This repository was archived by the owner on Mar 6, 2026. It is now read-only.
Setting job_config.source_format is ignored by load_table_from_dataframe #261
Currently on PARQUET is supported. If someone sets the format to something other than that, we should probably raise a ValueError.
Might want to consider support CSV / JSON / AVRO, but also don't want to explode the number of (already numerous) code paths we have for DataFrame loading.