Skip to main content


— copy data from a SQL query to a file


COPY { <table_name> [ (<column_name> [, ...] ) ] | ( <query> ) }
TO <target_location>
[ WITH (<option> [, ...]) ]

where <option> can be one of:

FORMAT => <format_name>
<format_specific_option> => <value>


COPY TO copies the content of a Hyper database table to one or more files, overriding existing files if there are any.

If a column list is specified, only those specified columns will be written, in the order they appear in the column list.

Depending on the target location, Hyper will write data, e.g., to a local file or an AWS S3 bucket. More information on the available locations can be found in External Locations.


The name (optionally database- or schema-qualified) of an existing table.
An optional list of columns to be copied. If no column list is specified, all columns of the table will be copied.
A SELECT command, VALUES command or any other SQL query.
Location to write the data to. See External Location documentation for more information.
FORMAT => format_name
Selects the data format to be written. This option can be omitted in case the format can be inferred from the file extension. Supported formats are depicted in detail in External Formats.
format_specific_option => value
A format-specific option. The available options for each respective format can be found in External Formats.


Copy a table to a CSV file in the working directory of the Hyper server, having a custom delimiter. The schema of the CSV file will be the same as the schema of the table products:

COPY products TO './products.csv' WITH ( FORMAT => 'csv', DELIMITER => '|' )

Same but writing to multiple CSV files (assume the table products is large enough so that the resulting CSV file will be larger than 500000 bytes):

COPY products to './products.csv'
WITH ( FORMAT => 'csv', max_file_size => 500000 )

Copy to an Apache Parquet file to Amazon S3 using empty credentials and inferring the bucket region. The file format is inferred from the file extension:

COPY products TO s3_location(
access_key_id => '',
secret_access_key => ''

Same but with explicit Amazon S3 credentials and bucket region:

COPY products
TO s3_location(
access_key_id => 'ACCESSKEYID12EXAMPLE',
secret_access_key => 'sWfssWSmnME5X/36dsf3G/cbyDzErEXAMPLE123',
region => 'us-east-1'

Instead of copying the table products as a whole, use a query instead to do whatever operation is needed:

COPY (SELECT name FROM products ORDER BY price LIMIT 100)
TO s3_location(
access_key_id => 'ACCESSKEYID12EXAMPLE',
secret_access_key => 'sWfssWSmnME5X/36dsf3G/cbyDzErEXAMPLE123',
region => 'us-east-1'


Files named in a COPY command are written directly by the server, not by the client application. Therefore, they must reside on or be accessible to the database server machine, not the client. They must be accessible to and writable by the Hyper user (the user ID the server runs as), not the client.

COPY input and output is affected by date_style.

COPY stops operation at the first error.

Hyper also supports the PostgreSQL syntax of the COPY command, which is slightly different from the syntax depicted here. This is only supported for PostgreSQL compatibility. When writing SQL for Hyper, we recommend using the syntax documented here.