apache-airflow-providers-amazon 8.24.0rc2__tar.gz → 8.25.0rc1__tar.gz

{apache_airflow_providers_amazon-8.24.0rc2 → apache_airflow_providers_amazon-8.25.0rc1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: apache-airflow-providers-amazon
-Version: 8.24.0rc2
+Version: 8.25.0rc1
 Summary: Provider package apache-airflow-providers-amazon for Apache Airflow
 Keywords: airflow-provider,amazon,airflow,integration
 Author-email: Apache Software Foundation <dev@airflow.apache.org>
@@ -46,15 +46,15 @@ Requires-Dist: apache-airflow-providers-imap ; extra == "imap"
 Requires-Dist: apache-airflow-providers-microsoft-azure ; extra == "microsoft.azure"
 Requires-Dist: apache-airflow-providers-mongo ; extra == "mongo"
 Requires-Dist: apache-airflow-providers-openlineage ; extra == "openlineage"
-Requires-Dist: pandas>=1.5.3,<2.2 ; extra == "pandas" and (python_version<"3.12")
-Requires-Dist: pandas>=2.1.1,<2.2 ; extra == "pandas" and (python_version>="3.12")
+Requires-Dist: pandas>=2.1.2,<2.2 ; extra == "pandas" and (python_version>="3.9")
+Requires-Dist: pandas>=1.5.3,<2.2 ; extra == "pandas" and (python_version<"3.9")
 Requires-Dist: python3-saml>=1.16.0 ; extra == "python3-saml"
 Requires-Dist: s3fs>=2023.10.0 ; extra == "s3fs"
 Requires-Dist: apache-airflow-providers-salesforce ; extra == "salesforce"
 Requires-Dist: apache-airflow-providers-ssh ; extra == "ssh"
 Project-URL: Bug Tracker, https://github.com/apache/airflow/issues
-Project-URL: Changelog, https://airflow.apache.org/docs/apache-airflow-providers-amazon/8.24.0/changelog.html
-Project-URL: Documentation, https://airflow.apache.org/docs/apache-airflow-providers-amazon/8.24.0
+Project-URL: Changelog, https://airflow.apache.org/docs/apache-airflow-providers-amazon/8.25.0/changelog.html
+Project-URL: Documentation, https://airflow.apache.org/docs/apache-airflow-providers-amazon/8.25.0
 Project-URL: Slack Chat, https://s.apache.org/airflow-slack
 Project-URL: Source Code, https://github.com/apache/airflow
 Project-URL: Twitter, https://twitter.com/ApacheAirflow
@@ -121,7 +121,7 @@ Provides-Extra: ssh
 
 Package ``apache-airflow-providers-amazon``
 
-Release: ``8.24.0.rc2``
+Release: ``8.25.0.rc1``
 
 
 Amazon integration (including `Amazon Web Services (AWS) <https://aws.amazon.com/>`__).
@@ -134,7 +134,7 @@ This is a provider package for ``amazon`` provider. All classes for this provide
 are in ``airflow.providers.amazon`` python package.
 
 You can find package information and changelog for the provider
-in the `documentation <https://airflow.apache.org/docs/apache-airflow-providers-amazon/8.24.0/>`_.
+in the `documentation <https://airflow.apache.org/docs/apache-airflow-providers-amazon/8.25.0/>`_.
 
 Installation
 ------------
@@ -198,4 +198,4 @@ Dependent package
 ======================================================================================================================  ===================
 
 The changelog for the provider package can be found in the
-`changelog <https://airflow.apache.org/docs/apache-airflow-providers-amazon/8.24.0/changelog.html>`_.
+`changelog <https://airflow.apache.org/docs/apache-airflow-providers-amazon/8.25.0/changelog.html>`_.

{apache_airflow_providers_amazon-8.24.0rc2 → apache_airflow_providers_amazon-8.25.0rc1}/README.rst

@@ -42,7 +42,7 @@
 
 Package ``apache-airflow-providers-amazon``
 
-Release: ``8.24.0.rc2``
+Release: ``8.25.0.rc1``
 
 
 Amazon integration (including `Amazon Web Services (AWS) <https://aws.amazon.com/>`__).
@@ -55,7 +55,7 @@ This is a provider package for ``amazon`` provider. All classes for this provide
 are in ``airflow.providers.amazon`` python package.
 
 You can find package information and changelog for the provider
-in the `documentation <https://airflow.apache.org/docs/apache-airflow-providers-amazon/8.24.0/>`_.
+in the `documentation <https://airflow.apache.org/docs/apache-airflow-providers-amazon/8.25.0/>`_.
 
 Installation
 ------------
@@ -119,4 +119,4 @@ Dependent package
 ======================================================================================================================  ===================
 
 The changelog for the provider package can be found in the
-`changelog <https://airflow.apache.org/docs/apache-airflow-providers-amazon/8.24.0/changelog.html>`_.
+`changelog <https://airflow.apache.org/docs/apache-airflow-providers-amazon/8.25.0/changelog.html>`_.

{apache_airflow_providers_amazon-8.24.0rc2 → apache_airflow_providers_amazon-8.25.0rc1}/airflow/providers/amazon/__init__.py

@@ -29,7 +29,7 @@ from airflow import __version__ as airflow_version
 
 __all__ = ["__version__"]
 
-__version__ = "8.24.0"
+__version__ = "8.25.0"
 
 if packaging.version.parse(packaging.version.parse(airflow_version).base_version) < packaging.version.parse(
     "2.7.0"

{apache_airflow_providers_amazon-8.24.0rc2 → apache_airflow_providers_amazon-8.25.0rc1}/airflow/providers/amazon/aws/hooks/comprehend.py

@@ -16,6 +16,7 @@
 # under the License.
 from __future__ import annotations
 
+from airflow.exceptions import AirflowException
 from airflow.providers.amazon.aws.hooks.base_aws import AwsBaseHook
 
 
@@ -35,3 +36,35 @@ class ComprehendHook(AwsBaseHook):
     def __init__(self, *args, **kwargs) -> None:
         kwargs["client_type"] = "comprehend"
         super().__init__(*args, **kwargs)
+
+    def validate_document_classifier_training_status(
+        self, document_classifier_arn: str, fail_on_warnings: bool = False
+    ) -> None:
+        """
+        Log the Information about the document classifier.
+
+        NumberOfLabels
+        NumberOfTrainedDocuments
+        NumberOfTestDocuments
+        EvaluationMetrics
+
+        """
+        response = self.conn.describe_document_classifier(DocumentClassifierArn=document_classifier_arn)
+
+        status = response["DocumentClassifierProperties"]["Status"]
+
+        if status == "TRAINED_WITH_WARNING":
+            self.log.info(
+                "AWS Comprehend document classifier training completed with %s, Message: %s please review the skipped files folder in the output location %s",
+                status,
+                response["DocumentClassifierProperties"]["Message"],
+                response["DocumentClassifierProperties"]["OutputDataConfig"]["S3Uri"],
+            )
+
+            if fail_on_warnings:
+                raise AirflowException("Warnings in AWS Comprehend document classifier training.")
+
+        self.log.info(
+            "AWS Comprehend document classifier metadata: %s",
+            response["DocumentClassifierProperties"]["ClassifierMetadata"],
+        )

{apache_airflow_providers_amazon-8.24.0rc2 → apache_airflow_providers_amazon-8.25.0rc1}/airflow/providers/amazon/aws/hooks/redshift_sql.py

@@ -20,14 +20,19 @@ from functools import cached_property
 from typing import TYPE_CHECKING
 
 import redshift_connector
+from packaging.version import Version
 from redshift_connector import Connection as RedshiftConnection
 from sqlalchemy import create_engine
 from sqlalchemy.engine.url import URL
 
+from airflow import __version__ as AIRFLOW_VERSION
 from airflow.exceptions import AirflowException
 from airflow.providers.amazon.aws.hooks.base_aws import AwsBaseHook
 from airflow.providers.common.sql.hooks.sql import DbApiHook
 
+_IS_AIRFLOW_2_10_OR_HIGHER = Version(Version(AIRFLOW_VERSION).base_version) >= Version("2.10.0")
+
+
 if TYPE_CHECKING:
     from airflow.models.connection import Connection
     from airflow.providers.openlineage.sqlparser import DatabaseInfo
@@ -257,4 +262,6 @@ class RedshiftSQLHook(DbApiHook):
 
     def get_openlineage_default_schema(self) -> str | None:
         """Return current schema. This is usually changed with ``SEARCH_PATH`` parameter."""
-        return self.get_first("SELECT CURRENT_SCHEMA();")[0]
+        if _IS_AIRFLOW_2_10_OR_HIGHER:
+            return self.get_first("SELECT CURRENT_SCHEMA();")[0]
+        return super().get_openlineage_default_schema()

{apache_airflow_providers_amazon-8.24.0rc2 → apache_airflow_providers_amazon-8.25.0rc1}/airflow/providers/amazon/aws/operators/comprehend.py

@@ -23,7 +23,10 @@ from airflow.configuration import conf
 from airflow.exceptions import AirflowException
 from airflow.providers.amazon.aws.hooks.comprehend import ComprehendHook
 from airflow.providers.amazon.aws.operators.base_aws import AwsBaseOperator
-from airflow.providers.amazon.aws.triggers.comprehend import ComprehendPiiEntitiesDetectionJobCompletedTrigger
+from airflow.providers.amazon.aws.triggers.comprehend import (
+    ComprehendCreateDocumentClassifierCompletedTrigger,
+    ComprehendPiiEntitiesDetectionJobCompletedTrigger,
+)
 from airflow.providers.amazon.aws.utils import validate_execute_complete_event
 from airflow.providers.amazon.aws.utils.mixins import aws_template_fields
 from airflow.utils.timezone import utcnow
@@ -190,3 +193,147 @@ class ComprehendStartPiiEntitiesDetectionJobOperator(ComprehendBaseOperator):
 
         self.log.info("Comprehend pii entities detection job `%s` complete.", event["job_id"])
         return event["job_id"]
+
+
+class ComprehendCreateDocumentClassifierOperator(AwsBaseOperator[ComprehendHook]):
+    """
+    Create a comprehend document classifier that can categorize documents.
+
+    Provide a set of training documents that are labeled with the categories.
+
+    .. seealso::
+        For more information on how to use this operator, take a look at the guide:
+        :ref:`howto/operator:ComprehendCreateDocumentClassifierOperator`
+
+    :param document_classifier_name: The name of the document classifier. (templated)
+    :param input_data_config: Specifies the format and location of the input data for the job. (templated)
+    :param mode: Indicates the mode in which the classifier will be trained. (templated)
+    :param data_access_role_arn: The Amazon Resource Name (ARN) of the IAM role that grants Amazon Comprehend
+        read access to your input data. (templated)
+    :param language_code: The language of the input documents. You can specify any of the languages supported by
+        Amazon Comprehend. All documents must be in the same language. (templated)
+    :param fail_on_warnings: If set to True, the document classifier training job will throw an error when the
+        status is TRAINED_WITH_WARNING. (default False)
+    :param output_data_config: Specifies the location for the output files from a custom classifier job.
+        This parameter is required for a request that creates a native document model. (templated)
+    :param document_classifier_kwargs: Any optional parameters to pass to the document classifier. (templated)
+
+    :param wait_for_completion: Whether to wait for job to stop. (default: True)
+    :param waiter_delay: Time in seconds to wait between status checks. (default: 60)
+    :param waiter_max_attempts: Maximum number of attempts to check for job completion. (default: 20)
+    :param deferrable: If True, the operator will wait asynchronously for the job to stop.
+        This implies waiting for completion. This mode requires aiobotocore module to be installed.
+        (default: False)
+    :param aws_conn_id: The Airflow connection used for AWS credentials.
+        If this is ``None`` or empty then the default boto3 behaviour is used. If
+        running Airflow in a distributed manner and aws_conn_id is None or
+        empty, then default boto3 configuration would be used (and must be
+        maintained on each worker node).
+    :param region_name: AWS region_name. If not specified then the default boto3 behaviour is used.
+    :param verify: Whether to verify SSL certificates. See:
+        https://boto3.amazonaws.com/v1/documentation/api/latest/reference/core/session.html
+    :param botocore_config: Configuration dictionary (key-values) for botocore client. See:
+        https://botocore.amazonaws.com/v1/documentation/api/latest/reference/config.html
+    """
+
+    aws_hook_class = ComprehendHook
+
+    template_fields: Sequence[str] = aws_template_fields(
+        "document_classifier_name",
+        "input_data_config",
+        "mode",
+        "data_access_role_arn",
+        "language_code",
+        "output_data_config",
+        "document_classifier_kwargs",
+    )
+
+    template_fields_renderers: dict = {
+        "input_data_config": "json",
+        "output_data_config": "json",
+        "document_classifier_kwargs": "json",
+    }
+
+    def __init__(
+        self,
+        document_classifier_name: str,
+        input_data_config: dict[str, Any],
+        mode: str,
+        data_access_role_arn: str,
+        language_code: str,
+        fail_on_warnings: bool = False,
+        output_data_config: dict[str, Any] | None = None,
+        document_classifier_kwargs: dict[str, Any] | None = None,
+        wait_for_completion: bool = True,
+        waiter_delay: int = 60,
+        waiter_max_attempts: int = 20,
+        deferrable: bool = conf.getboolean("operators", "default_deferrable", fallback=False),
+        aws_conn_id: str | None = "aws_default",
+        **kwargs,
+    ):
+        super().__init__(**kwargs)
+        self.document_classifier_name = document_classifier_name
+        self.input_data_config = input_data_config
+        self.mode = mode
+        self.data_access_role_arn = data_access_role_arn
+        self.language_code = language_code
+        self.fail_on_warnings = fail_on_warnings
+        self.output_data_config = output_data_config
+        self.document_classifier_kwargs = document_classifier_kwargs or {}
+        self.wait_for_completion = wait_for_completion
+        self.waiter_delay = waiter_delay
+        self.waiter_max_attempts = waiter_max_attempts
+        self.deferrable = deferrable
+        self.aws_conn_id = aws_conn_id
+
+    def execute(self, context: Context) -> str:
+        if self.output_data_config:
+            self.document_classifier_kwargs["OutputDataConfig"] = self.output_data_config
+
+        document_classifier_arn = self.hook.conn.create_document_classifier(
+            DocumentClassifierName=self.document_classifier_name,
+            InputDataConfig=self.input_data_config,
+            Mode=self.mode,
+            DataAccessRoleArn=self.data_access_role_arn,
+            LanguageCode=self.language_code,
+            **self.document_classifier_kwargs,
+        )["DocumentClassifierArn"]
+
+        message_description = f"document classifier {document_classifier_arn} to complete."
+        if self.deferrable:
+            self.log.info("Deferring %s", message_description)
+            self.defer(
+                trigger=ComprehendCreateDocumentClassifierCompletedTrigger(
+                    document_classifier_arn=document_classifier_arn,
+                    waiter_delay=self.waiter_delay,
+                    waiter_max_attempts=self.waiter_max_attempts,
+                    aws_conn_id=self.aws_conn_id,
+                ),
+                method_name="execute_complete",
+            )
+        elif self.wait_for_completion:
+            self.log.info("Waiting for %s", message_description)
+
+            self.hook.get_waiter("create_document_classifier_complete").wait(
+                DocumentClassifierArn=document_classifier_arn,
+                WaiterConfig={"Delay": self.waiter_delay, "MaxAttempts": self.waiter_max_attempts},
+            )
+
+            self.hook.validate_document_classifier_training_status(
+                document_classifier_arn=document_classifier_arn, fail_on_warnings=self.fail_on_warnings
+            )
+
+        return document_classifier_arn
+
+    def execute_complete(self, context: Context, event: dict[str, Any] | None = None) -> str:
+        event = validate_execute_complete_event(event)
+        if event["status"] != "success":
+            raise AirflowException("Error while running comprehend create document classifier: %s", event)
+
+        self.hook.validate_document_classifier_training_status(
+            document_classifier_arn=event["document_classifier_arn"], fail_on_warnings=self.fail_on_warnings
+        )
+
+        self.log.info("Comprehend document classifier `%s` complete.", event["document_classifier_arn"])
+
+        return event["document_classifier_arn"]

{apache_airflow_providers_amazon-8.24.0rc2 → apache_airflow_providers_amazon-8.25.0rc1}/airflow/providers/amazon/aws/sensors/comprehend.py

@@ -23,7 +23,10 @@ from airflow.configuration import conf
 from airflow.exceptions import AirflowException, AirflowSkipException
 from airflow.providers.amazon.aws.hooks.comprehend import ComprehendHook
 from airflow.providers.amazon.aws.sensors.base_aws import AwsBaseSensor
-from airflow.providers.amazon.aws.triggers.comprehend import ComprehendPiiEntitiesDetectionJobCompletedTrigger
+from airflow.providers.amazon.aws.triggers.comprehend import (
+    ComprehendCreateDocumentClassifierCompletedTrigger,
+    ComprehendPiiEntitiesDetectionJobCompletedTrigger,
+)
 from airflow.providers.amazon.aws.utils.mixins import aws_template_fields
 
 if TYPE_CHECKING:
@@ -145,3 +148,111 @@ class ComprehendStartPiiEntitiesDetectionJobCompletedSensor(ComprehendBaseSensor
         return self.hook.conn.describe_pii_entities_detection_job(JobId=self.job_id)[
             "PiiEntitiesDetectionJobProperties"
         ]["JobStatus"]
+
+
+class ComprehendCreateDocumentClassifierCompletedSensor(AwsBaseSensor[ComprehendHook]):
+    """
+    Poll the state of the document classifier until it reaches a completed state; fails if the job fails.
+
+    .. seealso::
+        For more information on how to use this sensor, take a look at the guide:
+        :ref:`howto/sensor:ComprehendCreateDocumentClassifierCompletedSensor`
+
+    :param document_classifier_arn: The arn of the Comprehend document classifier.
+    :param fail_on_warnings: If set to True, the document classifier training job will throw an error when the
+        status is TRAINED_WITH_WARNING. (default False)
+
+    :param deferrable: If True, the sensor will operate in deferrable mode. This mode requires aiobotocore
+        module to be installed.
+        (default: False, but can be overridden in config file by setting default_deferrable to True)
+    :param poke_interval: Polling period in seconds to check for the status of the job. (default: 120)
+    :param max_retries: Number of times before returning the current state. (default: 75)
+    :param aws_conn_id: The Airflow connection used for AWS credentials.
+        If this is ``None`` or empty then the default boto3 behaviour is used. If
+        running Airflow in a distributed manner and aws_conn_id is None or
+        empty, then default boto3 configuration would be used (and must be
+        maintained on each worker node).
+    :param region_name: AWS region_name. If not specified then the default boto3 behaviour is used.
+    :param verify: Whether to verify SSL certificates. See:
+        https://boto3.amazonaws.com/v1/documentation/api/latest/reference/core/session.html
+    :param botocore_config: Configuration dictionary (key-values) for botocore client. See:
+        https://botocore.amazonaws.com/v1/documentation/api/latest/reference/config.html
+    """
+
+    aws_hook_class = ComprehendHook
+
+    INTERMEDIATE_STATES: tuple[str, ...] = (
+        "SUBMITTED",
+        "TRAINING",
+    )
+    FAILURE_STATES: tuple[str, ...] = (
+        "DELETING",
+        "STOP_REQUESTED",
+        "STOPPED",
+        "IN_ERROR",
+    )
+    SUCCESS_STATES: tuple[str, ...] = ("TRAINED", "TRAINED_WITH_WARNING")
+    FAILURE_MESSAGE = "Comprehend document classifier failed."
+
+    template_fields: Sequence[str] = aws_template_fields("document_classifier_arn")
+
+    def __init__(
+        self,
+        *,
+        document_classifier_arn: str,
+        fail_on_warnings: bool = False,
+        max_retries: int = 75,
+        poke_interval: int = 120,
+        deferrable: bool = conf.getboolean("operators", "default_deferrable", fallback=False),
+        aws_conn_id: str | None = "aws_default",
+        **kwargs: Any,
+    ) -> None:
+        super().__init__(**kwargs)
+        self.document_classifier_arn = document_classifier_arn
+        self.fail_on_warnings = fail_on_warnings
+        self.max_retries = max_retries
+        self.poke_interval = poke_interval
+        self.deferrable = deferrable
+        self.aws_conn_id = aws_conn_id
+
+    def execute(self, context: Context) -> Any:
+        if self.deferrable:
+            self.defer(
+                trigger=ComprehendCreateDocumentClassifierCompletedTrigger(
+                    document_classifier_arn=self.document_classifier_arn,
+                    waiter_delay=int(self.poke_interval),
+                    waiter_max_attempts=self.max_retries,
+                    aws_conn_id=self.aws_conn_id,
+                ),
+                method_name="poke",
+            )
+        else:
+            super().execute(context=context)
+
+    def poke(self, context: Context, **kwargs) -> bool:
+        status = self.hook.conn.describe_document_classifier(
+            DocumentClassifierArn=self.document_classifier_arn
+        )["DocumentClassifierProperties"]["Status"]
+
+        self.log.info(
+            "Poking for AWS Comprehend document classifier arn: %s status: %s",
+            self.document_classifier_arn,
+            status,
+        )
+
+        if status in self.FAILURE_STATES:
+            # TODO: remove this if block when min_airflow_version is set to higher than 2.7.1
+            if self.soft_fail:
+                raise AirflowSkipException(self.FAILURE_MESSAGE)
+            raise AirflowException(self.FAILURE_MESSAGE)
+
+        if status in self.SUCCESS_STATES:
+            self.hook.validate_document_classifier_training_status(
+                document_classifier_arn=self.document_classifier_arn, fail_on_warnings=self.fail_on_warnings
+            )
+
+            self.log.info("Comprehend document classifier `%s` complete.", self.document_classifier_arn)
+
+            return True
+
+        return False

{apache_airflow_providers_amazon-8.24.0rc2 → apache_airflow_providers_amazon-8.25.0rc1}/airflow/providers/amazon/aws/transfers/redshift_to_s3.py

@@ -128,7 +128,7 @@ class RedshiftToS3Operator(BaseOperator):
         self, credentials_block: str, select_query: str, s3_key: str, unload_options: str
     ) -> str:
         # Un-escape already escaped queries
-        select_query = re.sub(r"''(.+)''", r"'\1'", select_query)
+        select_query = re.sub(r"''(.+?)''", r"'\1'", select_query)
         return f"""
                     UNLOAD ($${select_query}$$)
                     TO 's3://{self.s3_bucket}/{s3_key}'

{apache_airflow_providers_amazon-8.24.0rc2 → apache_airflow_providers_amazon-8.25.0rc1}/airflow/providers/amazon/aws/triggers/comprehend.py

@@ -59,3 +59,39 @@ class ComprehendPiiEntitiesDetectionJobCompletedTrigger(AwsBaseWaiterTrigger):
 
     def hook(self) -> AwsGenericHook:
         return ComprehendHook(aws_conn_id=self.aws_conn_id)
+
+
+class ComprehendCreateDocumentClassifierCompletedTrigger(AwsBaseWaiterTrigger):
+    """
+    Trigger when a Comprehend document classifier is complete.
+
+    :param document_classifier_arn: The arn of the Comprehend document classifier.
+    :param waiter_delay: The amount of time in seconds to wait between attempts. (default: 120)
+    :param waiter_max_attempts: The maximum number of attempts to be made. (default: 75)
+    :param aws_conn_id: The Airflow connection used for AWS credentials.
+    """
+
+    def __init__(
+        self,
+        *,
+        document_classifier_arn: str,
+        waiter_delay: int = 120,
+        waiter_max_attempts: int = 75,
+        aws_conn_id: str | None = "aws_default",
+    ) -> None:
+        super().__init__(
+            serialized_fields={"document_classifier_arn": document_classifier_arn},
+            waiter_name="create_document_classifier_complete",
+            waiter_args={"DocumentClassifierArn": document_classifier_arn},
+            failure_message="Comprehend create document classifier failed.",
+            status_message="Status of Comprehend create document classifier is",
+            status_queries=["DocumentClassifierProperties.Status"],
+            return_key="document_classifier_arn",
+            return_value=document_classifier_arn,
+            waiter_delay=waiter_delay,
+            waiter_max_attempts=waiter_max_attempts,
+            aws_conn_id=aws_conn_id,
+        )
+
+    def hook(self) -> AwsGenericHook:
+        return ComprehendHook(aws_conn_id=self.aws_conn_id)

{apache_airflow_providers_amazon-8.24.0rc2 → apache_airflow_providers_amazon-8.25.0rc1}/airflow/providers/amazon/aws/utils/__init__.py

@@ -20,10 +20,9 @@ import logging
 import re
 from datetime import datetime, timezone
 from enum import Enum
+from importlib import metadata
 from typing import Any
 
-import importlib_metadata
-
 from airflow.exceptions import AirflowException
 from airflow.utils.helpers import prune_dict
 from airflow.version import version
@@ -78,7 +77,7 @@ def get_airflow_version() -> tuple[int, ...]:
 
 def get_botocore_version() -> tuple[int, ...]:
     """Return the version number of the installed botocore package in the form of a tuple[int,...]."""
-    return tuple(map(int, importlib_metadata.version("botocore").split(".")[:3]))
+    return tuple(map(int, metadata.version("botocore").split(".")[:3]))
 
 
 def validate_execute_complete_event(event: dict[str, Any] | None = None) -> dict[str, Any]:

apache_airflow_providers_amazon-8.25.0rc1/airflow/providers/amazon/aws/waiters/comprehend.json

@@ -0,0 +1,104 @@
+{
+    "version": 2,
+    "waiters": {
+        "pii_entities_detection_job_complete": {
+            "delay": 120,
+            "maxAttempts": 75,
+            "operation": "DescribePiiEntitiesDetectionJob",
+            "acceptors": [
+                {
+                    "matcher": "path",
+                    "argument": "PiiEntitiesDetectionJobProperties.JobStatus",
+                    "expected": "SUBMITTED",
+                    "state": "retry"
+                },
+                {
+                    "matcher": "path",
+                    "argument": "PiiEntitiesDetectionJobProperties.JobStatus",
+                    "expected": "IN_PROGRESS",
+                    "state": "retry"
+                },
+                {
+                    "matcher": "path",
+                    "argument": "PiiEntitiesDetectionJobProperties.JobStatus",
+                    "expected": "COMPLETED",
+                    "state": "success"
+                },
+                {
+                    "matcher": "path",
+                    "argument": "PiiEntitiesDetectionJobProperties.JobStatus",
+                    "expected": "FAILED",
+                    "state": "failure"
+                },
+                {
+                    "matcher": "path",
+                    "argument": "PiiEntitiesDetectionJobProperties.JobStatus",
+                    "expected": "STOP_REQUESTED",
+                    "state": "failure"
+                },
+                {
+                    "matcher": "path",
+                    "argument": "PiiEntitiesDetectionJobProperties.JobStatus",
+                    "expected": "STOPPED",
+                    "state": "failure"
+                }
+
+            ]
+        },
+         "create_document_classifier_complete": {
+            "delay": 120,
+            "maxAttempts": 75,
+            "operation": "DescribeDocumentClassifier",
+            "acceptors": [
+                {
+                    "matcher": "path",
+                    "argument": "DocumentClassifierProperties.Status",
+                    "expected": "SUBMITTED",
+                    "state": "retry"
+                },
+                {
+                    "matcher": "path",
+                    "argument": "DocumentClassifierProperties.Status",
+                    "expected": "TRAINING",
+                    "state": "retry"
+                },
+                {
+                    "matcher": "path",
+                    "argument": "DocumentClassifierProperties.Status",
+                    "expected": "DELETING",
+                    "state": "failure"
+                },
+                {
+                    "matcher": "path",
+                    "argument": "DocumentClassifierProperties.Status",
+                    "expected": "STOP_REQUESTED",
+                    "state": "failure"
+                },
+                {
+                    "matcher": "path",
+                    "argument": "DocumentClassifierProperties.Status",
+                    "expected": "STOPPED",
+                    "state": "failure"
+                },
+                {
+                    "matcher": "path",
+                    "argument": "DocumentClassifierProperties.Status",
+                    "expected": "IN_ERROR",
+                    "state": "failure"
+                },
+                {
+                    "matcher": "path",
+                    "argument": "DocumentClassifierProperties.Status",
+                    "expected": "TRAINED",
+                    "state": "success"
+                },
+                {
+                    "matcher": "path",
+                    "argument": "DocumentClassifierProperties.Status",
+                    "expected": "TRAINED_WITH_WARNING",
+                    "state": "success"
+                }
+            ]
+        }
+    }
+}

{apache_airflow_providers_amazon-8.24.0rc2 → apache_airflow_providers_amazon-8.25.0rc1}/airflow/providers/amazon/get_provider_info.py

@@ -28,8 +28,9 @@ def get_provider_info():
         "name": "Amazon",
         "description": "Amazon integration (including `Amazon Web Services (AWS) <https://aws.amazon.com/>`__).\n",
         "state": "ready",
-        "source-date-epoch": 1717051501,
+        "source-date-epoch": 1718602112,
         "versions": [
+            "8.25.0",
             "8.24.0",
             "8.23.0",
             "8.22.0",
@@ -109,8 +110,8 @@ def get_provider_info():
             {
                 "name": "pandas",
                 "dependencies": [
-                    'pandas>=1.5.3,<2.2;python_version<"3.12"',
-                    'pandas>=2.1.1,<2.2;python_version>="3.12"',
+                    'pandas>=2.1.2,<2.2;python_version>="3.9"',
+                    'pandas>=1.5.3,<2.2;python_version<"3.9"',
                 ],
             },
             {"name": "aiobotocore", "dependencies": ["aiobotocore[boto3]>=2.13.0"]},