored in snowflake as varchars. Is there some way for Airbyte to recognize that these are timetamps and save them as such?
1. I can't fix this with dbt - we're planning on using the api to allow clients to create new tables and if we have to insert a manual step in the middle the flow breaks down
2. I tried adding {"parse_dates": true} to the reader options to see if I could get pandas to recognize the dates. I've also tried to manually set the columns to datetimes, and that doesn't work as well {"dtype": {"trans_date":"datetime64"}}.
a. This does work when I just try to read the file using pandas.read_csv(), i.e. in a python interpreter and not through Airbyte
3. I tried dumping to postgres instead of Snowflake to see if it was a destination issue, and it was a varchar there as well
Any ideas? Thanks