Categories:

Data metric functions

INVALID_JSON_COUNT (system data metric function)¶

Returns the count of non-NULL column values that are not valid JSON. Use this function to validate columns that are expected to contain JSON payloads.

This topic provides the syntax for calling the function directly. To learn how to associate the function with a table or view so it runs at regular intervals, see Associate a DMF.

Syntax¶

SNOWFLAKE.CORE.INVALID_JSON_COUNT(<query>)

Arguments¶

query

Specifies a SQL query that projects a single column.

Allowed data types¶

The column projected by the query must have the VARCHAR data type.

Returns¶

The function returns a NUMBER value. NULL values are excluded from the count.

Example¶

Measure the number of invalid JSON values in the payload column:

SELECT SNOWFLAKE.CORE.INVALID_JSON_COUNT(
  SELECT
    payload
  FROM events.tables.logs
);

The following example creates a table with sample JSON payloads, and then counts the values that aren’t valid JSON:

CREATE OR REPLACE TABLE json_payloads (payload VARCHAR);

INSERT INTO json_payloads (payload) VALUES
  ('{"id": 1, "status": "active"}'),  -- valid JSON object
  ('[1, 2, 3]'),                      -- valid JSON array
  ('{"id": 2, "status": }'),          -- not valid JSON: missing value
  (NULL);                             -- NULL values aren't counted

SELECT SNOWFLAKE.CORE.INVALID_JSON_COUNT(
  SELECT
    payload
  FROM json_payloads
);

In this example, the function returns 1, for the one value that isn’t valid JSON.