One of the quieter releases into public preview is that we can now infer schemas for CSVs. To test this follow the following instructions:
- Create an internal stage
create stage frosty_stage;
- Create a csv file format
create or replace file format frosty_csv
field_optionally_enclosed_by = '"'
skip_header = 1; - PUT the following csv file into that internal stage. Download Here
Now you need to infer the schema and produce the following:

Remember, if you want to participate:
Another week, more fun. Cheers
I actually went on to create the table with the inferred schema and loaded the data by parsing the csv header
thanks for the challenge — had to `put` the file outside my solution, but good to explore these new functionalities (also useful to combina with other metadata functions!)