apache-airflow-providers-google 10.12.0rc1__py3-none-any.whl → 10.13.0__py3-none-any.whl

airflow/providers/google/cloud/operators/gcs.py

@@ -188,11 +188,11 @@ class GCSListObjectsOperator(GoogleCloudBaseOperator):
         folder in ``data`` bucket. ::
 
             GCS_Files = GCSListOperator(
-                task_id='GCS_Files',
-                bucket='data',
-                prefix='sales/sales-2017/',
-                match_glob='**/*/.avro',
-                gcp_conn_id=google_cloud_conn_id
+                task_id="GCS_Files",
+                bucket="data",
+                prefix="sales/sales-2017/",
+                match_glob="**/*/.avro",
+                gcp_conn_id=google_cloud_conn_id,
             )
     """
 
@@ -313,6 +313,7 @@ class GCSDeleteObjectsOperator(GoogleCloudBaseOperator):
             )
             raise ValueError(err_message)
 
+        self._objects: list[str] = []
         super().__init__(**kwargs)
 
     def execute(self, context: Context) -> None:
@@ -322,13 +323,47 @@ class GCSDeleteObjectsOperator(GoogleCloudBaseOperator):
         )
 
         if self.objects is not None:
-            objects = self.objects
+            self._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:
+            self._objects = hook.list(bucket_name=self.bucket_name, prefix=self.prefix)
+        self.log.info("Deleting %s objects from %s", len(self._objects), self.bucket_name)
+        for object_name in self._objects:
             hook.delete(bucket_name=self.bucket_name, object_name=object_name)
 
+    def get_openlineage_facets_on_complete(self, task_instance):
+        """Implementing on_complete as execute() resolves object names."""
+        from openlineage.client.facet import (
+            LifecycleStateChange,
+            LifecycleStateChangeDatasetFacet,
+            LifecycleStateChangeDatasetFacetPreviousIdentifier,
+        )
+        from openlineage.client.run import Dataset
+
+        from airflow.providers.openlineage.extractors import OperatorLineage
+
+        if not self._objects:
+            return OperatorLineage()
+
+        bucket_url = f"gs://{self.bucket_name}"
+        input_datasets = [
+            Dataset(
+                namespace=bucket_url,
+                name=object_name,
+                facets={
+                    "lifecycleStateChange": LifecycleStateChangeDatasetFacet(
+                        lifecycleStateChange=LifecycleStateChange.DROP.value,
+                        previousIdentifier=LifecycleStateChangeDatasetFacetPreviousIdentifier(
+                            namespace=bucket_url,
+                            name=object_name,
+                        ),
+                    )
+                },
+            )
+            for object_name in self._objects
+        ]
+
+        return OperatorLineage(inputs=input_datasets)
+
 
 class GCSBucketCreateAclEntryOperator(GoogleCloudBaseOperator):
     """
@@ -596,6 +631,22 @@ class GCSFileTransformOperator(GoogleCloudBaseOperator):
                 filename=destination_file.name,
             )
 
+    def get_openlineage_facets_on_start(self):
+        from openlineage.client.run import Dataset
+
+        from airflow.providers.openlineage.extractors import OperatorLineage
+
+        input_dataset = Dataset(
+            namespace=f"gs://{self.source_bucket}",
+            name=self.source_object,
+        )
+        output_dataset = Dataset(
+            namespace=f"gs://{self.destination_bucket}",
+            name=self.destination_object,
+        )
+
+        return OperatorLineage(inputs=[input_dataset], outputs=[output_dataset])
+
 
 class GCSTimeSpanFileTransformOperator(GoogleCloudBaseOperator):
     """
@@ -722,6 +773,9 @@ class GCSTimeSpanFileTransformOperator(GoogleCloudBaseOperator):
         self.upload_continue_on_fail = upload_continue_on_fail
         self.upload_num_attempts = upload_num_attempts
 
+        self._source_object_names: list[str] = []
+        self._destination_object_names: list[str] = []
+
     def execute(self, context: Context) -> list[str]:
         # Define intervals and prefixes.
         try:
@@ -773,7 +827,7 @@ class GCSTimeSpanFileTransformOperator(GoogleCloudBaseOperator):
         )
 
         # Fetch list of files.
-        blobs_to_transform = source_hook.list_by_timespan(
+        self._source_object_names = source_hook.list_by_timespan(
             bucket_name=self.source_bucket,
             prefix=source_prefix_interp,
             timespan_start=timespan_start,
@@ -785,7 +839,7 @@ class GCSTimeSpanFileTransformOperator(GoogleCloudBaseOperator):
             temp_output_dir_path = Path(temp_output_dir)
 
             # TODO: download in parallel.
-            for blob_to_transform in blobs_to_transform:
+            for blob_to_transform in self._source_object_names:
                 destination_file = temp_input_dir_path / blob_to_transform
                 destination_file.parent.mkdir(parents=True, exist_ok=True)
                 try:
@@ -822,8 +876,6 @@ class GCSTimeSpanFileTransformOperator(GoogleCloudBaseOperator):
 
             self.log.info("Transformation succeeded. Output temporarily located at %s", temp_output_dir_path)
 
-            files_uploaded = []
-
             # TODO: upload in parallel.
             for upload_file in temp_output_dir_path.glob("**/*"):
                 if upload_file.is_dir():
@@ -844,12 +896,35 @@ class GCSTimeSpanFileTransformOperator(GoogleCloudBaseOperator):
                         chunk_size=self.chunk_size,
                         num_max_attempts=self.upload_num_attempts,
                     )
-                    files_uploaded.append(str(upload_file_name))
+                    self._destination_object_names.append(str(upload_file_name))
                 except GoogleCloudError:
                     if not self.upload_continue_on_fail:
                         raise
 
-            return files_uploaded
+            return self._destination_object_names
+
+    def get_openlineage_facets_on_complete(self, task_instance):
+        """Implementing on_complete as execute() resolves object names."""
+        from openlineage.client.run import Dataset
+
+        from airflow.providers.openlineage.extractors import OperatorLineage
+
+        input_datasets = [
+            Dataset(
+                namespace=f"gs://{self.source_bucket}",
+                name=object_name,
+            )
+            for object_name in self._source_object_names
+        ]
+        output_datasets = [
+            Dataset(
+                namespace=f"gs://{self.destination_bucket}",
+                name=object_name,
+            )
+            for object_name in self._destination_object_names
+        ]
+
+        return OperatorLineage(inputs=input_datasets, outputs=output_datasets)
 
 
 class GCSDeleteBucketOperator(GoogleCloudBaseOperator):

airflow/providers/google/cloud/operators/kubernetes_engine.py

@@ -192,15 +192,14 @@ class GKECreateClusterOperator(GoogleCloudBaseOperator):
     The **minimum** required to define a cluster to create is:
 
     ``dict()`` ::
-        cluster_def = {'name': 'my-cluster-name',
-                       'initial_node_count': 1}
+        cluster_def = {"name": "my-cluster-name", "initial_node_count": 1}
 
     or
 
     ``Cluster`` proto ::
         from google.cloud.container_v1.types import Cluster
 
-        cluster_def = Cluster(name='my-cluster-name', initial_node_count=1)
+        cluster_def = Cluster(name="my-cluster-name", initial_node_count=1)
 
     **Operator Creation**: ::
 

airflow/providers/google/cloud/operators/pubsub.py

@@ -58,24 +58,22 @@ class PubSubCreateTopicOperator(GoogleCloudBaseOperator):
     By default, if the topic already exists, this operator will
     not cause the DAG to fail. ::
 
-        with DAG('successful DAG') as dag:
-            (
-                PubSubCreateTopicOperator(project_id='my-project', topic='my_new_topic')
-                >> PubSubCreateTopicOperator(project_id='my-project', topic='my_new_topic')
-            )
+        with DAG("successful DAG") as dag:
+            create_topic = PubSubCreateTopicOperator(project_id="my-project", topic="my_new_topic")
+            create_topic_again = PubSubCreateTopicOperator(project_id="my-project", topic="my_new_topic")
+
+            create_topic >> create_topic_again
 
     The operator can be configured to fail if the topic already exists. ::
 
-        with DAG('failing DAG') as dag:
-            (
-                PubSubCreateTopicOperator(project_id='my-project', topic='my_new_topic')
-                >> PubSubCreateTopicOperator(
-                    project_id='my-project',
-                    topic='my_new_topic',
-                    fail_if_exists=True,
-                )
+        with DAG("failing DAG") as dag:
+            create_topic = PubSubCreateTopicOperator(project_id="my-project", topic="my_new_topic")
+            create_topic_again = PubSubCreateTopicOperator(
+                project_id="my-project", topic="my_new_topic", fail_if_exists=True
             )
 
+            create_topic >> create_topic_again
+
     Both ``project_id`` and ``topic`` are templated so you can use Jinja templating in their values.
 
     :param project_id: Optional, the Google Cloud project ID where the topic will be created.
@@ -197,43 +195,35 @@ class PubSubCreateSubscriptionOperator(GoogleCloudBaseOperator):
     By default, if the subscription already exists, this operator will
     not cause the DAG to fail. However, the topic must exist in the project. ::
 
-        with DAG('successful DAG') as dag:
-            (
-                PubSubCreateSubscriptionOperator(
-                    project_id='my-project',
-                    topic='my-topic',
-                    subscription='my-subscription'
-                )
-                >> PubSubCreateSubscriptionOperator(
-                    project_id='my-project',
-                    topic='my-topic',
-                    subscription='my-subscription',
-                )
+        with DAG("successful DAG") as dag:
+            create_subscription = PubSubCreateSubscriptionOperator(
+                project_id="my-project", topic="my-topic", subscription="my-subscription"
+            )
+            create_subscription_again = PubSubCreateSubscriptionOperator(
+                project_id="my-project", topic="my-topic", subscription="my-subscription"
             )
 
+            create_subscription >> create_subscription_again
+
+
     The operator can be configured to fail if the subscription already exists.
     ::
 
-        with DAG('failing DAG') as dag:
-            (
-                PubSubCreateSubscriptionOperator(
-                    project_id='my-project',
-                    topic='my-topic',
-                    subscription='my-subscription',
-                )
-                >> PubSubCreateSubscriptionOperator(
-                    project_id='my-project',
-                    topic='my-topic',
-                    subscription='my-subscription',
-                    fail_if_exists=True,
-                )
+        with DAG("failing DAG") as dag:
+            create_subscription = PubSubCreateSubscriptionOperator(
+                project_id="my-project", topic="my-topic", subscription="my-subscription"
             )
+            create_subscription_again = PubSubCreateSubscriptionOperator(
+                project_id="my-project", topic="my-topic", subscription="my-subscription", fail_if_exists=True
+            )
+
+            create_subscription >> create_subscription_again
 
     Finally, subscription is not required. If not passed, the operator will
     generated a universally unique identifier for the subscription's name. ::
 
-        with DAG('DAG') as dag:
-            PubSubCreateSubscriptionOperator(project_id='my-project', topic='my-topic')
+        with DAG("DAG") as dag:
+            PubSubCreateSubscriptionOperator(project_id="my-project", topic="my-topic")
 
     ``project_id``, ``topic``, ``subscription``, ``subscription_project_id`` and
     ``impersonation_chain`` are templated so you can use Jinja templating in their values.
@@ -410,14 +400,16 @@ class PubSubDeleteTopicOperator(GoogleCloudBaseOperator):
     By default, if the topic does not exist, this operator will
     not cause the DAG to fail. ::
 
-        with DAG('successful DAG') as dag:
-            PubSubDeleteTopicOperator(project_id='my-project', topic='non_existing_topic')
+        with DAG("successful DAG") as dag:
+            PubSubDeleteTopicOperator(project_id="my-project", topic="non_existing_topic")
 
     The operator can be configured to fail if the topic does not exist. ::
 
-        with DAG('failing DAG') as dag:
+        with DAG("failing DAG") as dag:
             PubSubDeleteTopicOperator(
-                project_id='my-project', topic='non_existing_topic', fail_if_not_exists=True,
+                project_id="my-project",
+                topic="non_existing_topic",
+                fail_if_not_exists=True,
             )
 
     Both ``project_id`` and ``topic`` are templated so you can use Jinja templating in their values.
@@ -506,16 +498,18 @@ class PubSubDeleteSubscriptionOperator(GoogleCloudBaseOperator):
     By default, if the subscription does not exist, this operator will
     not cause the DAG to fail. ::
 
-        with DAG('successful DAG') as dag:
-            PubSubDeleteSubscriptionOperator(project_id='my-project', subscription='non-existing')
+        with DAG("successful DAG") as dag:
+            PubSubDeleteSubscriptionOperator(project_id="my-project", subscription="non-existing")
 
     The operator can be configured to fail if the subscription already exists.
 
     ::
 
-        with DAG('failing DAG') as dag:
+        with DAG("failing DAG") as dag:
             PubSubDeleteSubscriptionOperator(
-                project_id='my-project', subscription='non-existing', fail_if_not_exists=True,
+                project_id="my-project",
+                subscription="non-existing",
+                fail_if_not_exists=True,
             )
 
     ``project_id``, and ``subscription`` are templated so you can use Jinja templating in their values.
@@ -605,15 +599,13 @@ class PubSubPublishMessageOperator(GoogleCloudBaseOperator):
     in a single Google Cloud project. If the topic does not exist, this
     task will fail. ::
 
-        m1 = {'data': b'Hello, World!',
-              'attributes': {'type': 'greeting'}
-             }
-        m2 = {'data': b'Knock, knock'}
-        m3 = {'attributes': {'foo': ''}}
+        m1 = {"data": b"Hello, World!", "attributes": {"type": "greeting"}}
+        m2 = {"data": b"Knock, knock"}
+        m3 = {"attributes": {"foo": ""}}
 
         t1 = PubSubPublishMessageOperator(
-            project_id='my-project',
-            topic='my_topic',
+            project_id="my-project",
+            topic="my_topic",
             messages=[m1, m2, m3],
             create_topic=True,
             dag=dag,

airflow/providers/google/cloud/secrets/secret_manager.py

@@ -20,12 +20,16 @@ from __future__ import annotations
 import logging
 import re
 import warnings
+from typing import Sequence
 
 from google.auth.exceptions import DefaultCredentialsError
 
 from airflow.exceptions import AirflowException, AirflowProviderDeprecationWarning
 from airflow.providers.google.cloud._internal_client.secret_manager_client import _SecretManagerClient
-from airflow.providers.google.cloud.utils.credentials_provider import get_credentials_and_project_id
+from airflow.providers.google.cloud.utils.credentials_provider import (
+    _get_target_principal_and_delegates,
+    get_credentials_and_project_id,
+)
 from airflow.secrets import BaseSecretsBackend
 from airflow.utils.log.logging_mixin import LoggingMixin
 
@@ -76,6 +80,14 @@ class CloudSecretManagerBackend(BaseSecretsBackend, LoggingMixin):
     :param project_id: Project ID to read the secrets from. If not passed, the project ID from credentials
         will be used.
     :param sep: Separator used to concatenate connections_prefix and conn_id. Default: "-"
+    :param impersonation_chain: Optional 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.
     """
 
     def __init__(
@@ -89,6 +101,7 @@ class CloudSecretManagerBackend(BaseSecretsBackend, LoggingMixin):
         gcp_scopes: str | None = None,
         project_id: str | None = None,
         sep: str = "-",
+        impersonation_chain: str | Sequence[str] | None = None,
         **kwargs,
     ) -> None:
         super().__init__(**kwargs)
@@ -103,11 +116,19 @@ class CloudSecretManagerBackend(BaseSecretsBackend, LoggingMixin):
                     f"follows that pattern {SECRET_ID_PATTERN}"
                 )
         try:
+            if impersonation_chain:
+                target_principal, delegates = _get_target_principal_and_delegates(impersonation_chain)
+            else:
+                target_principal = None
+                delegates = None
+
             self.credentials, self.project_id = get_credentials_and_project_id(
                 keyfile_dict=gcp_keyfile_dict,
                 key_path=gcp_key_path,
                 credential_config_file=gcp_credential_config_file,
                 scopes=gcp_scopes,
+                target_principal=target_principal,
+                delegates=delegates,
             )
         except (DefaultCredentialsError, FileNotFoundError):
             log.exception(

airflow/providers/google/cloud/sensors/cloud_composer.py

@@ -19,9 +19,10 @@
 
 from __future__ import annotations
 
+import warnings
 from typing import TYPE_CHECKING, Any, Sequence
 
-from airflow.exceptions import AirflowException, AirflowSkipException
+from airflow.exceptions import AirflowException, AirflowProviderDeprecationWarning, AirflowSkipException
 from airflow.providers.google.cloud.triggers.cloud_composer import CloudComposerExecutionTrigger
 from airflow.sensors.base import BaseSensorOperator
 
@@ -33,6 +34,11 @@ class CloudComposerEnvironmentSensor(BaseSensorOperator):
     """
     Check the status of the Cloud Composer Environment task.
 
+    This Sensor is deprecated. You can achieve the same functionality by using Cloud Composer Operators
+    CloudComposerCreateEnvironmentOperator, CloudComposerDeleteEnvironmentOperator and
+    CloudComposerUpdateEnvironmentOperator in  deferrable or non-deferrable mode, since every operator
+    gives user a possibility to wait (asynchronously or synchronously) until Operation will be finished.
+
     :param project_id: Required. The ID of the Google Cloud project that the service belongs to.
     :param region: Required. The ID of the Google Cloud region that the service belongs to.
     :param operation_name: The name of the operation resource
@@ -59,6 +65,13 @@ class CloudComposerEnvironmentSensor(BaseSensorOperator):
         pooling_period_seconds: int = 30,
         **kwargs,
     ):
+        warnings.warn(
+            f"The `{self.__class__.__name__}` operator is deprecated. You can achieve the same functionality "
+            f"by using operators in deferrable or non-deferrable mode, since every operator for Cloud "
+            f"Composer will wait for the operation to complete.",
+            AirflowProviderDeprecationWarning,
+            stacklevel=2,
+        )
         super().__init__(**kwargs)
         self.project_id = project_id
         self.region = region

airflow/providers/google/cloud/sensors/dataplex.py

@@ -259,3 +259,121 @@ class DataplexDataQualityJobStatusSensor(BaseSensorOperator):
                     raise AirflowSkipException(message)
                 raise AirflowDataQualityScanException(message)
         return job_status == DataScanJob.State.SUCCEEDED
+
+
+class DataplexDataProfileJobStatusSensor(BaseSensorOperator):
+    """
+    Check the status of the Dataplex DataProfile job.
+
+    :param project_id: Required. The ID of the Google Cloud project that the task belongs to.
+    :param region: Required. The ID of the Google Cloud region that the task belongs to.
+    :param data_scan_id: Required. Data Quality scan identifier.
+    :param job_id: Required. Job ID.
+    :param api_version: The version of the api that will be requested for example 'v3'.
+    :param retry: A retry object used  to retry requests. If `None` is specified, requests
+        will not be retried.
+    :param metadata: Additional metadata that is provided to the method.
+    :param gcp_conn_id: The connection ID to use when fetching connection info.
+    :param impersonation_chain: Optional 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).
+    :param result_timeout: Value in seconds for which operator will wait for the Data Quality scan result.
+        Throws exception if there is no result found after specified amount of seconds.
+
+    :return: Boolean indicating if the job run has reached the ``DataScanJob.State.SUCCEEDED``.
+    """
+
+    template_fields = ["job_id"]
+
+    def __init__(
+        self,
+        project_id: str,
+        region: str,
+        data_scan_id: str,
+        job_id: str,
+        api_version: str = "v1",
+        retry: Retry | _MethodDefault = DEFAULT,
+        metadata: Sequence[tuple[str, str]] = (),
+        gcp_conn_id: str = "google_cloud_default",
+        impersonation_chain: str | Sequence[str] | None = None,
+        result_timeout: float = 60.0 * 10,
+        start_sensor_time: float | None = None,
+        *args,
+        **kwargs,
+    ) -> None:
+        super().__init__(*args, **kwargs)
+        self.project_id = project_id
+        self.region = region
+        self.data_scan_id = data_scan_id
+        self.job_id = job_id
+        self.api_version = api_version
+        self.retry = retry
+        self.metadata = metadata
+        self.gcp_conn_id = gcp_conn_id
+        self.impersonation_chain = impersonation_chain
+        self.result_timeout = result_timeout
+        self.start_sensor_time = start_sensor_time
+
+    def _duration(self):
+        if not self.start_sensor_time:
+            self.start_sensor_time = time.monotonic()
+        return time.monotonic() - self.start_sensor_time
+
+    def poke(self, context: Context) -> bool:
+        self.log.info("Waiting for job %s to be %s", self.job_id, DataScanJob.State.SUCCEEDED)
+        if self.result_timeout:
+            duration = self._duration()
+            if duration > self.result_timeout:
+                # TODO: remove this if check when min_airflow_version is set to higher than 2.7.1
+                message = (
+                    f"Timeout: Data Profile scan {self.job_id} is not ready after {self.result_timeout}s"
+                )
+                if self.soft_fail:
+                    raise AirflowSkipException(message)
+                raise AirflowDataQualityScanResultTimeoutException(message)
+
+        hook = DataplexHook(
+            gcp_conn_id=self.gcp_conn_id,
+            api_version=self.api_version,
+            impersonation_chain=self.impersonation_chain,
+        )
+
+        try:
+            job = hook.get_data_scan_job(
+                project_id=self.project_id,
+                region=self.region,
+                data_scan_id=self.data_scan_id,
+                job_id=self.job_id,
+                timeout=self.timeout,
+                retry=self.retry,
+                metadata=self.metadata,
+            )
+        except GoogleAPICallError as e:
+            # TODO: remove this if check when min_airflow_version is set to higher than 2.7.1
+            message = f"Error occurred when trying to retrieve Data Profile scan job: {self.data_scan_id}"
+            if self.soft_fail:
+                raise AirflowSkipException(message, e)
+            raise AirflowException(message, e)
+
+        job_status = job.state
+        self.log.info(
+            "Current status of the Dataplex Data Profile scan job %s => %s", self.job_id, job_status
+        )
+        if job_status == DataScanJob.State.FAILED:
+            # TODO: remove this if check when min_airflow_version is set to higher than 2.7.1
+            message = f"Data Profile scan job failed: {self.job_id}"
+            if self.soft_fail:
+                raise AirflowSkipException(message)
+            raise AirflowException(message)
+        if job_status == DataScanJob.State.CANCELLED:
+            # TODO: remove this if check when min_airflow_version is set to higher than 2.7.1
+            message = f"Data Profile scan job cancelled: {self.job_id}"
+            if self.soft_fail:
+                raise AirflowSkipException(message)
+            raise AirflowException(message)
+        return job_status == DataScanJob.State.SUCCEEDED

airflow/providers/google/cloud/sensors/gcs.py

@@ -50,6 +50,7 @@ class GCSObjectExistenceSensor(BaseSensorOperator):
     :param bucket: The Google Cloud Storage bucket where the object is.
     :param object: The name of the object to check in the Google cloud
         storage bucket.
+    :param use_glob: When set to True the object parameter is interpreted as glob
     :param google_cloud_conn_id: The connection ID to use when
         connecting to Google Cloud Storage.
     :param impersonation_chain: Optional service account to impersonate using short-term
@@ -75,6 +76,7 @@ class GCSObjectExistenceSensor(BaseSensorOperator):
         *,
         bucket: str,
         object: str,
+        use_glob: bool = False,
         google_cloud_conn_id: str = "google_cloud_default",
         impersonation_chain: str | Sequence[str] | None = None,
         retry: Retry = DEFAULT_RETRY,
@@ -84,7 +86,9 @@ class GCSObjectExistenceSensor(BaseSensorOperator):
         super().__init__(**kwargs)
         self.bucket = bucket
         self.object = object
+        self.use_glob = use_glob
         self.google_cloud_conn_id = google_cloud_conn_id
+        self._matches: list[str] = []
         self.impersonation_chain = impersonation_chain
         self.retry = retry
 
@@ -96,7 +100,11 @@ class GCSObjectExistenceSensor(BaseSensorOperator):
             gcp_conn_id=self.google_cloud_conn_id,
             impersonation_chain=self.impersonation_chain,
         )
-        return hook.exists(self.bucket, self.object, self.retry)
+        if self.use_glob:
+            self._matches = hook.list(self.bucket, match_glob=self.object)
+            return bool(self._matches)
+        else:
+            return hook.exists(self.bucket, self.object, self.retry)
 
     def execute(self, context: Context) -> None:
         """Airflow runs this method on the worker and defers using the trigger."""
@@ -109,6 +117,7 @@ class GCSObjectExistenceSensor(BaseSensorOperator):
                     trigger=GCSBlobTrigger(
                         bucket=self.bucket,
                         object_name=self.object,
+                        use_glob=self.use_glob,
                         poke_interval=self.poke_interval,
                         google_cloud_conn_id=self.google_cloud_conn_id,
                         hook_params={

airflow/providers/google/cloud/transfers/adls_to_gcs.py

@@ -58,12 +58,12 @@ class ADLSToGCSOperator(ADLSListOperator):
         resulting gcs path will be ``gs://mybucket/hello/world.avro`` ::
 
             copy_single_file = AdlsToGoogleCloudStorageOperator(
-                task_id='copy_single_file',
-                src_adls='hello/world.avro',
-                dest_gcs='gs://mybucket',
+                task_id="copy_single_file",
+                src_adls="hello/world.avro",
+                dest_gcs="gs://mybucket",
                 replace=False,
-                azure_data_lake_conn_id='azure_data_lake_default',
-                gcp_conn_id='google_cloud_default'
+                azure_data_lake_conn_id="azure_data_lake_default",
+                gcp_conn_id="google_cloud_default",
             )
 
         The following Operator would copy all parquet files from ADLS