Skip to main content

entry_point

💡Did you know...
Available from dbt v1.11 or with the dbt "Latest" release track.
functions/<filename>.yml
functions:
  - name: <function name>
    config:
      entry_point: <string> # required for Python UDFs

Definition​

When creating Python UDFs, specify the Python function to be called in entry_point.

Python UDFs are currently supported in Snowflake and BigQuery. Each warehouse uses a different name for the entry point function. The following table shows what they’re called:

WarehouseHow entry_point is used
SnowflakeBecomes the HANDLER name in LANGUAGE PYTHON UDF
BigQueryBecomes the entry_point in OPTIONS(...)

Example​

For example, if you have a Python UDF in functions/my_function.py with the following code which uses the function main as the entry point:

functions/my_function.py
import re

def _digits_only(s: str) -> bool:
    return bool(re.search(r'^[0-9]+$', s or ''))

def _to_flag(is_match: bool) -> int:
    return 1 if is_match else 0

def main(a_string: str) -> int:
    """
    This is used as the entry point for the UDF.
    Returns 1 if a_string represents a positive integer (e.g., "10"),
    else 0.
    """
    return _to_flag(_digits_only(a_string))

After defining the UDF, you can specify main as the entry_point in the YAML file. entry_point: main points to the main function as the entry point for the UDF, while _digits_only and _to_flag are helper functions.

functions/schema.yml
functions:
  - name: is_positive_int
    description: Returns 1 if a_string matches ^[0-9]+$, else 0
    config:
      runtime_version: "3.11"    # required
      entry_point: main          # required: points to the function above
    arguments:
      - name: a_string
        data_type: string
    returns:
      data_type: integer

Was this page helpful?

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

0
Edit this page