[Optional] NOS generates the ROWFORMAT clause.
For JSON data, encoding_format is:
{"record_delimiter":"\n", "character_set":"cs_value"}
For CSV data, encoding_format is:
{"field_delimiter":"fd_value", "record_delimiter":"\n", "character_set":"cs_value"}
For Parquet data, encoding_format is:
{"character_set":"cs_value"}
If the Payload column is defined for a CSV file, and its data type is DATASET CSV WITH SCHEMA, the schema specification takes precedence over ROWFORMAT.
- record_delimiter
- Specify that the record delimiter is the line feed character, "\n". The key name and "\n" and case-sensitive.
- character_set
- Specify the field character set. The key name is case-sensitive, but cs_value is not.
- field_delimiter
- Specify the field delimiter. The key name and fd_value are case-sensitive.