Tutorial: Loading JSON data into a relational table (2024)

Introduction¶

When uploading JSON data into a table, you have these options:

• Store JSON objects natively in a VARIANT type column (as shown in Tutorial: Bulk loading from a local file system using COPY).

• Store JSON object natively in an intermediate table and then use FLATTEN function to extract JSON elements into separate columns in a table (as shown in Tutorial: JSON basics for Snowflake)

• Transform JSON elements directly into table columns as shown in this tutorial.

The COPY command in this tutorial uses a SELECT statement to query for individual elements in a staged JSON file.

The example commands provided in this tutorial includes a PUT statement.We recommend executing these commands in SnowSQL which supports the PUT command.Clients such as Snowsight and the Classic Console donot support the PUT command.

Prerequisites¶

For this tutorial you need to:

• Download a Snowflake provided JSON data file.

• Create a database, a table, and a virtual warehouse for this tutorial.

Database, table, and virtual warehouse are basic Snowflake objects required formost Snowflake activities.

{ "location": { "state_city": "MA-Lexington", "zip": "40503" }, "sale_date": "2017-3-5", "price": "275836"}

 create or replace database mydatabase; use schema mydatabase.public;CREATE OR REPLACE TEMPORARY TABLE home_sales ( city STRING, zip STRING, state STRING, type STRING DEFAULT 'Residential', sale_date timestamp_ntz, price STRING );create or replace warehouse mywarehouse with warehouse_size='X-SMALL' auto_suspend = 120 auto_resume = true initially_suspended=true;use warehouse mywarehouse;

Create file format object¶

Execute the CREATE FILE FORMAT commandto create the sf_tut_json_format file format.

CREATE OR REPLACE FILE FORMAT sf_tut_json_format TYPE = JSON;

TYPE = 'JSON' indicates the source file format type. CSV is the default file format type.

Create stage object¶

Execute CREATE STAGE to create theinternal sf_tut_stage stage.

CREATE OR REPLACE TEMPORARY STAGE sf_tut_stage FILE_FORMAT = sf_tut_json_format;

Copy

Similar to temporary tables, temporary stages are automatically droppedat the end of the session.

Stage the data file¶

Execute the PUT command to upload the JSON file from your local file system to thenamed stage.

• Linux or macOS

  See Also

  Convert JSON to CSV Online | Popular Tools | Gigasheet

  PUT file:///tmp/load/sales.json @sf_tut_stage AUTO_COMPRESS=TRUE;

  Copy

• Windows

  PUT file://C:\temp\load\sales.json @sf_tut_stage AUTO_COMPRESS=TRUE;

  Copy

Copy data into the target table¶

Load the sales.json.gz staged data file into the home_sales table.

COPY INTO home_sales(city, state, zip, sale_date, price) FROM (SELECT SUBSTR($1:location.state_city,4), SUBSTR($1:location.state_city,1,2), $1:location.zip, to_timestamp_ntz($1:sale_date), $1:price FROM @sf_tut_stage/sales.json.gz t) ON_ERROR = 'continue';

Note the $1 in the SELECT query refers to the single column where the JSON is stored.The query also uses the following functions:

• The SUBSTR , SUBSTRING function to extract city and state values from state_city JSON key.

• The TO_TIMESTAMP / TO_TIMESTAMP_* to cast the sale_date JSON key value to a timestamp.

Execute the following query to verify data is copied.

SELECT * from home_sales;

Remove the successfully copied data files¶

After you verify that you successfully copied data from your stage into the tables,you can remove data files from the internal stage using the REMOVEcommand to save on data storage.

REMOVE @sf_tut_stage/sales.json.gz;

Copy

Clean up¶

Execute the following DROP <object> commands to return your system to its state before you began the tutorial:

DROP DATABASE IF EXISTS mydatabase;DROP WAREHOUSE IF EXISTS mywarehouse;

Copy

Dropping the database automatically removes all child database objects such as tables.