target-snowflake target for loading data into Snowflake

pip install target-snowflake==0.2.4


Target Snowflake


PyPI version

A Singer Snowflake target, for use with Singer streams generated by Singer taps.

Snowflake Connector



pip install target-snowflake


  1. Follow the Best Practices for setting up separate tap and target virtualenvs to avoid version conflicts.

  2. Create a config file at ~/ with Snowflake connection information and target Snowflake schema and warehouse.

      "snowflake_account": "",
      "snowflake_username": "myuser",
      "snowflake_role": "myrole",
      "snowflake_password": "1234",
      "snowflake_database": "my_analytics",
      "snowflake_schema": "mytapname",
      "snowflake_warehouse": "dw"

1. Run `target-snowfkajke` against a [Singer]( tap.

 ~/.virtualenvs/tap-something/bin/tap-something \
   | ~/.virtualenvs/target-snowflake/bin/target-snowflake \
     --config ~/ >> state.json

If you are running windows, the following is equivalent:

venvs\tap-exchangeratesapi\Scripts\tap-exchangeratesapi.exe | ^
venvs\target-snowflake\Scripts\target-snowlfake.exe ^
--config target_snowflake_config.json


The fields available to be specified in the config file are specified here.

Field Type Default Details
snowflake_account ["string"] N/A ACCOUNT might require the region and cloud platform where your account is located, in the form of: <your_account_name>.<region_id>.<cloud> (e.g. Refer to Snowflake's documentation about Account
snowflake_username ["string"] N/A
snowflake_password ["string", "null"] null
snowflake_role ["string"] null If not specified, Snowflake will use the user's default role.
snowflake_database ["string"] N/A
snowflake_authenticator ["string"] "snowflake" Speifies the authentication provider for snowflake to use. Valud options are the internal one ("snowflake"), a browser session ("externalbrowser"), or Okta ("https://<your_okta_account_name>"). See the snowflake docs for more details.
snowflake_schema ["string", "null"] "PUBLIC"
snowflake_warehouse ["string"] N/A
invalid_records_detect ["boolean", "null"] true Include false in your config to disable crashing on invalid records
invalid_records_threshold ["integer", "null"] 0 Include a positive value n in your config to allow at most n invalid records per stream before giving up.
disable_collection ["string", "null"] false Include true in your config to disable Singer Usage Logging.
logging_level ["string", "null"] "INFO" The level for logging. Set to DEBUG to get things like queries executed, timing of those queries, etc. See Python's Logger Levels for information about valid values.
persist_empty_tables ["boolean", "null"] False Whether the Target should create tables which have no records present in Remote.
state_support ["boolean", "null"] True Whether the Target should emit STATE messages to stdout for further consumption. In this mode, which is on by default, STATE messages are buffered in memory until all the records that occurred before them are flushed according to the batch flushing schedule the target is configured with.
target_s3 ["object", "null"] N/A When included, use S3 to stage files. See S3 below

S3 Config.json

Field Type Default Details
aws_access_key_id ["string"] N/A
aws_secret_access_key ["string"] N/A
bucket ["string"] N/A Bucket where staging files should be uploaded to.
key_prefix ["string", "null"] "" Prefix for staging file uploads to allow for better delineation of tmp files


  • Snowflake SQL Identifiers:
    • Although Snowflake supports quoted identifiers to have non-alphanumeric values, target-snowflake limits identifiers to uppercase alphanumerics, and underscores
    • This is done to make querability/useability in Snowflake simpler, so as to not require users to have to use sometimes cumbersome quotes to query their data
  • Requires a JSON Schema for every stream.
  • Only string, string with date-time format, integer, number, boolean, object, and array types with or without null are supported. Arrays can have any of the other types listed, including objects as types within items.
    • Example of JSON Schema types that work
      • ['number']
      • ['string']
      • ['string', 'null']
      • ['string', 'integer']
      • ['integer', 'number']
    • Exmaple of JSON Schema types that DO NOT work
      • ['any']
      • ['null']
  • JSON Schema combinations such as anyOf and allOf are not supported.
  • JSON Schema $ref is partially supported:
    • NOTE: The following limitations are known to NOT fail gracefully
    • Presently you cannot have any circular or recursive $refs
    • $refs must be present within the schema:
      • URI's do not work
      • if the $ref is broken, the behaviour is considered unexpected
  • Any values which are the string \\N will be streamed to Snowflake as the literal null


Target Snowflake is sponsored by Data Mill (Data Mill Services, LLC)

Data Mill helps organizations utilize modern data infrastructure and data science to power analytics, products, and services.

Copyright Data Mill Services, LLC 2018