读取以逗号作为定界符并在psql中转义引号的csv文件

Question

I want to read a csv file which is separated by comma ( , ) but want to ignore comma within the double quotes ( "" ). I want to store the result into a table.

Example:

abc,00.000.00.00,00:00:00:00:00:00,Sun Nov 01 00:00:00 EST 0000,Sun Nov 01 00:00:00 EST 0000,"Apple, Inc.",abcd-0000abc-a,abcd-abcd-a0000-00

Here I don't want to split on Apple, .

I know there exists csv reader in python and I can use it in plpython but that's slow considering millions of such strings! I would like a pure psql method!

Answer 1

Here is an example of reading a CSV file with an External Table using the CSV format.

CREATE EXTERNAL TABLE ext_expenses ( name text, 
date date,  amount float4, category text, desc1 text ) 
LOCATION ('gpfdist://etlhost-1:8081/*.txt', 
          'gpfdist://etlhost-2:8082/*.txt')
FORMAT 'CSV' ( DELIMITER ',' )
LOG ERRORS SEGMENT REJECT LIMIT 5;

This was taken from the Greenplum docs too.

http://gpdb.docs.pivotal.io/530/admin_guide/external/g-example-4-single-gpfdist-instance-with-error-logging.html