Using ogr2ogr to load a CSV into Postgres

README.md

The absolute easiest way to get a CSV into a postgresql table is to use ogr2ogr with AUTODETECT_TYPE=YES.

I learned a while back that this is what cartoDB uses to import your CSV into postgis (with a lot of other parameters added)

ogr2ogr -f PostgreSQL PG:"host=localhost user=postgres dbname=postgres password=password"  docs.csv -oo AUTODETECT_TYPE=YES

agreed. Works well.
Furthermore, if you wanted to only load certain columns from the CSV, you could pass an -sql flag like this:
PostgreSQL PG:"host=localhost user=postgres dbname=postgres password=password" docs.csv -oo AUTODETECT_TYPE=YES -sql "SELECT col1, col2 FROM docs"
Notice how you must omit the file extension (.csv) from the SQL query.

This interestingly doesn't work with CSVs with only one column. You have to "trick" it by adding a comma right after the first column in the header or by adding a stub second column in the header.

This interestingly doesn't work with CSVs with only one column. You have to "trick" it by adding a comma right after the first column in the header or by adding a stub second column in the header.

Sure, but you’d never need to use the -sql filter if the CSV only contains 1 column.