If you need to export SQL to a file, particularly a CSV you can do so using the
COPY command. The
COPY command has a few different options we will use to achieve this goal, namely
TO will specifiy where to dump the results.
CSV specifies what type of file to create, while
HEADER will make sure the first row of the file is allocated to column headers.
DELIMITER denotes the character to separate the columns from within the file.
Below is an example of the command I tried to use to handle this:
COPY (SELECT * FROM users where users.organization_id='123') TO '/path/to/newfile' WITH DELIMITER ',' CSV HEADER;
I mentioned I “tried to” use this command, as often happens in tech it was not so straightforward much like solving the sql ordering timeout. This time I received an error because I didn’t have root access to the sql server and was met with the error:
ERROR: must be superuser or a member of the pg_write_server_files role to COPY to a file
HINT: Anyone can COPY to stdout or from stdin. psql’s \copy command also works for anyone.
So I listened to the HINT and printed my results to STDOUT in my terminal. This is what the resulting command looked like:
COPY (SELECT * FROM users where users.organization_id='123') TO STDOUT;
Printing to STDOUT is more helpful if you have unlimited scroll back set in your terminal. If you’re using iterm on a Mac you can enable this through: preferences -> profiles -> unlimited scroll back.
Without that you would only see the last 1000 rows printed.