Standard SQL user-defined functions

BigQuery supports user-defined functions (UDFs). A UDF lets you create a function by using another SQL expression or JavaScript. These functions accept columns of input and perform actions, returning the result of those actions as a value. For information on user-defined functions in legacy SQL, see User-defined functions in legacy SQL.

UDFs can either be persistent or temporary. You can reuse persistent UDFs across multiple queries and temporary UDFs in a single query.

UDF syntax

To create a persistent UDF, use the following syntax:

CREATE [OR REPLACE] FUNCTION [IF NOT EXISTS]
    [[project_name.]dataset_name.]function_name
    ([named_parameter[, ...]])
  [RETURNS data_type]
  { sql_function_definition | javascript_function_definition }

To create a temporary UDF, use the following syntax:

CREATE [OR REPLACE] {TEMPORARY | TEMP} FUNCTION [IF NOT EXISTS]
    function_name
    ([named_parameter[, ...]])
  [RETURNS data_type]
  { sql_function_definition | javascript_function_definition }

named_parameter:
  param_name param_type

sql_function_definition:
  AS (sql_expression)

javascript_function_definition:
  [determinism_specifier]
  LANGUAGE js
  [OPTIONS (library = library_array)]
  AS javascript_code

determinism_specifier:
  { DETERMINISTIC | NOT DETERMINISTIC }

This syntax consists of the following components:

  • CREATE { FUNCTION | OR REPLACE FUNCTION | FUNCTION IF NOT EXISTS }. Creates or updates a function. To replace any existing function with the same name, use the OR REPLACE keyword. To treat the query as successful and take no action if a function with the same name already exists, use the IF NOT EXISTS clause.

  • project_name is the name of the project where you are creating the function. Defaults to the project that runs this DDL query. If the project name contains special characters such as colons, it should be quoted in backticks ` (example: `google.com:my_project`).

  • dataset_name is the name of the dataset where you are creating the function. Defaults to the defaultDataset in the request.

  • named_parameter. Consists of a comma-separated param_name and param_type pair. The value of param_type is a BigQuery data type. For a SQL UDF, the value of param_type can also be ANY TYPE.

  • determinism_specifier. Applies only to JavaScript user-defined functions. Provides a hint to BigQuery as to whether the query result can be cached. Can be one of the following values:

    • DETERMINISTIC: The function always returns the same result when passed the same arguments. The query result is potentially cacheable. For example, if the function add_one(i) always returns i + 1, the function is deterministic.

    • NOT DETERMINISTIC: The function does not always return the same result when passed the same arguments, and therefore is not cacheable. For example, if add_random(i) returns i + rand(), the function is not deterministic and BigQuery will not use cached results.

      If all of the invoked functions are DETERMINISTIC, BigQuery will try to cache the result, unless the results can't be cached for other reasons. For more information, see Using cached query results.

  • [RETURNS data_type]. Specifies the data type that the function returns.

    • If the function is defined in SQL, then the RETURNS clause is optional. If the RETURNS clause is omitted, then BigQuery infers the result type of the function from the SQL function body when a query calls the function.
    • If the function is defined in JavaScript, then the RETURNS clause is required. For more information about allowed values for data_type, see Supported JavaScript UDF data types.
  • AS (sql_expression). Specifies the SQL expression that defines the function.

  • [OPTIONS (library = library_array)]. For a JavaScript UDF, specifies an array of JavaScript libraries to include in the function definition.

  • AS javascript_code. Specifies the definition of a JavaScript function. javascript_code is a string literal.

To delete a persistent user-defined function, use the following syntax:

DROP FUNCTION [IF EXISTS] [[project_name.]dataset_name.]function_name

Temporary UDFs expire as soon as the query finishes, so DROP FUNCTION statements are only supported for temporary UDFs in scripts and procedures.

SQL UDFs

Create SQL UDFs using the following syntax:

CREATE [OR REPLACE] [TEMPORARY | TEMP] FUNCTION [IF NOT EXISTS]
    [[`project_name`.]dataset_name.]function_name
    ([named_parameter[, ...]])
  [RETURNS data_type]
  AS (sql_expression)

named_parameter:
  param_name param_type

Templated SQL UDF parameters

A templated parameter with param_type = ANY TYPE can match more than one argument type when the function is called.

  • If more than one parameter has type ANY TYPE, BigQuery does not enforce any type relationship between these arguments.
  • The function return type cannot be ANY TYPE. It must be either omitted, which means to be automatically determined based on sql_expression, or an explicit type.
  • Passing the function arguments of types that are incompatible with the function definition results in an error at call time.

SQL UDF examples

The following example shows a UDF that employs a SQL function.

CREATE TEMP FUNCTION addFourAndDivide(x INT64, y INT64) AS ((x + 4) / y);
WITH numbers AS
  (SELECT 1 as val
  UNION ALL
  SELECT 3 as val
  UNION ALL
  SELECT 4 as val
  UNION ALL
  SELECT 5 as val)
SELECT val, addFourAndDivide(val, 2) AS result
FROM numbers;

+-----+--------+
| val | result |
+-----+--------+
| 1   | 2.5    |
| 3   | 3.5    |
| 4   | 4      |
| 5   | 4.5    |
+-----+--------+

The following example shows a SQL UDF that uses a templated parameter. The resulting function accepts arguments of various types.

CREATE TEMP FUNCTION addFourAndDivideAny(x ANY TYPE, y ANY TYPE) AS (
  (x + 4) / y
);
SELECT addFourAndDivideAny(3, 4) AS integer_output,
       addFourAndDivideAny(1.59, 3.14) AS floating_point_output;

+----------------+-----------------------+
| integer_output | floating_point_output |
+----------------+-----------------------+
| 1.75           | 1.7802547770700636    |
+----------------+-----------------------+

The following example shows a SQL UDF that uses a templated parameter to return the last element of an array of any type.

CREATE TEMP FUNCTION lastArrayElement(arr ANY TYPE) AS (
  arr[ORDINAL(ARRAY_LENGTH(arr))]
);
SELECT
  names[OFFSET(0)] AS first_name,
  lastArrayElement(names) AS last_name
FROM (
  SELECT ['Fred', 'McFeely', 'Rogers'] AS names UNION ALL
  SELECT ['Marie', 'Skłodowska', 'Curie']
);

+------------+-----------+
| first_name | last_name |
+------------+-----------+
| Fred       | Rogers    |
| Marie      | Curie     |
+------------+-----------+

JavaScript UDFs

Create JavaScript UDFs using the following structure.

CREATE [OR REPLACE] [TEMPORARY | TEMP] FUNCTION [IF NOT EXISTS]
    [[`project_name`.]dataset_name.]function_name
    ([named_parameter[, ...]])
  RETURNS data_type
  [DETERMINISTIC | NOT DETERMINISTIC]
  LANGUAGE js
  [OPTIONS (library = library_array)]
  AS javascript_code

Supported JavaScript UDF data types

Some SQL types have a direct mapping to JavaScript types, but others do not. BigQuery represents types in the following manner:

BigQuery data type JavaScript data type
ARRAY ARRAY
BOOL BOOLEAN
BYTES base64-encoded STRING
FLOAT64 NUMBER
NUMERIC If a NUMERIC value can be represented exactly as an IEEE 754 floating-point value and has no fractional part, it is encoded as a Number. These values are in the range [-253, 253]. Otherwise, it is encoded as a String.
STRING STRING
STRUCT OBJECT where each STRUCT field is a named field
TIMESTAMP DATE with a microsecond field containing the microsecond fraction of the timestamp
DATE DATE

Because JavaScript does not support a 64-bit integer type, INT64 is unsupported as an input type for JavaScript UDFs. Instead, use FLOAT64 to represent integer values as a number, or STRING to represent integer values as a string.

BigQuery does support INT64 as a return type in JavaScript UDFs. In this case, the JavaScript function body can return either a JavaScript Number or a String. BigQuery then converts either of these types to INT64.

If the return value of the JavaScript UDF is a Promise, BigQuery waits for the Promise until it is settled. If the Promise settles into a fulfilled state, BigQuery returns its result. If the Promise settles into a rejected state, BigQuery returns an error.

Quoting rules

You must enclose JavaScript code in quotes. For simple, one line code snippets, you can use a standard quoted string:

CREATE TEMP FUNCTION plusOne(x FLOAT64)
RETURNS FLOAT64
LANGUAGE js
AS "return x+1;";
SELECT val, plusOne(val) AS result
FROM UNNEST([1, 2, 3, 4, 5]) AS val;

+-----------+-----------+
| val       | result    |
+-----------+-----------+
| 1         | 2         |
| 2         | 3         |
| 3         | 4         |
| 4         | 5         |
| 5         | 6         |
+-----------+-----------+

In cases where the snippet contains quotes, or consists of multiple lines, use triple-quoted blocks:

CREATE TEMP FUNCTION customGreeting(a STRING)
RETURNS STRING
LANGUAGE js AS """
  var d = new Date();
  if (d.getHours() < 12) {
    return 'Good Morning, ' + a + '!';
  } else {
    return 'Good Evening, ' + a + '!';
  }
  """;
SELECT customGreeting(names) as everyone
FROM UNNEST(["Hannah", "Max", "Jakob"]) AS names;
+-----------------------+
| everyone              |
+-----------------------+
| Good Morning, Hannah! |
| Good Morning, Max!    |
| Good Morning, Jakob!  |
+-----------------------+

Including JavaScript libraries

You can extend your JavaScript UDFs using the OPTIONS section. This section lets you specify external code libraries for the UDF.

CREATE TEMP FUNCTION myFunc(a FLOAT64, b STRING)
  RETURNS STRING
  LANGUAGE js
  OPTIONS (
    library=["gs://my-bucket/path/to/lib1.js", "gs://my-bucket/path/to/lib2.js"]
  )
  AS
"""
    // Assumes 'doInterestingStuff' is defined in one of the library files.
    return doInterestingStuff(a, b);
""";

SELECT myFunc(3.14, 'foo');

In the preceding example, code in lib1.js and lib2.js is available to any code in the [external_code] section of the UDF.

JavaScript UDF examples

CREATE TEMP FUNCTION multiplyInputs(x FLOAT64, y FLOAT64)
RETURNS FLOAT64
LANGUAGE js AS """
  return x*y;
""";
WITH numbers AS
  (SELECT 1 AS x, 5 as y
  UNION ALL
  SELECT 2 AS x, 10 as y
  UNION ALL
  SELECT 3 as x, 15 as y)
SELECT x, y, multiplyInputs(x, y) as product
FROM numbers;

+-----+-----+--------------+
| x   | y   | product      |
+-----+-----+--------------+
| 1   | 5   | 5            |
| 2   | 10  | 20           |
| 3   | 15  | 45           |
+-----+-----+--------------+

You can pass the result of a UDF as input to another UDF. For example:

CREATE TEMP FUNCTION multiplyInputs(x FLOAT64, y FLOAT64)
RETURNS FLOAT64
LANGUAGE js AS """
  return x*y;
""";
CREATE TEMP FUNCTION divideByTwo(x FLOAT64)
RETURNS FLOAT64
LANGUAGE js AS """
  return x/2;
""";
WITH numbers AS
  (SELECT 1 AS x, 5 as y
  UNION ALL
  SELECT 2 AS x, 10 as y
  UNION ALL
  SELECT 3 as x, 15 as y)
SELECT x,
  y,
  multiplyInputs(divideByTwo(x), divideByTwo(y)) as half_product
FROM numbers;

+-----+-----+--------------+
| x   | y   | half_product |
+-----+-----+--------------+
| 1   | 5   | 1.25         |
| 2   | 10  | 5            |
| 3   | 15  | 11.25        |
+-----+-----+--------------+

The following example sums the values of all fields named foo in the given JSON string.

CREATE TEMP FUNCTION SumFieldsNamedFoo(json_row STRING)
  RETURNS FLOAT64
  LANGUAGE js AS """
function SumFoo(obj) {
  var sum = 0;
  for (var field in obj) {
    if (obj.hasOwnProperty(field) && obj[field] != null) {
      if (typeof obj[field] == "object") {
        sum += SumFoo(obj[field]);
      } else if (field == "foo") {
        sum += obj[field];
      }
    }
  }
  return sum;
}
var row = JSON.parse(json_row);
return SumFoo(row);
""";

WITH Input AS (
  SELECT STRUCT(1 AS foo, 2 AS bar, STRUCT('foo' AS x, 3.14 AS foo) AS baz) AS s, 10 AS foo UNION ALL
  SELECT NULL, 4 AS foo UNION ALL
  SELECT STRUCT(NULL, 2 AS bar, STRUCT('fizz' AS x, 1.59 AS foo) AS baz) AS s, NULL AS foo
)
SELECT
  TO_JSON_STRING(t) AS json_row,
  SumFieldsNamedFoo(TO_JSON_STRING(t)) AS foo_sum
FROM Input AS t;
+---------------------------------------------------------------------+---------+
| json_row                                                            | foo_sum |
+---------------------------------------------------------------------+---------+
| {"s":{"foo":1,"bar":2,"baz":{"x":"foo","foo":3.14}},"foo":10}       | 14.14   |
| {"s":null,"foo":4}                                                  | 4       |
| {"s":{"foo":null,"bar":2,"baz":{"x":"fizz","foo":1.59}},"foo":null} | 1.59    |
+---------------------------------------------------------------------+---------+

For more information on how BigQuery data types map to JavaScript types, see Supported JavaScript UDF data types

Best practices for JavaScript UDFs

Prefilter your input

If your input can be easily filtered down before being passed to a JavaScript UDF, your query will likely be faster and cheaper.

Avoid persistent mutable state

Do not store or access mutable state across JavaScript UDF calls.

Use memory efficiently

The JavaScript processing environment has limited memory available per query. JavaScript UDF queries that accumulate too much local state might fail due to memory exhaustion.

Running a query with a UDF

Using the Cloud Console

You can use the Cloud Console to run queries using one or more UDFs.

  1. Click Compose new query.
  2. In the Query Editor pane, enter the UDF statement. For example:

    CREATE TEMPORARY FUNCTION timesTwo(x FLOAT64)
    RETURNS FLOAT64
      LANGUAGE js AS """
      return x*2;
    """;
  3. Below the UDF statement, enter your SQL query. For example:

    SELECT timesTwo(numbers) AS doubles
    FROM UNNEST([1, 2, 3, 4, 5]) AS numbers;
  4. Click Run query. The query results display underneath the buttons.

Using the bq command-line tool

You can use the bq command-line tool from the Cloud SDK to run a query containing one or more UDFs.

Use the following syntax to run a query with a UDF:

bq query <statement_with_udf_and_query>

Authorized UDFs

An authorized UDF is a UDF that is authorized to access a particular dataset. The UDF can query tables in the dataset, even if the user who calls the UDF does not have access to those tables.

Authorized UDFs let you share query results with particular users or groups without giving those users or groups access to the underlying tables. For example, an authorized UDF can compute an aggregation over data or look up a table value and use that value in a computation.

To authorize a UDF, you can use the Google Cloud Console, the REST API, or the bq command-line tool:

Console

  1. Open the BigQuery page in the Cloud Console.

    Go to the BigQuery page

  2. In the navigation panel, in the Resources section, expand your project and select a dataset.

  3. In the details panel, click Authorize Routines.

  4. In the Authorized routines page, in the Authorize routine section, select the project ID, dataset ID, and routine ID for the UDF that you want to authorize.

  5. Click Add authorization.

API

  1. Call the datasets.get method to fetch the dataset that you want the UDF to access. The response body contains a representation of the Dataset resource.

  2. Add the following JSON object to the access array in the Dataset resource:

    {
     "routine": {
       "datasetId": "DATASET_NAME",
       "projectId": "PROJECT_ID",
       "routineId": "ROUTINE_NAME"
     }
    }
    

    Where:

    • DATASET_NAME is the name of the dataset that contains the UDF.
    • PROJECT_ID is the project ID of the project that contains the UDF.
    • ROUTINE_NAME is the name of the UDF.
  3. Call the dataset.update method with the modified Dataset representation.

bq

  1. Use the bq show command to get the JSON representation of the dataset that you want the UDF to access. The output from the command is a JSON representation of the Dataset resource. Save the result to a local file.

    bq show --format=prettyjson TARGET_DATASET > dataset.json
    

    Replace TARGET_DATASET with the name of the dataset that the UDF will have access to.

  2. Edit the file to add the following JSON object to the access array in the Dataset resource:

    {
     "routine": {
       "datasetId": "DATASET_NAME",
       "projectId": "PROJECT_ID",
       "routineId": "ROUTINE_NAME"
     }
    }
    

    Where:

    • DATASET_NAME is the name of the dataset that contains the UDF.
    • PROJECT_ID is the project ID of the project that contains the UDF.
    • ROUTINE_NAME is the name of the UDF.
  3. Use the bq update command to update the dataset.

    bq update --source dataset.json TARGET_DATASET
    

Authorized UDF example

The following is an end-to-end example of creating and using an authorized UDF.

  1. Create two datasets named private_dataset and public_dataset. For more information about creating a dataset, see Creating a dataset.

  2. Run the following statement to create a table named private_table in private_dataset:

    CREATE OR REPLACE TABLE private_dataset.private_table
    AS SELECT key FROM UNNEST(['key1', 'key1','key2','key3']) key;
    
  3. Run the following statement to create a UDF named count_key in public_dataset. The UDF includes a SELECT statement on private_table.

    CREATE OR REPLACE FUNCTION public_dataset.count_key(input_key STRING)
    RETURNS INT64
    AS
    ((SELECT COUNT(1) FROM private_dataset.private_table t WHERE t.key = input_key));
    
  4. Grant the bigquery.dataViewer role to a user on the public_dataset dataset. This role includes the bigquery.routines.get permission, which lets the user call the function. For information about how to assign access controls to a dataset, see Controlling access to datasets.

  5. At this point, the user has permission to call the count_key function but cannot access the table in private_dataset. If the user tries to call the function, they get an error message similar to the following:

    Access Denied: Table myproject:private_dataset.private_table: User does
    not have permission to query table myproject:private_dataset.private_table.
    
  6. Using the bq command-line tool, run the show command as follows:

    bq show --format=prettyjson private_dataset > dataset.json
    

    The output is saved to a local file named dataset.json.

  7. Edit dataset.json to add the following JSON object to the access array:

    {
     "routine": {
       "datasetId": "public_dataset",
       "projectId": "PROJECT_ID",
       "routineId": "count_key"
     }
    }
    

    Replace PROJECT_ID with the project ID for public_dataset.

  8. Using the bq command-line tool, run the update command as follows:

    bq update --source dataset.json private_dataset
    
  9. To verify that the UDF has access to private_dataset, the user can run the following query:

    SELECT public_dataset.count_key('key1');
    

Adding descriptions to UDFs

To add a description to a UDF, follow these steps:

Console

  1. Open the BigQuery web UI in the Cloud Console.

    Go to the Cloud Console

  2. In the Resources pane, select your function.

  3. In the Details pane, click the pencil icon next to Description to edit the description text.

  4. In the dialog, enter a description in the box or edit the existing description. Click Update to save the new description text.

Alternatively, you can use a Standard SQL query to update the description using the description parameter of the OPTIONS field. In the Query editor box, enter your function definition, then add the following line:

OPTIONS (description="DESCRIPTION") AS """

Replace DESCRIPTION with the description you would like to add.

bq

Using the bq query syntax from UDFs and the bq command line tool, you can edit a function's description from the command line. Specify standard SQL with a --nouse_legacy_sql or -- use_legacy_sql=false flag, then enter your function definition. Add the following line to your definition to set the description parameter in the OPTIONS field:

OPTIONS (description="DESCRIPTION") AS """

Replace DESCRIPTION with the description you would like to add.

Limitations

The following limitations apply to temporary and persistent user-defined functions:

  • The DOM objects Window, Document, and Node, and functions that require them, are not supported.
  • JavaScript functions that rely on native code are not supported.
  • A JavaScript UDF can time out and prevent your query from completing. Timeouts can be as short as 5 minutes, but can vary depending on several factors, including how much user CPU time your function consumes and how large your inputs and outputs to the JavaScript function are.
  • Bitwise operations in JavaScript handle only the most significant 32 bits.
  • UDFs are subject to certain rate limits and quota limits. For more information, see UDF limits.

The following limitations apply to persistent user-defined functions:

  • Each dataset can only contain one persistent UDF with the same name. However, you can create a UDF whose name is the same as the name of a table in the same dataset.
  • When referencing a persistent UDF from another persistent UDF or a logical view, you must qualify the name with the dataset. For example:
    CREATE FUNCTION mydataset.referringFunction() AS (mydataset.referencedFunction());

The following limitations apply to temporary user-defined functions.

  • When creating a temporary UDF, function_name cannot contain periods.
  • Views and persistent UDFs cannot reference temporary UDFs.