airbyte-cdk 6.18.1__tar.gz → 6.19.0__tar.gz

{airbyte_cdk-6.18.1 → airbyte_cdk-6.19.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: airbyte-cdk
-Version: 6.18.1
+Version: 6.19.0
 Summary: A framework for writing Airbyte Connectors.
 License: MIT
 Keywords: airbyte,connector-development-kit,cdk

{airbyte_cdk-6.18.1 → airbyte_cdk-6.19.0}/airbyte_cdk/sources/declarative/declarative_component_schema.yaml

@@ -678,7 +678,7 @@ definitions:
     properties:
       type:
         type: string
-        enum: [ CustomSchemaNormalization ]
+        enum: [CustomSchemaNormalization]
       class_name:
         title: Class Name
         description: Fully-qualified name of the class that will be implementing the custom normalization. The format is `source_<name>.<package>.<class_name>`.
@@ -2886,6 +2886,7 @@ definitions:
       parser:
         anyOf:
           - "$ref": "#/definitions/GzipParser"
+          - "$ref": "#/definitions/JsonParser"
           - "$ref": "#/definitions/JsonLineParser"
           - "$ref": "#/definitions/CsvParser"
   #    PARSERS
@@ -2902,6 +2903,20 @@ definitions:
         anyOf:
           - "$ref": "#/definitions/JsonLineParser"
           - "$ref": "#/definitions/CsvParser"
+          - "$ref": "#/definitions/JsonParser"
+  JsonParser:
+    title: JsonParser
+    description: Parser used for parsing str, bytes, or bytearray data and returning data in a dictionary format.
+    type: object
+    required:
+      - type
+    properties:
+      type:
+        type: string
+        enum: [JsonParser]
+      encoding:
+        type: string
+        default: utf-8
   JsonLineParser:
     type: object
     required:

{airbyte_cdk-6.18.1 → airbyte_cdk-6.19.0}/airbyte_cdk/sources/declarative/decoders/composite_raw_decoder.py

@@ -7,9 +7,12 @@ from dataclasses import dataclass
 from io import BufferedIOBase, TextIOWrapper
 from typing import Any, Generator, MutableMapping, Optional
 
+import orjson
 import requests
 
+from airbyte_cdk.models import FailureType
 from airbyte_cdk.sources.declarative.decoders.decoder import Decoder
+from airbyte_cdk.utils import AirbyteTracedException
 
 logger = logging.getLogger("airbyte")
 
@@ -42,6 +45,46 @@ class GzipParser(Parser):
             yield from self.inner_parser.parse(gzipobj)
 
 
+@dataclass
+class JsonParser(Parser):
+    encoding: str = "utf-8"
+
+    def parse(self, data: BufferedIOBase) -> Generator[MutableMapping[str, Any], None, None]:
+        """
+        Attempts to deserialize data using orjson library. As an extra layer of safety we fallback on the json library to deserialize the data.
+        """
+        raw_data = data.read()
+        body_json = self._parse_orjson(raw_data) or self._parse_json(raw_data)
+
+        if body_json is None:
+            raise AirbyteTracedException(
+                message="Response JSON data failed to be parsed. See logs for more information.",
+                internal_message=f"Response JSON data failed to be parsed.",
+                failure_type=FailureType.system_error,
+            )
+
+        if isinstance(body_json, list):
+            yield from body_json
+        else:
+            yield from [body_json]
+
+    def _parse_orjson(self, raw_data: bytes) -> Optional[Any]:
+        try:
+            return orjson.loads(raw_data.decode(self.encoding))
+        except Exception as exc:
+            logger.debug(
+                f"Failed to parse JSON data using orjson library. Falling back to json library. {exc}"
+            )
+            return None
+
+    def _parse_json(self, raw_data: bytes) -> Optional[Any]:
+        try:
+            return json.loads(raw_data.decode(self.encoding))
+        except Exception as exc:
+            logger.error(f"Failed to parse JSON data using json library. {exc}")
+            return None
+
+
 @dataclass
 class JsonLineParser(Parser):
     encoding: Optional[str] = "utf-8"

{airbyte_cdk-6.18.1 → airbyte_cdk-6.19.0}/airbyte_cdk/sources/declarative/models/declarative_component_schema.py

@@ -1201,6 +1201,14 @@ class LegacySessionTokenAuthenticator(BaseModel):
     parameters: Optional[Dict[str, Any]] = Field(None, alias="$parameters")
 
 
+class JsonParser(BaseModel):
+    class Config:
+        extra = Extra.allow
+
+    type: Literal["JsonParser"]
+    encoding: Optional[str] = "utf-8"
+
+
 class JsonLineParser(BaseModel):
     type: Literal["JsonLineParser"]
     encoding: Optional[str] = "utf-8"
@@ -1599,7 +1607,7 @@ class RecordSelector(BaseModel):
 
 class GzipParser(BaseModel):
     type: Literal["GzipParser"]
-    inner_parser: Union[JsonLineParser, CsvParser]
+    inner_parser: Union[JsonLineParser, CsvParser, JsonParser]
 
 
 class Spec(BaseModel):
@@ -1634,7 +1642,7 @@ class CompositeErrorHandler(BaseModel):
 
 class CompositeRawDecoder(BaseModel):
     type: Literal["CompositeRawDecoder"]
-    parser: Union[GzipParser, JsonLineParser, CsvParser]
+    parser: Union[GzipParser, JsonParser, JsonLineParser, CsvParser]
 
 
 class DeclarativeSource1(BaseModel):

{airbyte_cdk-6.18.1 → airbyte_cdk-6.19.0}/airbyte_cdk/sources/declarative/parsers/model_to_component_factory.py

@@ -72,6 +72,8 @@ from airbyte_cdk.sources.declarative.decoders.composite_raw_decoder import (
     CsvParser,
     GzipParser,
     JsonLineParser,
+    JsonParser,
+    Parser,
 )
 from airbyte_cdk.sources.declarative.extractors import (
     DpathExtractor,
@@ -247,6 +249,9 @@ from airbyte_cdk.sources.declarative.models.declarative_component_schema import
 from airbyte_cdk.sources.declarative.models.declarative_component_schema import (
     JsonLineParser as JsonLineParserModel,
 )
+from airbyte_cdk.sources.declarative.models.declarative_component_schema import (
+    JsonParser as JsonParserModel,
+)
 from airbyte_cdk.sources.declarative.models.declarative_component_schema import (
     JwtAuthenticator as JwtAuthenticatorModel,
 )
@@ -522,6 +527,7 @@ class ModelToComponentFactory:
             JsonDecoderModel: self.create_json_decoder,
             JsonlDecoderModel: self.create_jsonl_decoder,
             JsonLineParserModel: self.create_json_line_parser,
+            JsonParserModel: self.create_json_parser,
             GzipJsonDecoderModel: self.create_gzipjson_decoder,
             GzipParserModel: self.create_gzip_parser,
             KeysToLowerModel: self.create_keys_to_lower_transformation,
@@ -1032,17 +1038,17 @@ class ModelToComponentFactory:
         self, model: CursorPaginationModel, config: Config, decoder: Decoder, **kwargs: Any
     ) -> CursorPaginationStrategy:
         if isinstance(decoder, PaginationDecoderDecorator):
-            if not isinstance(decoder.decoder, (JsonDecoder, XmlDecoder)):
-                raise ValueError(
-                    f"Provided decoder of {type(decoder.decoder)=} is not supported. Please set JsonDecoder or XmlDecoder instead."
-                )
+            inner_decoder = decoder.decoder
+        else:
+            inner_decoder = decoder
+            decoder = PaginationDecoderDecorator(decoder=decoder)
+
+        if self._is_supported_decoder_for_pagination(inner_decoder):
             decoder_to_use = decoder
         else:
-            if not isinstance(decoder, (JsonDecoder, XmlDecoder)):
-                raise ValueError(
-                    f"Provided decoder of {type(decoder)=} is not supported. Please set JsonDecoder or XmlDecoder instead."
-                )
-            decoder_to_use = PaginationDecoderDecorator(decoder=decoder)
+            raise ValueError(
+                self._UNSUPPORTED_DECODER_ERROR.format(decoder_type=type(inner_decoder))
+            )
 
         return CursorPaginationStrategy(
             cursor_value=model.cursor_value,
@@ -1515,11 +1521,10 @@ class ModelToComponentFactory:
         cursor_used_for_stop_condition: Optional[DeclarativeCursor] = None,
     ) -> Union[DefaultPaginator, PaginatorTestReadDecorator]:
         if decoder:
-            if not isinstance(decoder, (JsonDecoder, XmlDecoder)):
-                raise ValueError(
-                    f"Provided decoder of {type(decoder)=} is not supported. Please set JsonDecoder or XmlDecoder instead."
-                )
-            decoder_to_use = PaginationDecoderDecorator(decoder=decoder)
+            if self._is_supported_decoder_for_pagination(decoder):
+                decoder_to_use = PaginationDecoderDecorator(decoder=decoder)
+            else:
+                raise ValueError(self._UNSUPPORTED_DECODER_ERROR.format(decoder_type=type(decoder)))
         else:
             decoder_to_use = PaginationDecoderDecorator(decoder=JsonDecoder(parameters={}))
         page_size_option = (
@@ -1748,6 +1753,11 @@ class ModelToComponentFactory:
     def create_json_decoder(model: JsonDecoderModel, config: Config, **kwargs: Any) -> JsonDecoder:
         return JsonDecoder(parameters={})
 
+    @staticmethod
+    def create_json_parser(model: JsonParserModel, config: Config, **kwargs: Any) -> JsonParser:
+        encoding = model.encoding if model.encoding else "utf-8"
+        return JsonParser(encoding=encoding)
+
     @staticmethod
     def create_jsonl_decoder(
         model: JsonlDecoderModel, config: Config, **kwargs: Any
@@ -1940,22 +1950,22 @@ class ModelToComponentFactory:
             message_repository=self._message_repository,
         )
 
-    @staticmethod
     def create_offset_increment(
-        model: OffsetIncrementModel, config: Config, decoder: Decoder, **kwargs: Any
+        self, model: OffsetIncrementModel, config: Config, decoder: Decoder, **kwargs: Any
     ) -> OffsetIncrement:
         if isinstance(decoder, PaginationDecoderDecorator):
-            if not isinstance(decoder.decoder, (JsonDecoder, XmlDecoder)):
-                raise ValueError(
-                    f"Provided decoder of {type(decoder.decoder)=} is not supported. Please set JsonDecoder or XmlDecoder instead."
-                )
+            inner_decoder = decoder.decoder
+        else:
+            inner_decoder = decoder
+            decoder = PaginationDecoderDecorator(decoder=decoder)
+
+        if self._is_supported_decoder_for_pagination(inner_decoder):
             decoder_to_use = decoder
         else:
-            if not isinstance(decoder, (JsonDecoder, XmlDecoder)):
-                raise ValueError(
-                    f"Provided decoder of {type(decoder)=} is not supported. Please set JsonDecoder or XmlDecoder instead."
-                )
-            decoder_to_use = PaginationDecoderDecorator(decoder=decoder)
+            raise ValueError(
+                self._UNSUPPORTED_DECODER_ERROR.format(decoder_type=type(inner_decoder))
+            )
+
         return OffsetIncrement(
             page_size=model.page_size,
             config=config,
@@ -2555,3 +2565,25 @@ class ModelToComponentFactory:
             components_mapping=components_mapping,
             parameters=model.parameters or {},
         )
+
+    _UNSUPPORTED_DECODER_ERROR = (
+        "Specified decoder of {decoder_type} is not supported for pagination."
+        "Please set as `JsonDecoder`, `XmlDecoder`, or a `CompositeRawDecoder` with an inner_parser of `JsonParser` or `GzipParser` instead."
+        "If using `GzipParser`, please ensure that the lowest level inner_parser is a `JsonParser`."
+    )
+
+    def _is_supported_decoder_for_pagination(self, decoder: Decoder) -> bool:
+        if isinstance(decoder, (JsonDecoder, XmlDecoder)):
+            return True
+        elif isinstance(decoder, CompositeRawDecoder):
+            return self._is_supported_parser_for_pagination(decoder.parser)
+        else:
+            return False
+
+    def _is_supported_parser_for_pagination(self, parser: Parser) -> bool:
+        if isinstance(parser, JsonParser):
+            return True
+        elif isinstance(parser, GzipParser):
+            return isinstance(parser.inner_parser, JsonParser)
+        else:
+            return False

{airbyte_cdk-6.18.1 → airbyte_cdk-6.19.0}/airbyte_cdk/sources/file_based/config/abstract_file_based_spec.py

@@ -31,6 +31,17 @@ class DeliverRawFiles(BaseModel):
 
     delivery_type: Literal["use_file_transfer"] = Field("use_file_transfer", const=True)
 
+    preserve_directory_structure: bool = Field(
+        title="Preserve Sub-Directories in File Paths",
+        description=(
+            "If enabled, sends subdirectory folder structure "
+            "along with source file names to the destination. "
+            "Otherwise, files will be synced by their names only. "
+            "This option is ignored when file-based replication is not enabled."
+        ),
+        default=True,
+    )
+
 
 class AbstractFileBasedSpec(BaseModel):
     """

{airbyte_cdk-6.18.1 → airbyte_cdk-6.19.0}/airbyte_cdk/sources/file_based/exceptions.py

@@ -111,6 +111,40 @@ class ErrorListingFiles(BaseFileBasedSourceError):
     pass
 
 
+class DuplicatedFilesError(BaseFileBasedSourceError):
+    def __init__(self, duplicated_files_names: List[dict[str, List[str]]], **kwargs: Any):
+        self._duplicated_files_names = duplicated_files_names
+        self._stream_name: str = kwargs["stream"]
+        super().__init__(self._format_duplicate_files_error_message(), **kwargs)
+
+    def _format_duplicate_files_error_message(self) -> str:
+        duplicated_files_messages = []
+        for duplicated_file in self._duplicated_files_names:
+            for duplicated_file_name, file_paths in duplicated_file.items():
+                file_duplicated_message = (
+                    f"{len(file_paths)} duplicates found for file name {duplicated_file_name}:\n\n"
+                    + "".join(f"\n - {file_paths}")
+                )
+                duplicated_files_messages.append(file_duplicated_message)
+
+        error_message = (
+            f"ERROR: Duplicate filenames found for stream {self._stream_name}. "
+            "Duplicate file names are not allowed if the Preserve Sub-Directories in File Paths option is disabled. "
+            "Please remove or rename the duplicate files before attempting to re-run the sync.\n\n"
+            + "\n".join(duplicated_files_messages)
+        )
+
+        return error_message
+
+    def __repr__(self) -> str:
+        """Return a string representation of the exception."""
+        class_name = self.__class__.__name__
+        properties_str = ", ".join(
+            f"{k}={v!r}" for k, v in self.__dict__.items() if not k.startswith("_")
+        )
+        return f"{class_name}({properties_str})"
+
+
 class CustomFileBasedException(AirbyteTracedException):
     """
     A specialized exception for file-based connectors.

{airbyte_cdk-6.18.1 → airbyte_cdk-6.19.0}/airbyte_cdk/sources/file_based/file_based_source.py

@@ -242,7 +242,7 @@ class FileBasedSource(ConcurrentSourceAdapter, ABC):
                         stream=self._make_default_stream(
                             stream_config=stream_config,
                             cursor=cursor,
-                            use_file_transfer=self._use_file_transfer(parsed_config),
+                            parsed_config=parsed_config,
                         ),
                         source=self,
                         logger=self.logger,
@@ -273,7 +273,7 @@ class FileBasedSource(ConcurrentSourceAdapter, ABC):
                         stream=self._make_default_stream(
                             stream_config=stream_config,
                             cursor=cursor,
-                            use_file_transfer=self._use_file_transfer(parsed_config),
+                            parsed_config=parsed_config,
                         ),
                         source=self,
                         logger=self.logger,
@@ -285,7 +285,7 @@ class FileBasedSource(ConcurrentSourceAdapter, ABC):
                     stream = self._make_default_stream(
                         stream_config=stream_config,
                         cursor=cursor,
-                        use_file_transfer=self._use_file_transfer(parsed_config),
+                        parsed_config=parsed_config,
                     )
 
                 streams.append(stream)
@@ -298,7 +298,7 @@ class FileBasedSource(ConcurrentSourceAdapter, ABC):
         self,
         stream_config: FileBasedStreamConfig,
         cursor: Optional[AbstractFileBasedCursor],
-        use_file_transfer: bool = False,
+        parsed_config: AbstractFileBasedSpec,
     ) -> AbstractFileBasedStream:
         return DefaultFileBasedStream(
             config=stream_config,
@@ -310,7 +310,8 @@ class FileBasedSource(ConcurrentSourceAdapter, ABC):
             validation_policy=self._validate_and_get_validation_policy(stream_config),
             errors_collector=self.errors_collector,
             cursor=cursor,
-            use_file_transfer=use_file_transfer,
+            use_file_transfer=self._use_file_transfer(parsed_config),
+            preserve_directory_structure=self._preserve_directory_structure(parsed_config),
         )
 
     def _get_stream_from_catalog(
@@ -385,3 +386,25 @@ class FileBasedSource(ConcurrentSourceAdapter, ABC):
             and parsed_config.delivery_method.delivery_type == "use_file_transfer"
         )
         return use_file_transfer
+
+    @staticmethod
+    def _preserve_directory_structure(parsed_config: AbstractFileBasedSpec) -> bool:
+        """
+        Determines whether to preserve directory structure during file transfer.
+
+        When enabled, files maintain their subdirectory paths in the destination.
+        When disabled, files are flattened to the root of the destination.
+
+        Args:
+            parsed_config: The parsed configuration containing delivery method settings
+
+        Returns:
+            True if directory structure should be preserved (default), False otherwise
+        """
+        if (
+            FileBasedSource._use_file_transfer(parsed_config)
+            and hasattr(parsed_config.delivery_method, "preserve_directory_structure")
+            and parsed_config.delivery_method.preserve_directory_structure is not None
+        ):
+            return parsed_config.delivery_method.preserve_directory_structure
+        return True

{airbyte_cdk-6.18.1 → airbyte_cdk-6.19.0}/airbyte_cdk/sources/file_based/file_based_stream_reader.py

@@ -135,6 +135,17 @@ class AbstractFileBasedStreamReader(ABC):
             return use_file_transfer
         return False
 
+    def preserve_directory_structure(self) -> bool:
+        # fall back to preserve subdirectories if config is not present or incomplete
+        if (
+            self.use_file_transfer()
+            and self.config
+            and hasattr(self.config.delivery_method, "preserve_directory_structure")
+            and self.config.delivery_method.preserve_directory_structure is not None
+        ):
+            return self.config.delivery_method.preserve_directory_structure
+        return True
+
     @abstractmethod
     def get_file(
         self, file: RemoteFile, local_directory: str, logger: logging.Logger
@@ -159,10 +170,13 @@ class AbstractFileBasedStreamReader(ABC):
         """
         ...
 
-    @staticmethod
-    def _get_file_transfer_paths(file: RemoteFile, local_directory: str) -> List[str]:
-        # Remove left slashes from source path format to make relative path for writing locally
-        file_relative_path = file.uri.lstrip("/")
+    def _get_file_transfer_paths(self, file: RemoteFile, local_directory: str) -> List[str]:
+        preserve_directory_structure = self.preserve_directory_structure()
+        if preserve_directory_structure:
+            # Remove left slashes from source path format to make relative path for writing locally
+            file_relative_path = file.uri.lstrip("/")
+        else:
+            file_relative_path = path.basename(file.uri)
         local_file_path = path.join(local_directory, file_relative_path)
 
         # Ensure the local directory exists

{airbyte_cdk-6.18.1 → airbyte_cdk-6.19.0}/airbyte_cdk/sources/file_based/file_types/unstructured_parser.py

@@ -2,6 +2,7 @@
 # Copyright (c) 2023 Airbyte, Inc., all rights reserved.
 #
 import logging
+import os
 import traceback
 from datetime import datetime
 from io import BytesIO, IOBase
@@ -42,12 +43,34 @@ unstructured_partition_pdf = None
 unstructured_partition_docx = None
 unstructured_partition_pptx = None
 
+AIRBYTE_NLTK_DATA_DIR = "/airbyte/nltk_data"
+TMP_NLTK_DATA_DIR = "/tmp/nltk_data"
+
+
+def get_nltk_temp_folder() -> str:
+    """
+    For non-root connectors /tmp is not currently writable, but we should allow it in the future.
+    It's safe to use /airbyte for now. Fallback to /tmp for local development.
+    """
+    try:
+        nltk_data_dir = AIRBYTE_NLTK_DATA_DIR
+        os.makedirs(nltk_data_dir, exist_ok=True)
+    except OSError:
+        nltk_data_dir = TMP_NLTK_DATA_DIR
+        os.makedirs(nltk_data_dir, exist_ok=True)
+    return nltk_data_dir
+
+
 try:
+    nltk_data_dir = get_nltk_temp_folder()
+    nltk.data.path.append(nltk_data_dir)
     nltk.data.find("tokenizers/punkt.zip")
     nltk.data.find("tokenizers/punkt_tab.zip")
+    nltk.data.find("tokenizers/averaged_perceptron_tagger_eng.zip")
 except LookupError:
-    nltk.download("punkt")
-    nltk.download("punkt_tab")
+    nltk.download("punkt", download_dir=nltk_data_dir, quiet=True)
+    nltk.download("punkt_tab", download_dir=nltk_data_dir, quiet=True)
+    nltk.download("averaged_perceptron_tagger_eng", download_dir=nltk_data_dir, quiet=True)
 
 
 def optional_decode(contents: Union[str, bytes]) -> str:

{airbyte_cdk-6.18.1 → airbyte_cdk-6.19.0}/airbyte_cdk/sources/file_based/stream/default_file_based_stream.py

@@ -5,14 +5,17 @@
 import asyncio
 import itertools
 import traceback
+from collections import defaultdict
 from copy import deepcopy
 from functools import cache
-from typing import Any, Dict, Iterable, List, Mapping, MutableMapping, Optional, Set, Union
+from os import path
+from typing import Any, Dict, Iterable, List, Mapping, MutableMapping, Optional, Set, Tuple, Union
 
 from airbyte_cdk.models import AirbyteLogMessage, AirbyteMessage, FailureType, Level
 from airbyte_cdk.models import Type as MessageType
 from airbyte_cdk.sources.file_based.config.file_based_stream_config import PrimaryKeyType
 from airbyte_cdk.sources.file_based.exceptions import (
+    DuplicatedFilesError,
     FileBasedSourceError,
     InvalidSchemaError,
     MissingSchemaError,
@@ -43,6 +46,8 @@ class DefaultFileBasedStream(AbstractFileBasedStream, IncrementalMixin):
     """
 
     FILE_TRANSFER_KW = "use_file_transfer"
+    PRESERVE_DIRECTORY_STRUCTURE_KW = "preserve_directory_structure"
+    FILES_KEY = "files"
     DATE_TIME_FORMAT = "%Y-%m-%dT%H:%M:%S.%fZ"
     ab_last_mod_col = "_ab_source_file_last_modified"
     ab_file_name_col = "_ab_source_file_url"
@@ -50,10 +55,15 @@ class DefaultFileBasedStream(AbstractFileBasedStream, IncrementalMixin):
     source_file_url = "source_file_url"
     airbyte_columns = [ab_last_mod_col, ab_file_name_col]
     use_file_transfer = False
+    preserve_directory_structure = True
 
     def __init__(self, **kwargs: Any):
         if self.FILE_TRANSFER_KW in kwargs:
             self.use_file_transfer = kwargs.pop(self.FILE_TRANSFER_KW, False)
+        if self.PRESERVE_DIRECTORY_STRUCTURE_KW in kwargs:
+            self.preserve_directory_structure = kwargs.pop(
+                self.PRESERVE_DIRECTORY_STRUCTURE_KW, True
+            )
         super().__init__(**kwargs)
 
     @property
@@ -98,15 +108,33 @@ class DefaultFileBasedStream(AbstractFileBasedStream, IncrementalMixin):
         else:
             return super()._filter_schema_invalid_properties(configured_catalog_json_schema)
 
+    def _duplicated_files_names(
+        self, slices: List[dict[str, List[RemoteFile]]]
+    ) -> List[dict[str, List[str]]]:
+        seen_file_names: Dict[str, List[str]] = defaultdict(list)
+        for file_slice in slices:
+            for file_found in file_slice[self.FILES_KEY]:
+                file_name = path.basename(file_found.uri)
+                seen_file_names[file_name].append(file_found.uri)
+        return [
+            {file_name: paths} for file_name, paths in seen_file_names.items() if len(paths) > 1
+        ]
+
     def compute_slices(self) -> Iterable[Optional[Mapping[str, Any]]]:
         # Sort files by last_modified, uri and return them grouped by last_modified
         all_files = self.list_files()
         files_to_read = self._cursor.get_files_to_sync(all_files, self.logger)
         sorted_files_to_read = sorted(files_to_read, key=lambda f: (f.last_modified, f.uri))
         slices = [
-            {"files": list(group[1])}
+            {self.FILES_KEY: list(group[1])}
             for group in itertools.groupby(sorted_files_to_read, lambda f: f.last_modified)
         ]
+        if slices and not self.preserve_directory_structure:
+            duplicated_files_names = self._duplicated_files_names(slices)
+            if duplicated_files_names:
+                raise DuplicatedFilesError(
+                    stream=self.name, duplicated_files_names=duplicated_files_names
+                )
         return slices
 
     def transform_record(

{airbyte_cdk-6.18.1 → airbyte_cdk-6.19.0}/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 = "6.18.1"  # Version will be calculated dynamically.
+version = "6.19.0"  # Version will be calculated dynamically.
 
 [tool.poetry-dynamic-versioning]
 enable = false