airbyte-cdk 7.1.0.post4.dev17948197380__tar.gz → 7.2.0__tar.gz

{airbyte_cdk-7.1.0.post4.dev17948197380 → airbyte_cdk-7.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: airbyte-cdk
-Version: 7.1.0.post4.dev17948197380
+Version: 7.2.0
 Summary: A framework for writing Airbyte Connectors.
 Home-page: https://airbyte.com
 License: MIT

{airbyte_cdk-7.1.0.post4.dev17948197380 → airbyte_cdk-7.2.0}/airbyte_cdk/sources/declarative/auth/jwt.py

@@ -6,15 +6,26 @@ import base64
 import json
 from dataclasses import InitVar, dataclass
 from datetime import datetime
-from typing import Any, Mapping, Optional, Union
+from typing import Any, Mapping, Optional, Union, cast
 
 import jwt
+from cryptography.hazmat.primitives import serialization
+from cryptography.hazmat.primitives.asymmetric.ec import EllipticCurvePrivateKey
+from cryptography.hazmat.primitives.asymmetric.ed448 import Ed448PrivateKey
+from cryptography.hazmat.primitives.asymmetric.ed25519 import Ed25519PrivateKey
+from cryptography.hazmat.primitives.asymmetric.rsa import RSAPrivateKey
+from cryptography.hazmat.primitives.asymmetric.types import PrivateKeyTypes
 
 from airbyte_cdk.sources.declarative.auth.declarative_authenticator import DeclarativeAuthenticator
 from airbyte_cdk.sources.declarative.interpolation.interpolated_boolean import InterpolatedBoolean
 from airbyte_cdk.sources.declarative.interpolation.interpolated_mapping import InterpolatedMapping
 from airbyte_cdk.sources.declarative.interpolation.interpolated_string import InterpolatedString
 
+# Type alias for keys that JWT library accepts
+JwtKeyTypes = Union[
+    RSAPrivateKey, EllipticCurvePrivateKey, Ed25519PrivateKey, Ed448PrivateKey, str, bytes
+]
+
 
 class JwtAlgorithm(str):
     """
@@ -74,6 +85,7 @@ class JwtAuthenticator(DeclarativeAuthenticator):
     aud: Optional[Union[InterpolatedString, str]] = None
     additional_jwt_headers: Optional[Mapping[str, Any]] = None
     additional_jwt_payload: Optional[Mapping[str, Any]] = None
+    passphrase: Optional[Union[InterpolatedString, str]] = None
 
     def __post_init__(self, parameters: Mapping[str, Any]) -> None:
         self._secret_key = InterpolatedString.create(self.secret_key, parameters=parameters)
@@ -103,6 +115,11 @@ class JwtAuthenticator(DeclarativeAuthenticator):
         self._additional_jwt_payload = InterpolatedMapping(
             self.additional_jwt_payload or {}, parameters=parameters
         )
+        self._passphrase = (
+            InterpolatedString.create(self.passphrase, parameters=parameters)
+            if self.passphrase
+            else None
+        )
 
     def _get_jwt_headers(self) -> dict[str, Any]:
         """
@@ -149,11 +166,21 @@ class JwtAuthenticator(DeclarativeAuthenticator):
         payload["nbf"] = nbf
         return payload
 
-    def _get_secret_key(self) -> str:
+    def _get_secret_key(self) -> JwtKeyTypes:
         """
         Returns the secret key used to sign the JWT.
         """
         secret_key: str = self._secret_key.eval(self.config, json_loads=json.loads)
+
+        if self._passphrase:
+            passphrase_value = self._passphrase.eval(self.config, json_loads=json.loads)
+            if passphrase_value:
+                private_key = serialization.load_pem_private_key(
+                    secret_key.encode(),
+                    password=passphrase_value.encode(),
+                )
+                return cast(JwtKeyTypes, private_key)
+
         return (
             base64.b64encode(secret_key.encode()).decode()
             if self._base64_encode_secret_key

{airbyte_cdk-7.1.0.post4.dev17948197380 → airbyte_cdk-7.2.0}/airbyte_cdk/sources/declarative/declarative_component_schema.yaml

@@ -1270,6 +1270,12 @@ definitions:
         title: Additional JWT Payload Properties
         description: Additional properties to be added to the JWT payload.
         additionalProperties: true
+      passphrase:
+        title: Passphrase
+        description: A passphrase/password used to encrypt the private key. Only provide a passphrase if required by the API for JWT authentication. The API will typically provide the passphrase when generating the public/private key pair.
+        type: string
+        examples:
+          - "{{ config['passphrase'] }}"
       $parameters:
         type: object
         additionalProperties: true

{airbyte_cdk-7.1.0.post4.dev17948197380 → airbyte_cdk-7.2.0}/airbyte_cdk/sources/declarative/interpolation/macros.py

@@ -6,6 +6,7 @@ import builtins
 import datetime
 import re
 import typing
+import uuid
 from typing import Optional, Union
 from urllib.parse import quote_plus
 
@@ -207,6 +208,16 @@ def camel_case_to_snake_case(value: str) -> str:
     return re.sub(r"(?<!^)(?=[A-Z])", "_", value).lower()
 
 
+def generate_uuid() -> str:
+    """
+    Generates a UUID4
+
+    Usage:
+    `"{{ generate_uuid() }}"`
+    """
+    return str(uuid.uuid4())
+
+
 _macros_list = [
     now_utc,
     today_utc,
@@ -220,5 +231,6 @@ _macros_list = [
     str_to_datetime,
     sanitize_url,
     camel_case_to_snake_case,
+    generate_uuid,
 ]
 macros = {f.__name__: f for f in _macros_list}

{airbyte_cdk-7.1.0.post4.dev17948197380 → airbyte_cdk-7.2.0}/airbyte_cdk/sources/declarative/models/declarative_component_schema.py

@@ -448,6 +448,12 @@ class JwtAuthenticator(BaseModel):
         description="Additional properties to be added to the JWT payload.",
         title="Additional JWT Payload Properties",
     )
+    passphrase: Optional[str] = Field(
+        None,
+        description="A passphrase/password used to encrypt the private key. Only provide a passphrase if required by the API for JWT authentication. The API will typically provide the passphrase when generating the public/private key pair.",
+        examples=["{{ config['passphrase'] }}"],
+        title="Passphrase",
+    )
     parameters: Optional[Dict[str, Any]] = Field(None, alias="$parameters")
 
 

{airbyte_cdk-7.1.0.post4.dev17948197380 → airbyte_cdk-7.2.0}/airbyte_cdk/sources/declarative/parsers/model_to_component_factory.py

@@ -1264,22 +1264,12 @@ class ModelToComponentFactory:
         component_definition: ComponentDefinition,
         stream_name: str,
         stream_namespace: Optional[str],
+        stream_state: MutableMapping[str, Any],
         config: Config,
         message_repository: Optional[MessageRepository] = None,
         runtime_lookback_window: Optional[datetime.timedelta] = None,
-        stream_state_migrations: Optional[List[Any]] = None,
         **kwargs: Any,
     ) -> ConcurrentCursor:
-        # Per-partition incremental streams can dynamically create child cursors which will pass their current
-        # state via the stream_state keyword argument. Incremental syncs without parent streams use the
-        # incoming state and connector_state_manager that is initialized when the component factory is created
-        stream_state = (
-            self._connector_state_manager.get_stream_state(stream_name, stream_namespace)
-            if "stream_state" not in kwargs
-            else kwargs["stream_state"]
-        )
-        stream_state = self.apply_stream_state_migrations(stream_state_migrations, stream_state)
-
         component_type = component_definition.get("type")
         if component_definition.get("type") != model_type.__name__:
             raise ValueError(
@@ -1498,21 +1488,11 @@ class ModelToComponentFactory:
         component_definition: ComponentDefinition,
         stream_name: str,
         stream_namespace: Optional[str],
+        stream_state: MutableMapping[str, Any],
         config: Config,
         message_repository: Optional[MessageRepository] = None,
-        stream_state_migrations: Optional[List[Any]] = None,
         **kwargs: Any,
     ) -> ConcurrentCursor:
-        # Per-partition incremental streams can dynamically create child cursors which will pass their current
-        # state via the stream_state keyword argument. Incremental syncs without parent streams use the
-        # incoming state and connector_state_manager that is initialized when the component factory is created
-        stream_state = (
-            self._connector_state_manager.get_stream_state(stream_name, stream_namespace)
-            if "stream_state" not in kwargs
-            else kwargs["stream_state"]
-        )
-        stream_state = self.apply_stream_state_migrations(stream_state_migrations, stream_state)
-
         component_type = component_definition.get("type")
         if component_definition.get("type") != model_type.__name__:
             raise ValueError(
@@ -1587,7 +1567,6 @@ class ModelToComponentFactory:
         config: Config,
         stream_state: MutableMapping[str, Any],
         partition_router: PartitionRouter,
-        stream_state_migrations: Optional[List[Any]] = None,
         attempt_to_create_cursor_if_not_provided: bool = False,
         **kwargs: Any,
     ) -> ConcurrentPerPartitionCursor:
@@ -1647,11 +1626,9 @@ class ModelToComponentFactory:
                 stream_namespace=stream_namespace,
                 config=config,
                 message_repository=NoopMessageRepository(),
-                # stream_state_migrations=stream_state_migrations,  # FIXME is it expected to run migration on per partition state too?
             )
         )
 
-        stream_state = self.apply_stream_state_migrations(stream_state_migrations, stream_state)
         # Per-partition state doesn't make sense for GroupingPartitionRouter, so force the global state
         use_global_cursor = isinstance(
             partition_router, GroupingPartitionRouter
@@ -1974,6 +1951,7 @@ class ModelToComponentFactory:
         self, model: DeclarativeStreamModel, config: Config, is_parent: bool = False, **kwargs: Any
     ) -> AbstractStream:
         primary_key = model.primary_key.__root__ if model.primary_key else None
+        self._migrate_state(model, config)
 
         partition_router = self._build_stream_slicer_from_partition_router(
             model.retriever,
@@ -2135,6 +2113,23 @@ class ModelToComponentFactory:
             supports_file_transfer=hasattr(model, "file_uploader") and bool(model.file_uploader),
         )
 
+    def _migrate_state(self, model: DeclarativeStreamModel, config: Config) -> None:
+        stream_name = model.name or ""
+        stream_state = self._connector_state_manager.get_stream_state(
+            stream_name=stream_name, namespace=None
+        )
+        if model.state_migrations:
+            state_transformations = [
+                self._create_component_from_model(state_migration, config, declarative_stream=model)
+                for state_migration in model.state_migrations
+            ]
+        else:
+            state_transformations = []
+        stream_state = self.apply_stream_state_migrations(state_transformations, stream_state)
+        self._connector_state_manager.update_state_for_stream(
+            stream_name=stream_name, namespace=None, value=stream_state
+        )
+
     def _is_stop_condition_on_cursor(self, model: DeclarativeStreamModel) -> bool:
         return bool(
             model.incremental_sync
@@ -2206,17 +2201,7 @@ class ModelToComponentFactory:
         config: Config,
     ) -> Cursor:
         stream_name = model.name or ""
-        stream_state = self._connector_state_manager.get_stream_state(
-            stream_name=stream_name, namespace=None
-        )
-
-        if model.state_migrations:
-            state_transformations = [
-                self._create_component_from_model(state_migration, config, declarative_stream=model)
-                for state_migration in model.state_migrations
-            ]
-        else:
-            state_transformations = []
+        stream_state = self._connector_state_manager.get_stream_state(stream_name, None)
 
         if (
             model.incremental_sync
@@ -2228,10 +2213,9 @@ class ModelToComponentFactory:
                 model_type=DatetimeBasedCursorModel,
                 component_definition=model.incremental_sync.__dict__,
                 stream_name=stream_name,
+                stream_state=stream_state,
                 stream_namespace=None,
                 config=config or {},
-                stream_state=stream_state,
-                stream_state_migrations=state_transformations,
                 partition_router=stream_slicer,
                 attempt_to_create_cursor_if_not_provided=True,  # FIXME can we remove that now?
             )
@@ -2242,8 +2226,8 @@ class ModelToComponentFactory:
                     component_definition=model.incremental_sync.__dict__,
                     stream_name=stream_name,
                     stream_namespace=None,
+                    stream_state=stream_state,
                     config=config or {},
-                    stream_state_migrations=state_transformations,
                 )
             elif type(model.incremental_sync) == DatetimeBasedCursorModel:
                 return self.create_concurrent_cursor_from_datetime_based_cursor(  # type: ignore # This is a known issue that we are creating and returning a ConcurrentCursor which does not technically implement the (low-code) StreamSlicer. However, (low-code) StreamSlicer and ConcurrentCursor both implement StreamSlicer.stream_slices() which is the primary method needed for checkpointing
@@ -2251,8 +2235,8 @@ class ModelToComponentFactory:
                     component_definition=model.incremental_sync.__dict__,
                     stream_name=stream_name,
                     stream_namespace=None,
+                    stream_state=stream_state,
                     config=config or {},
-                    stream_state_migrations=state_transformations,
                     attempt_to_create_cursor_if_not_provided=True,
                 )
             else:
@@ -2721,6 +2705,7 @@ class ModelToComponentFactory:
             aud=jwt_payload.aud,
             additional_jwt_headers=model.additional_jwt_headers,
             additional_jwt_payload=model.additional_jwt_payload,
+            passphrase=model.passphrase,
         )
 
     def create_list_partition_router(

{airbyte_cdk-7.1.0.post4.dev17948197380 → airbyte_cdk-7.2.0}/airbyte_cdk/sources/declarative/resolvers/config_components_resolver.py

@@ -177,11 +177,7 @@ class ConfigComponentsResolver(ComponentsResolver):
                 )
 
                 path = [path.eval(self.config, **kwargs) for path in resolved_component.field_path]
-                # Avoid parsing strings that are meant to be strings
-                if not (isinstance(value, str) and valid_types == (str,)):
-                    parsed_value = self._parse_yaml_if_possible(value)
-                else:
-                    parsed_value = value
+                parsed_value = self._parse_yaml_if_possible(value)
                 updated = dpath.set(updated_config, path, parsed_value)
 
                 if parsed_value and not updated and resolved_component.create_or_update:

{airbyte_cdk-7.1.0.post4.dev17948197380 → airbyte_cdk-7.2.0}/pyproject.toml

@@ -23,7 +23,7 @@ classifiers = [
 keywords = ["airbyte", "connector-development-kit", "cdk"]
 
 # Python CDK uses dynamic versioning: https://github.com/mtkennerly/poetry-dynamic-versioning
-version = "7.1.0.post4.dev17948197380"  # Version will be calculated dynamically.
+version = "7.2.0"  # Version will be calculated dynamically.
 
 [tool.poetry-dynamic-versioning]
 enable = false