How do you print the result of a PostgreSQL query in CSV or TSV format from the command line?
PostgresqlPostgresql Problem Overview
I'd like to execute a query from the shell (not in the interactive psql client) and have it print the CSV or TSV representation of the output to STDOUT. How do you do that with psql or one of the PostgreSQL command-line tools?
Postgresql Solutions
Solution 1 - Postgresql
If you are using PostgreSQL 8.2 or newer, use this for CSV:
psql -c "COPY (<select query>) TO STDOUT WITH CSV"
and this of TSV, with proper NULLs:
psql -c "COPY (<select query>) TO STDOUT WITH NULL AS ''"
The CSV form will properly quote any fields that contain the double-quote character. See the PostgreSQL documentation of your specific version for more details and options for COPY.
Solution 2 - Postgresql
Starting from Bohemian's answer, I found these flags useful:
psql my_database -U myuser -A -F , -X -t -f /path/to/query.sql -o /path/to/output.csv
- Unaligned output mode: -A
- Use comma as field delimiter: -F ,
- Do not read psqlrc: -X
- Tuples only (no header/footer): -t
- File containing SQL query: -f
- Output file: -o
Solution 3 - Postgresql
EDITED: Using -F
Use commas via -F and use "unaligned table output mode" -A:
psql my_database -U myuser -A -F , -c "select * from mytable"
Solution 4 - Postgresql
To specify a tsv use the delimiter '\t'
psql my_database -U myuser -F'\t' --no-align -f mysqlfile.sql -o outputfile.tsv
To specify a csv use the delimiter ','
psql my_database -U myuser -F',' --no-align -f mysqlfile.sql -o outputfile.csv
Solution 5 - Postgresql
Also possible is the copy command which allows you to specify header, delimiters and quoting options
psql my_database -U my_user -c "copy (select a.id,b.id from my_table_a as a inner join my_table_b as b on b.id = a.id) to STDOUT"
Solution 6 - Postgresql
You can specify the field separator with the -F command line parameter to psql
Solution 7 - Postgresql
Export AS TSV WITH HEADER
You can include the HEADER as follows:
\COPY (SELECT * FROM tca) TO '/.../metab/tca.tsv' WITH DELIMITER E'\t' CSV HEADER;
\COPY (SELECT * FROM tca) TO '/...a/metab/tca.tsv' WITH NULL AS '' DELIMITER E'\t' CSV HEADER;
E.g. (PSQL):
[metabolism]# \COPY (SELECT * FROM tca) TO '/mnt/Vancouver/programming/data/metabolism/tca.tsv' WITH NULL AS '' DELIMITER E'\t' CSV HEADER;
COPY 22
BASH:
[victoria@victoria tsv]$ pwd
/mnt/Vancouver/programming/data/metabolism/tsv
[victoria@victoria tsv]$ head -n3 tca.tsv
uuid src tgt rel rel_type
878b87de-0ca8-49a8-9f77-a24353e251d2 oxalosuccinic acid oxoglutaric acid 1.1.1.42 2
7fd9cf88-495b-491b-956e-294f19097923 isocitric acid oxoglutaric acid 1.1.1.41 2
[victoria@victoria csv]$
Solution 8 - Postgresql
The simplest way (using psql) seems to be by using --csv flag:
psql --csv -c "SELECT * FROM sometable"