.import¶

This command loads data into the database.

Syntax¶

usage: .import [type] [type options] [path]

The followings are the currently supported import types.

It should be noted that csv, avro and pipe support field position and name based loading, but json only supports name based loading.

See Import Formats for more information.

Loading¶

There are two possible ways of loading data. One is position based, and the other is name based.

Let us use the following CSV as an example.

Name,Value
1,2

Position Based Loading¶

With positioned loading, each ? in SQL is used to count the input field positions.

So for the following example, the first ? corresponds to column Name in the CSV file, and the second ? corresponds to column Value.

CREATE TABLE MyTable (a INTEGER, b INTEGER);
.import csv -h on lib/import1.csv
INSERT INTO MyTable VALUES (?, ?);

-- Check the data
SELECT * FROM MyTable ORDER BY a;
-- success --
A B
- -
1 2

Name Based Loading¶

With name based loading, source field name is put inside {{ and }} to specify the source field name.

.import csv -h on lib/import1.csv
INSERT INTO MyTable VALUES ({{Name}}, {{Value}});

-- Check the data
SELECT * FROM MyTable ORDER BY a;
-- success --
A B
- -
1 2

Batch Loading¶

Jaqy by default, uses batch loading which can improve the loading speed. See .set batchsize for more information.

Example¶

-- Loading data using field position
.import csv -h on lib/import1.csv
INSERT INTO MyTable VALUES (?, ?);

-- Loading data using names
.import csv -h on lib/import1.csv
INSERT INTO MyTable VALUES ({{a}}, {{b}});