airbyte-cdk 6.56.5.post3.dev15889419925__tar.gz → 6.56.6.post2.dev15930606991__tar.gz

{airbyte_cdk-6.56.5.post3.dev15889419925 → airbyte_cdk-6.56.6.post2.dev15930606991}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: airbyte-cdk
-Version: 6.56.5.post3.dev15889419925
+Version: 6.56.6.post2.dev15930606991
 Summary: A framework for writing Airbyte Connectors.
 Home-page: https://airbyte.com
 License: MIT

{airbyte_cdk-6.56.5.post3.dev15889419925 → airbyte_cdk-6.56.6.post2.dev15930606991}/airbyte_cdk/sources/declarative/declarative_component_schema.yaml

@@ -3650,7 +3650,6 @@ definitions:
           - "$ref": "#/definitions/GzipDecoder"
           - "$ref": "#/definitions/JsonDecoder"
           - "$ref": "#/definitions/JsonlDecoder"
-          - "$ref": "#/definitions/XmlDecoder"
   CsvDecoder:
     title: CSV
     description: "Select 'CSV' for response data that is formatted as CSV (comma-separated values). Can specify an encoding (default: 'utf-8') and a delimiter (default: ',')."

{airbyte_cdk-6.56.5.post3.dev15889419925 → airbyte_cdk-6.56.6.post2.dev15930606991}/airbyte_cdk/sources/declarative/models/declarative_component_schema.py

@@ -1,5 +1,3 @@
-# Copyright (c) 2025 Airbyte, Inc., all rights reserved.
-
 # generated by datamodel-codegen:
 #   filename:  declarative_component_schema.yaml
 
@@ -2022,7 +2020,7 @@ class RecordSelector(BaseModel):
 
 class GzipDecoder(BaseModel):
     type: Literal["GzipDecoder"]
-    decoder: Union[CsvDecoder, GzipDecoder, JsonDecoder, JsonlDecoder, XmlDecoder]
+    decoder: Union[CsvDecoder, GzipDecoder, JsonDecoder, JsonlDecoder]
 
 
 class RequestBodyGraphQL(BaseModel):

{airbyte_cdk-6.56.5.post3.dev15889419925 → airbyte_cdk-6.56.6.post2.dev15930606991}/airbyte_cdk/sources/declarative/parsers/manifest_normalizer.py

@@ -108,8 +108,7 @@ class ManifestNormalizer:
             ManifestNormalizationException: Caught internally and handled by returning the original manifest.
         """
         try:
-            self._deduplicate_minifest()
-            self._reference_schemas()
+            self._deduplicate_manifest()
 
             return self._normalized_manifest
         except ManifestNormalizationException:
@@ -131,7 +130,7 @@ class ManifestNormalizer:
 
         yield from []
 
-    def _deduplicate_minifest(self) -> None:
+    def _deduplicate_manifest(self) -> None:
         """
         Find commonalities in the input JSON structure and refactor it to avoid redundancy.
         """
@@ -141,9 +140,117 @@ class ManifestNormalizer:
             self._prepare_definitions()
             # replace duplicates with references, if any
             self._handle_duplicates(self._collect_duplicates())
+            # replace parent streams with $refs
+            self._replace_parent_streams_with_refs()
+            # clean dangling fields after resolving $refs
+            self._clean_dangling_fields()
         except Exception as e:
             raise ManifestNormalizationException(str(e))
 
+    def _replace_parent_streams_with_refs(self) -> None:
+        """
+        For each stream in the manifest, if it has a retriever.partition_router with parent_stream_configs,
+        replace any 'stream' fields in those configs that are dicts and deeply equal to another stream object
+        with a $ref to the correct stream index.
+        """
+        streams = self._normalized_manifest.get(STREAMS_TAG, [])
+
+        # Build a hash-to-index mapping for O(1) lookups
+        stream_hash_to_index = {}
+        for idx, stream in enumerate(streams):
+            stream_hash = self._hash_object(stream)
+            stream_hash_to_index[stream_hash] = idx
+
+        for idx, stream in enumerate(streams):
+            retriever = stream.get("retriever")
+            if not retriever:
+                continue
+            partition_router = retriever.get("partition_router")
+            routers = (
+                partition_router
+                if isinstance(partition_router, list)
+                else [partition_router]
+                if partition_router
+                else []
+            )
+            for router in routers:
+                if not isinstance(router, dict):
+                    continue
+                if router.get("type") != "SubstreamPartitionRouter":
+                    continue
+                parent_stream_configs = router.get("parent_stream_configs", [])
+                for parent_config in parent_stream_configs:
+                    if not isinstance(parent_config, dict):
+                        continue
+                    stream_ref = parent_config.get("stream")
+                    # Only replace if it's a dict and matches any stream in the manifest
+                    if stream_ref is not None and isinstance(stream_ref, dict):
+                        stream_ref_hash = self._hash_object(stream_ref)
+                        if stream_ref_hash in stream_hash_to_index:
+                            parent_config["stream"] = {
+                                "$ref": f"#/streams/{stream_hash_to_index[stream_ref_hash]}"
+                            }
+
+    def _clean_dangling_fields(self) -> None:
+        """
+        Clean the manifest by removing unused definitions and schemas.
+        This method removes any definitions or schemas that are not referenced by any $ref in the manifest.
+        """
+
+        def find_all_refs(obj: Dict[str, Any], refs: List[str]) -> None:
+            """
+            Recursively find all $ref paths in the object.
+
+            Args:
+                obj: The object to search through
+                refs: List to store found reference paths
+            """
+            if not isinstance(obj, dict):
+                return
+
+            for key, value in obj.items():
+                if key == "$ref" and isinstance(value, str):
+                    # Remove the leading #/ from the ref path
+                    refs.append(value[2:])
+                elif isinstance(value, dict):
+                    find_all_refs(value, refs)
+                elif isinstance(value, list):
+                    for item in value:
+                        if isinstance(item, dict):
+                            find_all_refs(item, refs)
+
+        def clean_section(section: Dict[str, Any], section_path: str) -> None:
+            """
+            Clean a section by removing unreferenced fields.
+
+            Args:
+                section: The section to clean
+                section_path: The path to this section in the manifest
+            """
+            for key in list(section.keys()):
+                current_path = f"{section_path}/{key}"
+                # Check if this path is referenced or is a parent of a referenced path
+                if not any(ref.startswith(current_path) for ref in all_refs):
+                    del section[key]
+
+        # Find all references in the manifest
+        all_refs: List[str] = []
+        find_all_refs(self._normalized_manifest, all_refs)
+
+        # Clean definitions
+        if DEF_TAG in self._normalized_manifest:
+            clean_section(self._normalized_manifest[DEF_TAG], DEF_TAG)
+            # Remove empty definitions section
+            if not self._normalized_manifest[DEF_TAG]:
+                del self._normalized_manifest[DEF_TAG]
+
+        # Clean schemas
+        if SCHEMAS_TAG in self._normalized_manifest:
+            clean_section(self._normalized_manifest[SCHEMAS_TAG], SCHEMAS_TAG)
+            # Remove empty schemas section
+            if not self._normalized_manifest[SCHEMAS_TAG]:
+                del self._normalized_manifest[SCHEMAS_TAG]
+
     def _prepare_definitions(self) -> None:
         """
         Clean the definitions in the manifest by removing unnecessary properties.
@@ -163,43 +270,6 @@ class ManifestNormalizer:
             if key != LINKED_TAG:
                 self._normalized_manifest[DEF_TAG].pop(key, None)
 
-    def _extract_stream_schema(self, stream: Dict[str, Any]) -> None:
-        """
-        Extract the schema from the stream and add it to the `schemas` tag.
-        """
-
-        stream_name = stream["name"]
-        # copy the value of the SCHEMA_TAG to the SCHEMAS_TAG with the stream name as key
-        schema = stream.get(SCHEMA_LOADER_TAG, {}).get(SCHEMA_TAG)
-        if not SCHEMAS_TAG in self._normalized_manifest.keys():
-            self._normalized_manifest[SCHEMAS_TAG] = {}
-        # add stream schema to the SCHEMAS_TAG
-        if not stream_name in self._normalized_manifest[SCHEMAS_TAG].keys():
-            # add the schema to the SCHEMAS_TAG with the stream name as key
-            self._normalized_manifest[SCHEMAS_TAG][stream_name] = schema
-
-    def _reference_schemas(self) -> None:
-        """
-        Set the schema reference for the given stream in the manifest.
-        This function modifies the manifest in place.
-        """
-
-        # reference the stream schema for the stream to where it's stored
-        if SCHEMAS_TAG in self._normalized_manifest.keys():
-            for stream in self._get_manifest_streams():
-                self._extract_stream_schema(stream)
-                self._set_stream_schema_ref(stream)
-
-    def _set_stream_schema_ref(self, stream: Dict[str, Any]) -> None:
-        """
-        Set the schema reference for the given stream in the manifest.
-        This function modifies the manifest in place.
-        """
-        stream_name = stream["name"]
-        if SCHEMAS_TAG in self._normalized_manifest.keys():
-            if stream_name in self._normalized_manifest[SCHEMAS_TAG]:
-                stream[SCHEMA_LOADER_TAG][SCHEMA_TAG] = self._create_schema_ref(stream_name)
-
     def _replace_duplicates_with_refs(self, duplicates: DuplicatesType) -> None:
         """
         Process duplicate objects and replace them with references.
@@ -447,16 +517,3 @@ class ManifestNormalizer:
         """
 
         return {"$ref": f"#/{DEF_TAG}/{LINKED_TAG}/{type_key}/{key}"}
-
-    def _create_schema_ref(self, key: str) -> Dict[str, str]:
-        """
-        Create a reference object for stream schema using the specified key.
-
-        Args:
-            key: The reference key to use
-
-        Returns:
-            A reference object in the proper format
-        """
-
-        return {"$ref": f"#/{SCHEMAS_TAG}/{key}"}

{airbyte_cdk-6.56.5.post3.dev15889419925 → airbyte_cdk-6.56.6.post2.dev15930606991}/airbyte_cdk/sources/declarative/resolvers/config_components_resolver.py

@@ -11,6 +11,7 @@ import dpath
 import yaml
 from typing_extensions import deprecated
 from yaml.parser import ParserError
+from yaml.scanner import ScannerError
 
 from airbyte_cdk.sources.declarative.interpolation import InterpolatedString
 from airbyte_cdk.sources.declarative.resolvers.components_resolver import (
@@ -185,4 +186,8 @@ class ConfigComponentsResolver(ComponentsResolver):
                 return yaml.safe_load(value)
             except ParserError:  # "{{ record[0] in ['cohortActiveUsers'] }}"   # not valid YAML
                 return value
+            except ScannerError as e:  # "%Y-%m-%d'   # not valid yaml
+                if "expected alphabetic or numeric character, but found '%'" in str(e):
+                    return value
+                raise e
         return value

{airbyte_cdk-6.56.5.post3.dev15889419925 → airbyte_cdk-6.56.6.post2.dev15930606991}/airbyte_cdk/test/entrypoint_wrapper.py

@@ -373,6 +373,7 @@ def read(
     expecting_exception: bool | None = None,  # Deprecated, use `expected_outcome` instead.
     *,
     expected_outcome: ExpectedOutcome | None = None,
+    debug: bool = False,
 ) -> EntrypointOutput:
     """
     config and state must be json serializable
@@ -394,6 +395,8 @@ def read(
             "--catalog",
             catalog_file,
         ]
+        if debug:
+            args.append("--debug")
         if state is not None:
             args.extend(
                 [

{airbyte_cdk-6.56.5.post3.dev15889419925 → airbyte_cdk-6.56.6.post2.dev15930606991}/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 = "6.56.5.post3.dev15889419925"  # Version will be calculated dynamically.
+version = "6.56.6.post2.dev15930606991"  # Version will be calculated dynamically.
 
 [tool.poetry-dynamic-versioning]
 enable = false