The first input value specified via the ON clause must be character based (for example, a CHAR/VARCHAR/CLOB or a DATASET STORAGE FORMAT CSV value) and be composed of data that conforms to the specified format (that is, it uses the specified field/record delimiters, or defaults if not specified). Any subsequent input values do not affect the result of conversion to Avro or JSON, but are passed through the table operator and given as additional output columns. This allows the resulting Avro or JSON documents to be associated with the source CSV data, which is very important when converting multiple CSV inputs.
CT my_table(id int, csvData VARCHAR(500));
INSERT INTO my_table(1, 'a,b,c,d,e,f\1,2,3,4,5,6\7,8,9,10,11,12');
INSERT INTO my_table(2, 'a,b,c,d,e,f\13,14,15,16,17,18
\19,20,21,22,23,24');
INSERT INTO my_table(3, 'a,b,c,d,e,f\25,26,27,28,29,30
\31,32,33,34,35,36');
SELECT id, data.toJSON() FROM CSV_TO_AVRO
(
ON (SELECT csvData, id FROM my_table)
USING SCHEMA('{"record_delimiter":"\\"}')
) as csvAvro
ORDER BY id, data.a;
id data.TOJSON()
----------- --------------------------------------------------------------
1 {"a":"1","b":"2","c":"3","d":"4","e":"5","f":"6"}
1 {"a":"7","b":"8","c":"9","d":"10","e":"11","f":"12"}
2 {"a":"13","b":"14","c":"15","d":"16","e":"17","f":"18"}
2 {"a":"19","b":"20","c":"21","d":"22","e":"23","f":"24"}
3 {"a":"25","b":"26","c":"27","d":"28","e":"29","f":"30"}
3 {"a":"31","b":"32","c":"33","d":"34","e":"35","f":"36"}