I am bulk inserting data into crateDB using
pandas.to_sql(tableId, 'crate://xxxxxxx:4200', if_exists='append', index=False, chunksize=20000)
The data contains a few faulty rows where one column of type 'numeric' contains a value of type 'object' / 'string'.
This results in a typecasting error while inserting.
When I set the chunksize parameter of to_sql() to 1, an exception is raised and I can catch the issue.
When chunksize is > 0, to_sql()
just continues and I cannot make sure that data was inserted properly.
What am I missing, and what would be a good approach to fix this problem?