def execute(self, context): hook = GoogleCloudStorageHook( google_cloud_storage_conn_id=self.google_cloud_storage_conn_id, delegate_to=self.delegate_to) if '*' in self.source_object: wildcard_position = self.source_object.index('*') objects = hook.list( self.source_bucket, prefix=self.source_object[:wildcard_position], delimiter=self.source_object[wildcard_position + 1:]) for source_object in objects: self.log.info('Executing copy of gs://{0}/{1} to ' 'gs://{2}/{3}/{1}'.format( self.source_bucket, source_object, self.destination_bucket, self.destination_object, source_object)) hook.copy( self.source_bucket, source_object, self.destination_bucket, "{}/{}".format(self.destination_object, source_object)) if self.move_object: hook.delete(self.source_bucket, source_object) else: self.log.info('Executing copy: %s, %s, %s, %s', self.source_bucket, self.source_object, self.destination_bucket or self.source_bucket, self.destination_object or self.source_object) hook.copy(self.source_bucket, self.source_object, self.destination_bucket, self.destination_object) if self.move_object: hook.delete(self.source_bucket, self.source_object)
def execute(self, context): hook = GoogleCloudStorageHook( google_cloud_storage_conn_id=self.google_cloud_storage_conn_id, delegate_to=self.delegate_to ) if '*' in self.source_object: wildcard_position = self.source_object.index('*') objects = hook.list(self.source_bucket, prefix=self.source_object[:wildcard_position], delimiter=self.source_object[wildcard_position + 1:]) for source_object in objects: self.log.info('Executing copy of gs://{0}/{1} to ' 'gs://{2}/{3}/{1}'.format(self.source_bucket, source_object, self.destination_bucket, self.destination_object, source_object)) hook.copy(self.source_bucket, source_object, self.destination_bucket, "{}/{}".format(self.destination_object, source_object)) if self.move_object: hook.delete(self.source_bucket, source_object) else: self.log.info('Executing copy: %s, %s, %s, %s', self.source_bucket, self.source_object, self.destination_bucket or self.source_bucket, self.destination_object or self.source_object) hook.copy(self.source_bucket, self.source_object, self.destination_bucket, self.destination_object) if self.move_object: hook.delete(self.source_bucket, self.source_object)
def execute(self, context): self.log.info('Exporting data to Cloud Storage bucket ' + self.bucket) if self.overwrite_existing and self.namespace: gcs_hook = GoogleCloudStorageHook(self.cloud_storage_conn_id) objects = gcs_hook.list(self.bucket, prefix=self.namespace) for o in objects: gcs_hook.delete(self.bucket, o) ds_hook = DatastoreHook(self.datastore_conn_id, self.delegate_to) result = ds_hook.export_to_storage_bucket( bucket=self.bucket, namespace=self.namespace, entity_filter=self.entity_filter, labels=self.labels) operation_name = result['name'] result = ds_hook.poll_operation_until_done( operation_name, self.polling_interval_in_seconds) state = result['metadata']['common']['state'] if state != 'SUCCESSFUL': raise AirflowException( 'Operation failed: result={}'.format(result)) return result
class GoogleDisplayVideo360ERFToBigQueryOperator(BaseOperator): """Upload Multiple Entity Read Files to specified big query dataset. """ def __init__(self, gcp_conn_id='google_cloud_default', report_body=None, yesterday=False, entity_type=None, file_creation_date=None, cloud_project_id=None, bq_table=None, schema=None, gcs_bucket=None, erf_bucket=None, partner_ids=[], write_disposition='WRITE_TRUNCATE', *args, **kwargs): super(GoogleDisplayVideo360ERFToBigQueryOperator, self).__init__(*args, **kwargs) self.gcp_conn_id = gcp_conn_id self.service = None self.bq_hook = None self.gcs_hook = None self.report_body = report_body self.erf_bucket = erf_bucket self.yesterday = yesterday self.cloud_project_id = cloud_project_id self.bq_table = bq_table self.gcs_bucket = gcs_bucket self.schema = schema self.entity_type = entity_type self.erf_object = 'entity/%s.0.%s.json' % (file_creation_date, entity_type) self.partner_ids = partner_ids self.write_disposition = write_disposition self.file_creation_date = file_creation_date def execute(self, context): if self.gcs_hook is None: self.gcs_hook = GoogleCloudStorageHook( google_cloud_storage_conn_id=self.gcp_conn_id) if self.bq_hook is None: self.bq_hook = BigQueryHook(bigquery_conn_id=self.gcp_conn_id) for i, partner_id in enumerate(self.partner_ids): filename = erf_utils.download_and_transform_erf(self, partner_id) entity_read_file_ndj = 'gs://%s/%s' % (self.gcs_bucket, filename) if i > 0: self.write_disposition = 'WRITE_APPEND' bq_base_cursor = self.bq_hook.get_conn().cursor() bq_base_cursor.run_load( destination_project_dataset_table=self.bq_table, schema_fields=self.schema, source_uris=[entity_read_file_ndj], source_format='NEWLINE_DELIMITED_JSON', write_disposition=self.write_disposition) self.gcs_hook.delete(self.gcs_bucket, filename)
def move_objects(source_bucket=None, destination_bucket=None, prefix=None, **kwargs): storage_objects = kwargs["ti"].xcom_pull(task_ids="list_files") hook = GoogleCloudStorageHook() for storage_object in storage_objects: destination_object = storage_object if prefix: destination_object = "{}/{}".format(prefix, storage_object) hook.copy(source_bucket, storage_object, destination_bucket, destination_object) hook.delete(source_bucket, storage_object)
def execute(self, context): hook = GoogleCloudStorageHook( google_cloud_storage_conn_id=self.google_cloud_storage_conn_id, delegate_to=self.delegate_to) log_message = 'Executing copy of gs://{0}/{1} to gs://{2}/{3}' if self.wildcard in self.source_object: prefix, delimiter = self.source_object.split(self.wildcard, 1) objects = hook.list(self.source_bucket, prefix=prefix, delimiter=delimiter) for source_object in objects: if self.last_modified_time is not None: # Check to see if object was modified after last_modified_time if hook.is_updated_after(self.source_bucket, source_object, self.last_modified_time): pass else: continue if self.destination_object is None: destination_object = source_object else: destination_object = source_object.replace( prefix, self.destination_object, 1) self.log.info( log_message.format(self.source_bucket, source_object, self.destination_bucket, destination_object)) hook.rewrite(self.source_bucket, source_object, self.destination_bucket, destination_object) if self.move_object: hook.delete(self.source_bucket, source_object) else: if self.last_modified_time is not None: if hook.is_updated_after(self.source_bucket, self.source_object, self.last_modified_time): pass else: return self.log.info( log_message.format( self.source_bucket, self.source_object, self.destination_bucket or self.source_bucket, self.destination_object or self.source_object)) hook.rewrite(self.source_bucket, self.source_object, self.destination_bucket, self.destination_object) if self.move_object: hook.delete(self.source_bucket, self.source_object)
def execute(self, context): hook = GoogleCloudStorageHook( google_cloud_storage_conn_id=self.google_cloud_storage_conn_id, delegate_to=self.delegate_to ) log_message = 'Executing copy of gs://{0}/{1} to gs://{2}/{3}' if self.wildcard in self.source_object: prefix, delimiter = self.source_object.split(self.wildcard, 1) objects = hook.list(self.source_bucket, prefix=prefix, delimiter=delimiter) for source_object in objects: if self.last_modified_time is not None: # Check to see if object was modified after last_modified_time if hook.is_updated_after(self.source_bucket, source_object, self.last_modified_time): pass else: continue if self.destination_object is None: destination_object = source_object else: destination_object = source_object.replace(prefix, self.destination_object, 1) self.log.info( log_message.format(self.source_bucket, source_object, self.destination_bucket, destination_object) ) hook.rewrite(self.source_bucket, source_object, self.destination_bucket, destination_object) if self.move_object: hook.delete(self.source_bucket, source_object) else: if self.last_modified_time is not None: if hook.is_updated_after(self.source_bucket, self.source_object, self.last_modified_time): pass else: return self.log.info( log_message.format(self.source_bucket, self.source_object, self.destination_bucket or self.source_bucket, self.destination_object or self.source_object) ) hook.rewrite(self.source_bucket, self.source_object, self.destination_bucket, self.destination_object) if self.move_object: hook.delete(self.source_bucket, self.source_object)
def execute(self, context): hook = GoogleCloudStorageHook( google_cloud_storage_conn_id=self.gcp_conn_id, delegate_to=self.delegate_to) if self.objects: objects = self.objects else: objects = hook.list(bucket_name=self.bucket_name, prefix=self.prefix) self.log.info("Deleting %s objects from %s", len(objects), self.bucket_name) for object_name in objects: hook.delete(bucket_name=self.bucket_name, object_name=object_name)
def clean_buckets(google_cloud_storage_conn_id, private_bucket, shared_bucket): import logging hook = GoogleCloudStorageHook(google_cloud_storage_conn_id) total = 0 # clean the entire bucket private = [(private_bucket, name) for name in hook.list(private_bucket)] shared = [(shared_bucket, name) for name in hook.list(shared_bucket)] for bucket_name, object_name in private + shared: logging.info(f"Deleting gs://{bucket_name}/{object_name}") hook.delete(bucket_name, object_name) total += 1 logging.info(f"Deleted {total} objects")
def move_objects(source_bucket=None, destination_bucket=None, prefix=None, **kwargs): storage_objects = kwargs['ti'].xcom_pull(task_ids='list_files') hook = GoogleCloudStorageHook() for storage_object in storage_objects: destination_object = storage_object if prefix: destination_object = f'{prefix}/{storage_object}' hook.copy(source_bucket, storage_object, destination_bucket, destination_object) hook.delete(source_bucket, storage_object)
def cleanup(**kwargs): ti = kwargs['ti'] gcs = GoogleCloudStorageHook() fileName = ti.xcom_pull(task_ids="censor") gcs.delete("workflowstorage", fileName) fileName = ti.xcom_pull(task_ids="compression") gcs.delete("workflowstorage", fileName) fileName = ti.xcom_pull(task_ids="conversion") gcs.delete("workflowstorage", fileName) fileName = ti.xcom_pull(task_ids="text2speech") gcs.delete("workflowstorage", fileName)
def execute(self, context): hook = GoogleCloudStorageHook( google_cloud_storage_conn_id=self.google_cloud_storage_conn_id, delegate_to=self.delegate_to) log_message = 'Executing copy of gs://{0}/{1} to gs://{2}/{3}' if self.wildcard in self.source_object: prefix, delimiter = self.source_object.split(self.wildcard, 1) objects = hook.list(self.source_bucket, prefix=prefix, delimiter=delimiter) for source_object in objects: if self.destination_object is None: destination_object = source_object else: destination_object = source_object.replace( prefix, self.destination_object, 1) self.log.info( log_message.format(self.source_bucket, source_object, self.destination_bucket, destination_object)) hook.copy(self.source_bucket, source_object, self.destination_bucket, destination_object) if self.move_object: hook.delete(self.source_bucket, source_object) else: self.log.info( log_message.format( self.source_bucket, self.source_object, self.destination_bucket or self.source_bucket, self.destination_object or self.source_object)) hook.copy(self.source_bucket, self.source_object, self.destination_bucket, self.destination_object) if self.move_object: hook.delete(self.source_bucket, self.source_object)
def execute(self, context): hook = GoogleCloudStorageHook( google_cloud_storage_conn_id=self.google_cloud_storage_conn_id, delegate_to=self.delegate_to ) log_message = 'Executing copy of gs://{0}/{1} to gs://{2}/{3}' if self.wildcard in self.source_object: prefix, delimiter = self.source_object.split(self.wildcard, 1) objects = hook.list(self.source_bucket, prefix=prefix, delimiter=delimiter) for source_object in objects: if self.destination_object is None: destination_object = source_object else: destination_object = source_object.replace(prefix, self.destination_object, 1) self.log.info( log_message.format(self.source_bucket, source_object, self.destination_bucket, destination_object) ) hook.rewrite(self.source_bucket, source_object, self.destination_bucket, destination_object) if self.move_object: hook.delete(self.source_bucket, source_object) else: self.log.info( log_message.format(self.source_bucket, self.source_object, self.destination_bucket or self.source_bucket, self.destination_object or self.source_object) ) hook.rewrite(self.source_bucket, self.source_object, self.destination_bucket, self.destination_object) if self.move_object: hook.delete(self.source_bucket, self.source_object)
def execute(self, context): self.log.info('Exporting data to Cloud Storage bucket ' + self.bucket) if self.overwrite_existing and self.namespace: gcs_hook = GoogleCloudStorageHook(self.cloud_storage_conn_id) objects = gcs_hook.list(self.bucket, prefix=self.namespace) for o in objects: gcs_hook.delete(self.bucket, o) ds_hook = DatastoreHook(self.datastore_conn_id, self.delegate_to) result = ds_hook.export_to_storage_bucket(bucket=self.bucket, namespace=self.namespace, entity_filter=self.entity_filter, labels=self.labels) operation_name = result['name'] result = ds_hook.poll_operation_until_done(operation_name, self.polling_interval_in_seconds) state = result['metadata']['common']['state'] if state != 'SUCCESSFUL': raise AirflowException('Operation failed: result={}'.format(result)) return result
class GcsToGDriveOperator(BaseOperator): """ Copies objects from a Google Cloud Storage service service to Google Drive service, with renaming if requested. Using this operator requires the following OAuth 2.0 scope: .. code-block:: none https://www.googleapis.com/auth/drive :param source_bucket: The source Google Cloud Storage bucket where the object is. (templated) :type source_bucket: str :param source_object: The source name of the object to copy in the Google cloud storage bucket. (templated) You can use only one wildcard for objects (filenames) within your bucket. The wildcard can appear inside the object name or at the end of the object name. Appending a wildcard to the bucket name is unsupported. :type source_object: str :param destination_object: The destination name of the object in the destination Google Drive service. (templated) If a wildcard is supplied in the source_object argument, this is the prefix that will be prepended to the final destination objects' paths. Note that the source path's part before the wildcard will be removed; if it needs to be retained it should be appended to destination_object. For example, with prefix ``foo/*`` and destination_object ``blah/``, the file ``foo/baz`` will be copied to ``blah/baz``; to retain the prefix write the destination_object as e.g. ``blah/foo``, in which case the copied file will be named ``blah/foo/baz``. :type destination_object: str :param move_object: When move object is True, the object is moved instead of copied to the new location. This is the equivalent of a mv command as opposed to a cp command. :type move_object: bool :param gcp_conn_id: (Optional) The connection ID used to connect to Google Cloud Platform. :type gcp_conn_id: str :param delegate_to: The account to impersonate, if any. For this to work, the service account making the request must have domain-wide delegation enabled. :type delegate_to: str """ template_fields = ("source_bucket", "source_object", "destination_object") ui_color = "#f0eee4" @apply_defaults def __init__(self, source_bucket, source_object, destination_object=None, move_object=False, gcp_conn_id="google_cloud_default", delegate_to=None, *args, **kwargs): super(GcsToGDriveOperator, self).__init__(*args, **kwargs) self.source_bucket = source_bucket self.source_object = source_object self.destination_object = destination_object self.move_object = move_object self.gcp_conn_id = gcp_conn_id self.delegate_to = delegate_to self.gcs_hook = None # type: Optional[GoogleCloudStorageHook] self.gdrive_hook = None # type: Optional[GoogleDriveHook] def execute(self, context): self.gcs_hook = GoogleCloudStorageHook( google_cloud_storage_conn_id=self.gcp_conn_id, delegate_to=self.delegate_to) self.gdrive_hook = GoogleDriveHook(gcp_conn_id=self.gcp_conn_id, delegate_to=self.delegate_to) if WILDCARD in self.source_object: total_wildcards = self.source_object.count(WILDCARD) if total_wildcards > 1: error_msg = ( "Only one wildcard '*' is allowed in source_object parameter. " "Found {} in {}.".format(total_wildcards, self.source_object)) raise AirflowException(error_msg) prefix, delimiter = self.source_object.split(WILDCARD, 1) objects = self.gcs_hook.list(self.source_bucket, prefix=prefix, delimiter=delimiter) for source_object in objects: if self.destination_object is None: destination_object = source_object else: destination_object = source_object.replace( prefix, self.destination_object, 1) self._copy_single_object(source_object=source_object, destination_object=destination_object) else: self._copy_single_object( source_object=self.source_object, destination_object=self.destination_object) def _copy_single_object(self, source_object, destination_object): self.log.info( "Executing copy of gs://%s/%s to gdrive://%s", self.source_bucket, source_object, destination_object, ) with tempfile.NamedTemporaryFile() as file: filename = file.name self.gcs_hook.download(bucket=self.source_bucket, object=source_object, filename=filename) self.gdrive_hook.upload_file(local_location=filename, remote_location=destination_object) if self.move_object: self.gcs_hook.delete(self.source_bucket, source_object)
class GoogleDisplayVideo360SDFToBigQueryOperator(BaseOperator): """Make a request to SDF API and upload the data to BQ.""" DEFAULT_SDF_TABLE_NAMES = { 'LINE_ITEM': 'SDFLineItem', 'AD_GROUP': 'SDFAdGroup', 'AD': 'SDFAd', 'INSERTION_ORDER': 'SDFInsertionOrder', 'CAMPAIGN': 'SDFCampaign' } SDF_API_RESPONSE_KEYS = { 'LINE_ITEM': 'lineItems', 'AD_GROUP': 'adGroups', 'AD': 'ads', 'INSERTION_ORDER': 'insertionOrders', 'CAMPAIGN': 'campaigns' } def __init__(self, gcp_conn_id='google_cloud_default', gcs_bucket=None, schema=None, bq_dataset=None, write_disposition=None, cloud_project_id=None, file_types=None, filter_ids=None, api_version=None, filter_type=None, table_names=DEFAULT_SDF_TABLE_NAMES, sdf_api_response_keys=SDF_API_RESPONSE_KEYS, *args, **kwargs): super(GoogleDisplayVideo360SDFToBigQueryOperator, self).__init__(*args, **kwargs) self.gcp_conn_id = gcp_conn_id self.service = None self.hook = None self.bq_hook = None self.gcs_hook = None self.gcs_bucket = gcs_bucket self.schema = schema self.bq_dataset = bq_dataset self.write_disposition = write_disposition self.cloud_project_id = cloud_project_id self.file_types = file_types self.filter_ids = filter_ids self.api_version = api_version self.filter_type = filter_type self.table_names = table_names self.sdf_api_response_keys = sdf_api_response_keys def execute(self, context): if self.hook is None: self.hook = GoogleDisplayVideo360Hook(gcp_conn_id=self.gcp_conn_id) if self.bq_hook is None: self.bq_hook = BigQueryHook(bigquery_conn_id=self.gcp_conn_id) if self.gcs_hook is None: self.gcs_hook = GoogleCloudStorageHook( google_cloud_storage_conn_id=self.gcp_conn_id) request_body = { 'fileTypes': self.file_types, 'filterType': self.filter_type, 'filterIds': self.filter_ids, 'version': self.api_version } logger.info('Request body: %s ' % request_body) request = self.hook.get_service().sdf().download(body=request_body) response = request.execute() for file_type in self.file_types: temp_file = None try: logger.info('Uploading SDF to GCS') temp_file = tempfile.NamedTemporaryFile(delete=False) response_key = self.sdf_api_response_keys.get(file_type) temp_file.write(response[response_key].encode('utf-8')) temp_file.close() filename = '%d_%s_%s_%s.json' % (time.time() * 1e+9, randint(1, 1000000), response_key, 'sdf') self.gcs_hook.upload(self.gcs_bucket, filename, temp_file.name) logger.info('SDF upload to GCS complete') finally: if temp_file: temp_file.close() os.unlink(temp_file.name) sdf_file = 'gs://%s/%s' % (self.gcs_bucket, filename) bq_table = self.table_names.get(file_type) bq_table = '%s.%s' % (self.bq_dataset, bq_table) schema = SDF_VERSIONED_SCHEMA_TYPES.get( self.api_version).get(file_type) try: bq_base_cursor = self.bq_hook.get_conn().cursor() logger.info('Uploading SDF to BigQuery') bq_base_cursor.run_load( destination_project_dataset_table=bq_table, schema_fields=schema, source_uris=[sdf_file], source_format='CSV', skip_leading_rows=1, write_disposition=self.write_disposition) finally: logger.info('Deleting SDF from GCS') self.gcs_hook.delete(self.gcs_bucket, filename)