feat(parser): add support for Pydantic v2

.github/workflows/quality_check_pydanticv2.yml

@@ -0,0 +1,76 @@
+name: Code quality - Pydanticv2
+
+# PROCESS
+#
+# 1. Install all dependencies and spin off containers for all supported Python versions
+# 2. Run code formatters and linters (various checks) for code standard
+# 3. Run static typing checker for potential bugs
+# 4. Run entire test suite for regressions except end-to-end (unit, functional, performance)
+# 5. Run static analysis (in addition to CodeQL) for common insecure code practices
+# 6. Run complexity baseline to avoid error-prone bugs and keep maintenance lower
+# 7. Collect and report on test coverage
+
+# USAGE
+#
+# Always triggered on new PRs, PR changes and PR merge.
+
+
+on:
+  pull_request:
+    paths:
+      - "aws_lambda_powertools/**"
+      - "tests/**"
+      - "pyproject.toml"
+      - "poetry.lock"
+      - "mypy.ini"
+    branches:
+      - develop
+  push:
+    paths:
+      - "aws_lambda_powertools/**"
+      - "tests/**"
+      - "pyproject.toml"
+      - "poetry.lock"
+      - "mypy.ini"
+    branches:
+      - develop
+
+permissions:
+  contents: read
+
+jobs:
+  quality_check:
+    runs-on: ubuntu-latest
+    strategy:
+      max-parallel: 4
+      matrix:
+        python-version: ["3.7", "3.8", "3.9", "3.10"]
+    env:
+      PYTHON: "${{ matrix.python-version }}"
+    permissions:
+      contents: read  # checkout code only
+    steps:
+      - uses: actions/checkout@c85c95e3d7251135ab7dc9ce3241c5835cc595a9  # v3.5.3
+      - name: Install poetry
+        run: pipx install poetry
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@bd6b4b6205c4dbad673328db7b31b7fab9e241c0 # v4.6.1
+        with:
+          python-version: ${{ matrix.python-version }}
+          cache: "poetry"
+      - name: Removing dev dependencies locked to Pydantic v1
+        run: poetry remove cfn-lint
+      - name: Replacing Pydantic v1 with v2 > 2.0.3
+        run: poetry add "pydantic=^2.0.3"
+      - name: Install dependencies
+        run: make dev
+      - name: Formatting and Linting
+        run: make lint
+      - name: Static type checking
+        run: make mypy
+      - name: Test with pytest
+        run: make test
+      - name: Security baseline
+        run: make security-baseline
+      - name: Complexity baseline
+        run: make complexity-baseline

aws_lambda_powertools/utilities/batch/base.py

@@ -348,6 +348,11 @@ def _to_batch_type(self, record: dict, event_type: EventType) -> EventSourceData
 
     def _to_batch_type(self, record: dict, event_type: EventType, model: Optional["BatchTypeModels"] = None):
         if model is not None:
+            # If a model is provided, we assume Pydantic is installed and we need to disable v2 warnings
+            from aws_lambda_powertools.utilities.parser.compat import disable_pydantic_v2_warning
+
+            disable_pydantic_v2_warning()
+
             return model.parse_obj(record)
         return self._DATA_CLASS_MAPPING[event_type](record)
 
@@ -500,8 +505,13 @@ def _process_record(self, record: dict) -> Union[SuccessResponse, FailureRespons
             # we need to handle that exception differently.
             # We check for a public attr in validation errors coming from Pydantic exceptions (subclass or not)
             # and we compare if it's coming from the same model that trigger the exception in the first place
-            model = getattr(exc, "model", None)
-            if model == self.model:
+
+            # Pydantic v1 raises a ValidationError with ErrorWrappers and store the model instance in a class variable.
+            # Pydantic v2 simplifies this by adding a title variable to store the model name directly.
+            model = getattr(exc, "model", None) or getattr(exc, "title", None)
+            model_name = getattr(self.model, "__name__", None)
+
+            if model == self.model or model == model_name:
                 return self._register_model_validation_error_record(record)
 
             return self.failure_handler(record=data, exception=sys.exc_info())
@@ -644,8 +654,13 @@ async def _async_process_record(self, record: dict) -> Union[SuccessResponse, Fa
             # we need to handle that exception differently.
             # We check for a public attr in validation errors coming from Pydantic exceptions (subclass or not)
             # and we compare if it's coming from the same model that trigger the exception in the first place
-            model = getattr(exc, "model", None)
-            if model == self.model:
+
+            # Pydantic v1 raises a ValidationError with ErrorWrappers and store the model instance in a class variable.
+            # Pydantic v2 simplifies this by adding a title variable to store the model name directly.
+            model = getattr(exc, "model", None) or getattr(exc, "title", None)
+            model_name = getattr(self.model, "__name__", None)
+
+            if model == self.model or model == model_name:
                 return self._register_model_validation_error_record(record)
 
             return self.failure_handler(record=data, exception=sys.exc_info())

aws_lambda_powertools/utilities/parser/compat.py

@@ -0,0 +1,34 @@
+import functools
+
+
+@functools.lru_cache(maxsize=None)
+def disable_pydantic_v2_warning():
+    """
+    Disables the Pydantic version 2 warning by filtering out the related warnings.
+
+    This function checks the version of Pydantic currently installed and if it is version 2,
+    it filters out the PydanticDeprecationWarning and PydanticDeprecatedSince20 warnings
+    to suppress them.
+
+    Since we only need to run the code once, we are using lru_cache to improve performance.
+
+    Note: This function assumes that Pydantic is installed.
+
+    Usage:
+        disable_pydantic_v2_warning()
+    """
+    try:
+        from pydantic import __version__
+
+        version = __version__.split(".")
+
+        if int(version[0]) == 2:
+            import warnings
+
+            from pydantic import PydanticDeprecatedSince20, PydanticDeprecationWarning
+
+            warnings.filterwarnings("ignore", category=PydanticDeprecationWarning)
+            warnings.filterwarnings("ignore", category=PydanticDeprecatedSince20)
+
+    except ImportError:
+        pass

aws_lambda_powertools/utilities/parser/envelopes/base.py

@@ -2,6 +2,7 @@
 from abc import ABC, abstractmethod
 from typing import Any, Dict, Optional, Type, TypeVar, Union
 
+from aws_lambda_powertools.utilities.parser.compat import disable_pydantic_v2_warning
 from aws_lambda_powertools.utilities.parser.types import Model
 
 logger = logging.getLogger(__name__)
@@ -26,6 +27,8 @@ def _parse(data: Optional[Union[Dict[str, Any], Any]], model: Type[Model]) -> Un
         Any
             Parsed data
         """
+        disable_pydantic_v2_warning()
+
         if data is None:
             logger.debug("Skipping parsing as event is None")
             return data

aws_lambda_powertools/utilities/parser/models/__init__.py

@@ -1,3 +1,7 @@
+from aws_lambda_powertools.utilities.parser.compat import disable_pydantic_v2_warning
+
+disable_pydantic_v2_warning()
+
 from .alb import AlbModel, AlbRequestContext, AlbRequestContextData
 from .apigw import (
     APIGatewayEventAuthorizer,

aws_lambda_powertools/utilities/parser/models/apigw.py

@@ -21,74 +21,74 @@ class ApiGatewayUserCert(BaseModel):
 
 
 class APIGatewayEventIdentity(BaseModel):
-    accessKey: Optional[str]
-    accountId: Optional[str]
-    apiKey: Optional[str]
-    apiKeyId: Optional[str]
-    caller: Optional[str]
-    cognitoAuthenticationProvider: Optional[str]
-    cognitoAuthenticationType: Optional[str]
-    cognitoIdentityId: Optional[str]
-    cognitoIdentityPoolId: Optional[str]
-    principalOrgId: Optional[str]
+    accessKey: Optional[str] = None
+    accountId: Optional[str] = None
+    apiKey: Optional[str] = None
+    apiKeyId: Optional[str] = None
+    caller: Optional[str] = None
+    cognitoAuthenticationProvider: Optional[str] = None
+    cognitoAuthenticationType: Optional[str] = None
+    cognitoIdentityId: Optional[str] = None
+    cognitoIdentityPoolId: Optional[str] = None
+    principalOrgId: Optional[str] = None
     # see #1562, temp workaround until API Gateway fixes it the Test button payload
     # removing it will not be considered a regression in the future
     sourceIp: Union[IPvAnyNetwork, Literal["test-invoke-source-ip"]]
-    user: Optional[str]
-    userAgent: Optional[str]
-    userArn: Optional[str]
-    clientCert: Optional[ApiGatewayUserCert]
+    user: Optional[str] = None
+    userAgent: Optional[str] = None
+    userArn: Optional[str] = None
+    clientCert: Optional[ApiGatewayUserCert] = None
 
 
 class APIGatewayEventAuthorizer(BaseModel):
-    claims: Optional[Dict[str, Any]]
-    scopes: Optional[List[str]]
+    claims: Optional[Dict[str, Any]] = None
+    scopes: Optional[List[str]] = None
 
 
 class APIGatewayEventRequestContext(BaseModel):
     accountId: str
     apiId: str
-    authorizer: Optional[APIGatewayEventAuthorizer]
+    authorizer: Optional[APIGatewayEventAuthorizer] = None
     stage: str
     protocol: str
     identity: APIGatewayEventIdentity
     requestId: str
     requestTime: str
     requestTimeEpoch: datetime
-    resourceId: Optional[str]
+    resourceId: Optional[str] = None
     resourcePath: str
-    domainName: Optional[str]
-    domainPrefix: Optional[str]
-    extendedRequestId: Optional[str]
+    domainName: Optional[str] = None
+    domainPrefix: Optional[str] = None
+    extendedRequestId: Optional[str] = None
     httpMethod: Literal["DELETE", "GET", "HEAD", "OPTIONS", "PATCH", "POST", "PUT"]
     path: str
-    connectedAt: Optional[datetime]
-    connectionId: Optional[str]
-    eventType: Optional[Literal["CONNECT", "MESSAGE", "DISCONNECT"]]
-    messageDirection: Optional[str]
-    messageId: Optional[str]
-    routeKey: Optional[str]
-    operationName: Optional[str]
+    connectedAt: Optional[datetime] = None
+    connectionId: Optional[str] = None
+    eventType: Optional[Literal["CONNECT", "MESSAGE", "DISCONNECT"]] = None
+    messageDirection: Optional[str] = None
+    messageId: Optional[str] = None
+    routeKey: Optional[str] = None
+    operationName: Optional[str] = None
 
-    @root_validator(allow_reuse=True)
+    @root_validator(allow_reuse=True, skip_on_failure=True)
     def check_message_id(cls, values):
         message_id, event_type = values.get("messageId"), values.get("eventType")
         if message_id is not None and event_type != "MESSAGE":
-            raise TypeError("messageId is available only when the `eventType` is `MESSAGE`")
+            raise ValueError("messageId is available only when the `eventType` is `MESSAGE`")
         return values
 
 
 class APIGatewayProxyEventModel(BaseModel):
-    version: Optional[str]
+    version: Optional[str] = None
     resource: str
     path: str
     httpMethod: Literal["DELETE", "GET", "HEAD", "OPTIONS", "PATCH", "POST", "PUT"]
     headers: Dict[str, str]
     multiValueHeaders: Dict[str, List[str]]
-    queryStringParameters: Optional[Dict[str, str]]
-    multiValueQueryStringParameters: Optional[Dict[str, List[str]]]
+    queryStringParameters: Optional[Dict[str, str]] = None
+    multiValueQueryStringParameters: Optional[Dict[str, List[str]]] = None
     requestContext: APIGatewayEventRequestContext
-    pathParameters: Optional[Dict[str, str]]
-    stageVariables: Optional[Dict[str, str]]
+    pathParameters: Optional[Dict[str, str]] = None
+    stageVariables: Optional[Dict[str, str]] = None
     isBase64Encoded: bool
-    body: Optional[Union[str, Type[BaseModel]]]
+    body: Optional[Union[str, Type[BaseModel]]] = None

aws_lambda_powertools/utilities/parser/models/apigwv2.py

@@ -14,13 +14,13 @@ class RequestContextV2AuthorizerIamCognito(BaseModel):
 
 
 class RequestContextV2AuthorizerIam(BaseModel):
-    accessKey: Optional[str]
-    accountId: Optional[str]
-    callerId: Optional[str]
-    principalOrgId: Optional[str]
-    userArn: Optional[str]
-    userId: Optional[str]
-    cognitoIdentity: Optional[RequestContextV2AuthorizerIamCognito]
+    accessKey: Optional[str] = None
+    accountId: Optional[str] = None
+    callerId: Optional[str] = None
+    principalOrgId: Optional[str] = None
+    userArn: Optional[str] = None
+    userId: Optional[str] = None
+    cognitoIdentity: Optional[RequestContextV2AuthorizerIamCognito] = None
 
 
 class RequestContextV2AuthorizerJwt(BaseModel):
@@ -29,8 +29,8 @@ class RequestContextV2AuthorizerJwt(BaseModel):
 
 
 class RequestContextV2Authorizer(BaseModel):
-    jwt: Optional[RequestContextV2AuthorizerJwt]
-    iam: Optional[RequestContextV2AuthorizerIam]
+    jwt: Optional[RequestContextV2AuthorizerJwt] = None
+    iam: Optional[RequestContextV2AuthorizerIam] = None
     lambda_value: Optional[Dict[str, Any]] = Field(None, alias="lambda")
 
 
@@ -45,7 +45,7 @@ class RequestContextV2Http(BaseModel):
 class RequestContextV2(BaseModel):
     accountId: str
     apiId: str
-    authorizer: Optional[RequestContextV2Authorizer]
+    authorizer: Optional[RequestContextV2Authorizer] = None
     domainName: str
     domainPrefix: str
     requestId: str
@@ -61,11 +61,11 @@ class APIGatewayProxyEventV2Model(BaseModel):
     routeKey: str
     rawPath: str
     rawQueryString: str
-    cookies: Optional[List[str]]
+    cookies: Optional[List[str]] = None
     headers: Dict[str, str]
-    queryStringParameters: Optional[Dict[str, str]]
-    pathParameters: Optional[Dict[str, str]]
-    stageVariables: Optional[Dict[str, str]]
+    queryStringParameters: Optional[Dict[str, str]] = None
+    pathParameters: Optional[Dict[str, str]] = None
+    stageVariables: Optional[Dict[str, str]] = None
     requestContext: RequestContextV2
-    body: Optional[Union[str, Type[BaseModel]]]
+    body: Optional[Union[str, Type[BaseModel]]] = None
     isBase64Encoded: bool

aws_lambda_powertools/utilities/parser/models/dynamodb.py

@@ -7,10 +7,10 @@
 
 
 class DynamoDBStreamChangedRecordModel(BaseModel):
-    ApproximateCreationDateTime: Optional[date]
+    ApproximateCreationDateTime: Optional[date] = None
     Keys: Dict[str, Dict[str, Any]]
-    NewImage: Optional[Union[Dict[str, Any], Type[BaseModel], BaseModel]]
-    OldImage: Optional[Union[Dict[str, Any], Type[BaseModel], BaseModel]]
+    NewImage: Optional[Union[Dict[str, Any], Type[BaseModel], BaseModel]] = None
+    OldImage: Optional[Union[Dict[str, Any], Type[BaseModel], BaseModel]] = None
     SequenceNumber: str
     SizeBytes: int
     StreamViewType: Literal["NEW_AND_OLD_IMAGES", "KEYS_ONLY", "NEW_IMAGE", "OLD_IMAGE"]
@@ -40,7 +40,7 @@ class DynamoDBStreamRecordModel(BaseModel):
     awsRegion: str
     eventSourceARN: str
     dynamodb: DynamoDBStreamChangedRecordModel
-    userIdentity: Optional[UserIdentity]
+    userIdentity: Optional[UserIdentity] = None
 
 
 class DynamoDBStreamModel(BaseModel):

aws_lambda_powertools/utilities/parser/models/kafka.py

@@ -19,8 +19,8 @@ class KafkaRecordModel(BaseModel):
     value: Union[str, Type[BaseModel]]
     headers: List[Dict[str, bytes]]
 
-    # validators
-    _decode_key = validator("key", allow_reuse=True)(base64_decode)
+    # Added type ignore to keep compatibility between Pydantic v1 and v2
+    _decode_key = validator("key", allow_reuse=True)(base64_decode)  # type: ignore[type-var, unused-ignore]
 
     @validator("value", pre=True, allow_reuse=True)
     def data_base64_decode(cls, value):

aws_lambda_powertools/utilities/parser/models/kinesis_firehose.py

@@ -17,7 +17,7 @@ class KinesisFirehoseRecord(BaseModel):
     data: Union[bytes, Type[BaseModel]]  # base64 encoded str is parsed into bytes
     recordId: str
     approximateArrivalTimestamp: PositiveInt
-    kinesisRecordMetadata: Optional[KinesisFirehoseRecordMetadata]
+    kinesisRecordMetadata: Optional[KinesisFirehoseRecordMetadata] = None
 
     @validator("data", pre=True, allow_reuse=True)
     def data_base64_decode(cls, value):
@@ -28,5 +28,5 @@ class KinesisFirehoseModel(BaseModel):
     invocationId: str
     deliveryStreamArn: str
     region: str
-    sourceKinesisStreamArn: Optional[str]
+    sourceKinesisStreamArn: Optional[str] = None
     records: List[KinesisFirehoseRecord]