Copy data from SQL to Google Cloud Storage in JSON, CSV, or Parquet format.

View on GitHub

Last Updated: Jan. 23, 2023

Access Instructions

Install the Google provider package into your Airflow environment.

Import the module into your DAG file and instantiate it with your desired params.


sqlRequiredThe SQL to execute.
bucketRequiredThe bucket to upload to.
filenameRequiredThe filename to use as the object name when uploading to Google Cloud Storage. A {} should be specified in the filename to allow the operator to inject file numbers in cases where the file is split due to size.
schema_filenameIf set, the filename to use as the object name when uploading a .json file containing the BigQuery schema fields for the table that was dumped from the database.
approx_max_file_size_bytesThis operator supports the ability to split large table dumps into multiple files (see notes in the filename param docs above). This param allows developers to specify the file size of the splits. Check to see the maximum allowed file size for a single object.
export_formatDesired format of files to be exported. (json, csv or parquet)
stringify_dictWhether to dump Dictionary type objects (such as JSON columns) as a string. Applies only to CSV/JSON export format.
field_delimiterThe delimiter to be used for CSV files.
null_markerThe null marker to be used for CSV files.
gzipOption to compress file for upload (does not apply to schemas).
schemaThe schema to use, if any. Should be a list of dict or a str. Pass a string if using Jinja template, otherwise, pass a list of dict. Examples could be seen: /schemas#specifying_a_json_schema_file
gcp_conn_id(Optional) The connection ID used to connect to Google Cloud.
delegate_toThe account to impersonate using domain-wide delegation of authority, if any. For this to work, the service account making the request must have domain-wide delegation enabled.
parametersa parameters dict that is substituted at query runtime.
impersonation_chainOptional service account to impersonate using short-term credentials, or chained list of accounts required to get the access_token of the last account in the list, which will be impersonated in the request. If set as a string, the account must grant the originating account the Service Account Token Creator IAM role. If set as a sequence, the identities from the list must grant Service Account Token Creator IAM role to the directly preceding identity, with first account from the list granting this role to the originating account (templated).
upload_metadatawhether to upload the row count metadata as blob metadata
exclude_columnsset of columns to exclude from transmission
partition_columnslist of columns to use for file partitioning. In order to use this parameter, you must sort your dataset by partition_columns. Do this by passing an ORDER BY clause to the sql query. Files are uploaded to GCS as objects with a hive style partitioning directory structure (templated).
write_on_emptyOptional parameter to specify whether to write a file if the export does not return any rows. Default is False so we will not write a file if the export returns no rows.


Copy data from SQL to Google Cloud Storage in JSON, CSV, or Parquet format.

Was this page helpful?