crawlee 0.6.13b17__py3-none-any.whl → 1.1.2b7__py3-none-any.whl

crawlee/_autoscaling/snapshotter.py

@@ -113,7 +113,7 @@ class Snapshotter:
         Args:
             config: The `Configuration` instance. Uses the global (default) one if not provided.
         """
-        config = service_locator.get_configuration()
+        config = config or service_locator.get_configuration()
 
         # Compute the maximum memory size based on the provided configuration. If `memory_mbytes` is provided,
         # it uses that value. Otherwise, it calculates the `max_memory_size` as a proportion of the system's

crawlee/_request.py

@@ -11,7 +11,7 @@ from yarl import URL
 from crawlee._types import EnqueueStrategy, HttpHeaders, HttpMethod, HttpPayload, JsonSerializable
 from crawlee._utils.crypto import crypto_random_object_id
 from crawlee._utils.docs import docs_group
-from crawlee._utils.requests import compute_unique_key, unique_key_to_request_id
+from crawlee._utils.requests import compute_unique_key
 from crawlee._utils.urls import validate_http_url
 
 if TYPE_CHECKING:
@@ -117,6 +117,7 @@ class UserData(BaseModel, MutableMapping[str, JsonSerializable]):
 user_data_adapter = TypeAdapter(UserData)
 
 
+@docs_group('Other')
 class RequestOptions(TypedDict):
     """Options that can be used to customize request creation.
 
@@ -163,11 +164,7 @@ class Request(BaseModel):
     ```
     """
 
-    model_config = ConfigDict(populate_by_name=True)
-
-    id: str
-    """A unique identifier for the request. Note that this is not used for deduplication, and should not be confused
-    with `unique_key`."""
+    model_config = ConfigDict(validate_by_name=True, validate_by_alias=True)
 
     unique_key: Annotated[str, Field(alias='uniqueKey')]
     """A unique key identifying the request. Two requests with the same `unique_key` are considered as pointing
@@ -188,9 +185,6 @@ class Request(BaseModel):
     method: HttpMethod = 'GET'
     """HTTP request method."""
 
-    headers: Annotated[HttpHeaders, Field(default_factory=HttpHeaders)] = HttpHeaders()
-    """HTTP request headers."""
-
     payload: Annotated[
         HttpPayload | None,
         BeforeValidator(lambda v: v.encode() if isinstance(v, str) else v),
@@ -198,23 +192,37 @@ class Request(BaseModel):
     ] = None
     """HTTP request payload."""
 
-    user_data: Annotated[
-        dict[str, JsonSerializable],  # Internally, the model contains `UserData`, this is just for convenience
-        Field(alias='userData', default_factory=lambda: UserData()),
-        PlainValidator(user_data_adapter.validate_python),
-        PlainSerializer(
-            lambda instance: user_data_adapter.dump_python(
-                instance,
-                by_alias=True,
-                exclude_none=True,
-                exclude_unset=True,
-                exclude_defaults=True,
-            )
-        ),
-    ] = {}
-    """Custom user data assigned to the request. Use this to save any request related data to the
-    request's scope, keeping them accessible on retries, failures etc.
-    """
+    # Workaround for pydantic 2.12 and mypy type checking issue for Annotated with default_factory
+    if TYPE_CHECKING:
+        headers: HttpHeaders = HttpHeaders()
+        """HTTP request headers."""
+
+        user_data: dict[str, JsonSerializable] = {}
+        """Custom user data assigned to the request. Use this to save any request related data to the
+        request's scope, keeping them accessible on retries, failures etc.
+        """
+
+    else:
+        headers: Annotated[HttpHeaders, Field(default_factory=HttpHeaders)]
+        """HTTP request headers."""
+
+        user_data: Annotated[
+            dict[str, JsonSerializable],  # Internally, the model contains `UserData`, this is just for convenience
+            Field(alias='userData', default_factory=lambda: UserData()),
+            PlainValidator(user_data_adapter.validate_python),
+            PlainSerializer(
+                lambda instance: user_data_adapter.dump_python(
+                    instance,
+                    by_alias=True,
+                    exclude_none=True,
+                    exclude_unset=True,
+                    exclude_defaults=True,
+                )
+            ),
+        ]
+        """Custom user data assigned to the request. Use this to save any request related data to the
+        request's scope, keeping them accessible on retries, failures etc.
+        """
 
     retry_count: Annotated[int, Field(alias='retryCount')] = 0
     """Number of times the request has been retried."""
@@ -239,7 +247,6 @@ class Request(BaseModel):
         label: str | None = None,
         session_id: str | None = None,
         unique_key: str | None = None,
-        id: str | None = None,
         keep_url_fragment: bool = False,
         use_extended_unique_key: bool = False,
         always_enqueue: bool = False,
@@ -264,8 +271,6 @@ class Request(BaseModel):
                 raised.
             unique_key: A unique key identifying the request. If not provided, it is automatically computed based on
                 the URL and other parameters. Requests with the same `unique_key` are treated as identical.
-            id: A unique identifier for the request. If not provided, it is automatically generated from the
-                `unique_key`.
             keep_url_fragment: Determines whether the URL fragment (e.g., `#section`) should be included in
                 the `unique_key` computation. This is only relevant when `unique_key` is not provided.
             use_extended_unique_key: Determines whether to include the HTTP method, ID Session and payload in the
@@ -294,14 +299,11 @@ class Request(BaseModel):
         )
 
         if always_enqueue:
-            unique_key = f'{unique_key}_{crypto_random_object_id()}'
-
-        id = id or unique_key_to_request_id(unique_key)
+            unique_key = f'{crypto_random_object_id()}|{unique_key}'
 
         request = cls(
             url=url,
             unique_key=unique_key,
-            id=id,
             method=method,
             headers=headers,
             payload=payload,

crawlee/_service_locator.py

@@ -11,6 +11,10 @@ from crawlee.storage_clients import FileSystemStorageClient, StorageClient
 if TYPE_CHECKING:
     from crawlee.storages._storage_instance_manager import StorageInstanceManager
 
+from logging import getLogger
+
+logger = getLogger(__name__)
+
 
 @docs_group('Configuration')
 class ServiceLocator:
@@ -19,23 +23,24 @@ class ServiceLocator:
     All services are initialized to its default value lazily.
     """
 
-    def __init__(self) -> None:
-        self._configuration: Configuration | None = None
-        self._event_manager: EventManager | None = None
-        self._storage_client: StorageClient | None = None
-        self._storage_instance_manager: StorageInstanceManager | None = None
+    global_storage_instance_manager: StorageInstanceManager | None = None
 
-        # Flags to check if the services were already set.
-        self._configuration_was_retrieved = False
-        self._event_manager_was_retrieved = False
-        self._storage_client_was_retrieved = False
+    def __init__(
+        self,
+        configuration: Configuration | None = None,
+        event_manager: EventManager | None = None,
+        storage_client: StorageClient | None = None,
+    ) -> None:
+        self._configuration = configuration
+        self._event_manager = event_manager
+        self._storage_client = storage_client
 
     def get_configuration(self) -> Configuration:
         """Get the configuration."""
         if self._configuration is None:
+            logger.debug('No configuration set, implicitly creating and using default Configuration.')
             self._configuration = Configuration()
 
-        self._configuration_was_retrieved = True
         return self._configuration
 
     def set_configuration(self, configuration: Configuration) -> None:
@@ -47,7 +52,10 @@ class ServiceLocator:
         Raises:
             ServiceConflictError: If the configuration has already been retrieved before.
         """
-        if self._configuration_was_retrieved:
+        if self._configuration is configuration:
+            # Same instance, no need to anything
+            return
+        if self._configuration:
             raise ServiceConflictError(Configuration, configuration, self._configuration)
 
         self._configuration = configuration
@@ -55,13 +63,14 @@ class ServiceLocator:
     def get_event_manager(self) -> EventManager:
         """Get the event manager."""
         if self._event_manager is None:
-            self._event_manager = (
-                LocalEventManager().from_config(config=self._configuration)
-                if self._configuration
-                else LocalEventManager.from_config()
-            )
+            logger.debug('No event manager set, implicitly creating and using default LocalEventManager.')
+            if self._configuration is None:
+                logger.debug(
+                    'Implicit creation of event manager will implicitly set configuration as side effect. '
+                    'It is advised to explicitly first set the configuration instead.'
+                )
+            self._event_manager = LocalEventManager().from_config(config=self._configuration)
 
-        self._event_manager_was_retrieved = True
         return self._event_manager
 
     def set_event_manager(self, event_manager: EventManager) -> None:
@@ -73,7 +82,10 @@ class ServiceLocator:
         Raises:
             ServiceConflictError: If the event manager has already been retrieved before.
         """
-        if self._event_manager_was_retrieved:
+        if self._event_manager is event_manager:
+            # Same instance, no need to anything
+            return
+        if self._event_manager:
             raise ServiceConflictError(EventManager, event_manager, self._event_manager)
 
         self._event_manager = event_manager
@@ -81,9 +93,14 @@ class ServiceLocator:
     def get_storage_client(self) -> StorageClient:
         """Get the storage client."""
         if self._storage_client is None:
+            logger.debug('No storage client set, implicitly creating and using default FileSystemStorageClient.')
+            if self._configuration is None:
+                logger.warning(
+                    'Implicit creation of storage client will implicitly set configuration as side effect. '
+                    'It is advised to explicitly first set the configuration instead.'
+                )
             self._storage_client = FileSystemStorageClient()
 
-        self._storage_client_was_retrieved = True
         return self._storage_client
 
     def set_storage_client(self, storage_client: StorageClient) -> None:
@@ -95,21 +112,24 @@ class ServiceLocator:
         Raises:
             ServiceConflictError: If the storage client has already been retrieved before.
         """
-        if self._storage_client_was_retrieved:
+        if self._storage_client is storage_client:
+            # Same instance, no need to anything
+            return
+        if self._storage_client:
             raise ServiceConflictError(StorageClient, storage_client, self._storage_client)
 
         self._storage_client = storage_client
 
     @property
     def storage_instance_manager(self) -> StorageInstanceManager:
-        """Get the storage instance manager."""
-        if self._storage_instance_manager is None:
+        """Get the storage instance manager. It is global manager shared by all instances of ServiceLocator."""
+        if ServiceLocator.global_storage_instance_manager is None:
             # Import here to avoid circular imports.
             from crawlee.storages._storage_instance_manager import StorageInstanceManager  # noqa: PLC0415
 
-            self._storage_instance_manager = StorageInstanceManager()
+            ServiceLocator.global_storage_instance_manager = StorageInstanceManager()
 
-        return self._storage_instance_manager
+        return ServiceLocator.global_storage_instance_manager
 
 
 service_locator = ServiceLocator()

crawlee/_types.py

@@ -3,17 +3,7 @@ from __future__ import annotations
 import dataclasses
 from collections.abc import Callable, Iterator, Mapping
 from dataclasses import dataclass
-from typing import (
-    TYPE_CHECKING,
-    Annotated,
-    Any,
-    Literal,
-    Protocol,
-    TypedDict,
-    TypeVar,
-    cast,
-    overload,
-)
+from typing import TYPE_CHECKING, Annotated, Any, Literal, Protocol, TypedDict, TypeVar, cast, overload
 
 from pydantic import ConfigDict, Field, PlainValidator, RootModel
 
@@ -25,7 +15,7 @@ if TYPE_CHECKING:
     import re
     from collections.abc import Callable, Coroutine, Sequence
 
-    from typing_extensions import NotRequired, Required, Unpack
+    from typing_extensions import NotRequired, Required, Self, Unpack
 
     from crawlee import Glob, Request
     from crawlee._request import RequestOptions
@@ -69,13 +59,17 @@ def _normalize_headers(headers: Mapping[str, str]) -> dict[str, str]:
 class HttpHeaders(RootModel, Mapping[str, str]):
     """A dictionary-like object representing HTTP headers."""
 
-    model_config = ConfigDict(populate_by_name=True)
+    model_config = ConfigDict(validate_by_name=True, validate_by_alias=True)
 
-    root: Annotated[
-        dict[str, str],
-        PlainValidator(lambda value: _normalize_headers(value)),
-        Field(default_factory=dict),
-    ] = {}
+    # Workaround for pydantic 2.12 and mypy type checking issue for Annotated with default_factory
+    if TYPE_CHECKING:
+        root: dict[str, str] = {}
+    else:
+        root: Annotated[
+            dict[str, str],
+            PlainValidator(lambda value: _normalize_headers(value)),
+            Field(default_factory=dict),
+        ]
 
     def __getitem__(self, key: str) -> str:
         return self.root[key.lower()]
@@ -110,9 +104,9 @@ class ConcurrencySettings:
     def __init__(
         self,
         min_concurrency: int = 1,
-        max_concurrency: int = 200,
+        max_concurrency: int = 100,
         max_tasks_per_minute: float = float('inf'),
-        desired_concurrency: int | None = None,
+        desired_concurrency: int = 10,
     ) -> None:
         """Initialize a new instance.
 
@@ -125,21 +119,24 @@ class ConcurrencySettings:
             desired_concurrency: The desired number of tasks that should be running parallel on the start of the pool,
                 if there is a large enough supply of them. By default, it is `min_concurrency`.
         """
-        if desired_concurrency is not None and desired_concurrency < 1:
-            raise ValueError('desired_concurrency must be 1 or larger')
-
         if min_concurrency < 1:
             raise ValueError('min_concurrency must be 1 or larger')
 
         if max_concurrency < min_concurrency:
             raise ValueError('max_concurrency cannot be less than min_concurrency')
 
+        if desired_concurrency < min_concurrency:
+            raise ValueError('desired_concurrency cannot be less than min_concurrency')
+
+        if desired_concurrency > max_concurrency:
+            raise ValueError('desired_concurrency cannot be greater than max_concurrency')
+
         if max_tasks_per_minute <= 0:
             raise ValueError('max_tasks_per_minute must be positive')
 
         self.min_concurrency = min_concurrency
         self.max_concurrency = max_concurrency
-        self.desired_concurrency = desired_concurrency if desired_concurrency is not None else min_concurrency
+        self.desired_concurrency = desired_concurrency
         self.max_tasks_per_minute = max_tasks_per_minute
 
 
@@ -180,6 +177,17 @@ class AddRequestsKwargs(EnqueueLinksKwargs):
     requests: Sequence[str | Request]
     """Requests to be added to the `RequestManager`."""
 
+    rq_id: str | None
+    """ID of the `RequestQueue` to add the requests to. Only one of `rq_id`, `rq_name` or `rq_alias` can be provided."""
+
+    rq_name: str | None
+    """Name of the `RequestQueue` to add the requests to. Only one of `rq_id`, `rq_name` or `rq_alias` can be provided.
+    """
+
+    rq_alias: str | None
+    """Alias of the `RequestQueue` to add the requests to. Only one of `rq_id`, `rq_name` or `rq_alias` can be provided.
+    """
+
 
 class PushDataKwargs(TypedDict):
     """Keyword arguments for dataset's `push_data` method."""
@@ -189,6 +197,7 @@ class PushDataFunctionCall(PushDataKwargs):
     data: list[dict[str, Any]] | dict[str, Any]
     dataset_id: str | None
     dataset_name: str | None
+    dataset_alias: str | None
 
 
 class KeyValueStoreInterface(Protocol):
@@ -255,21 +264,30 @@ class RequestHandlerRunResult:
         self._key_value_store_getter = key_value_store_getter
         self.add_requests_calls = list[AddRequestsKwargs]()
         self.push_data_calls = list[PushDataFunctionCall]()
-        self.key_value_store_changes = dict[tuple[str | None, str | None], KeyValueStoreChangeRecords]()
+        self.key_value_store_changes = dict[tuple[str | None, str | None, str | None], KeyValueStoreChangeRecords]()
 
     async def add_requests(
         self,
         requests: Sequence[str | Request],
+        rq_id: str | None = None,
+        rq_name: str | None = None,
+        rq_alias: str | None = None,
         **kwargs: Unpack[EnqueueLinksKwargs],
     ) -> None:
         """Track a call to the `add_requests` context helper."""
-        self.add_requests_calls.append(AddRequestsKwargs(requests=requests, **kwargs))
+        specified_params = sum(1 for param in [rq_id, rq_name, rq_alias] if param is not None)
+        if specified_params > 1:
+            raise ValueError('Only one of `rq_id`, `rq_name` or `rq_alias` can be provided.')
+        self.add_requests_calls.append(
+            AddRequestsKwargs(requests=requests, rq_id=rq_id, rq_name=rq_name, rq_alias=rq_alias, **kwargs)
+        )
 
     async def push_data(
         self,
         data: list[dict[str, Any]] | dict[str, Any],
         dataset_id: str | None = None,
         dataset_name: str | None = None,
+        dataset_alias: str | None = None,
         **kwargs: Unpack[PushDataKwargs],
     ) -> None:
         """Track a call to the `push_data` context helper."""
@@ -278,6 +296,7 @@ class RequestHandlerRunResult:
                 data=data,
                 dataset_id=dataset_id,
                 dataset_name=dataset_name,
+                dataset_alias=dataset_alias,
                 **kwargs,
             )
         )
@@ -287,13 +306,14 @@ class RequestHandlerRunResult:
         *,
         id: str | None = None,
         name: str | None = None,
+        alias: str | None = None,
     ) -> KeyValueStoreInterface:
-        if (id, name) not in self.key_value_store_changes:
-            self.key_value_store_changes[id, name] = KeyValueStoreChangeRecords(
-                await self._key_value_store_getter(id=id, name=name)
+        if (id, name, alias) not in self.key_value_store_changes:
+            self.key_value_store_changes[id, name, alias] = KeyValueStoreChangeRecords(
+                await self._key_value_store_getter(id=id, name=name, alias=alias)
             )
 
-        return self.key_value_store_changes[id, name]
+        return self.key_value_store_changes[id, name, alias]
 
 
 @docs_group('Functions')
@@ -307,12 +327,21 @@ class AddRequestsFunction(Protocol):
     def __call__(
         self,
         requests: Sequence[str | Request],
+        rq_id: str | None = None,
+        rq_name: str | None = None,
+        rq_alias: str | None = None,
         **kwargs: Unpack[EnqueueLinksKwargs],
     ) -> Coroutine[None, None, None]:
         """Call dunder method.
 
         Args:
             requests: Requests to be added to the `RequestManager`.
+            rq_id: ID of the `RequestQueue` to add the requests to. Only one of `rq_id`, `rq_name` or `rq_alias` can be
+                provided.
+            rq_name: Name of the `RequestQueue` to add the requests to. Only one of `rq_id`, `rq_name` or `rq_alias`
+                can be provided.
+            rq_alias: Alias of the `RequestQueue` to add the requests to. Only one of `rq_id`, `rq_name` or `rq_alias`
+                can be provided.
             **kwargs: Additional keyword arguments.
         """
 
@@ -340,12 +369,21 @@ class EnqueueLinksFunction(Protocol):
         label: str | None = None,
         user_data: dict[str, Any] | None = None,
         transform_request_function: Callable[[RequestOptions], RequestOptions | RequestTransformAction] | None = None,
+        rq_id: str | None = None,
+        rq_name: str | None = None,
+        rq_alias: str | None = None,
         **kwargs: Unpack[EnqueueLinksKwargs],
     ) -> Coroutine[None, None, None]: ...
 
     @overload
     def __call__(
-        self, *, requests: Sequence[str | Request] | None = None, **kwargs: Unpack[EnqueueLinksKwargs]
+        self,
+        *,
+        requests: Sequence[str | Request] | None = None,
+        rq_id: str | None = None,
+        rq_name: str | None = None,
+        rq_alias: str | None = None,
+        **kwargs: Unpack[EnqueueLinksKwargs],
     ) -> Coroutine[None, None, None]: ...
 
     def __call__(
@@ -356,6 +394,9 @@ class EnqueueLinksFunction(Protocol):
         user_data: dict[str, Any] | None = None,
         transform_request_function: Callable[[RequestOptions], RequestOptions | RequestTransformAction] | None = None,
         requests: Sequence[str | Request] | None = None,
+        rq_id: str | None = None,
+        rq_name: str | None = None,
+        rq_alias: str | None = None,
         **kwargs: Unpack[EnqueueLinksKwargs],
     ) -> Coroutine[None, None, None]:
         """Call enqueue links function.
@@ -373,6 +414,12 @@ class EnqueueLinksFunction(Protocol):
                 - `'skip'` to exclude the request from being enqueued,
                 - `'unchanged'` to use the original request options without modification.
             requests: Requests to be added to the `RequestManager`.
+            rq_id: ID of the `RequestQueue` to add the requests to. Only one of `rq_id`, `rq_name` or `rq_alias` can be
+                provided.
+            rq_name: Name of the `RequestQueue` to add the requests to. Only one of `rq_id`, `rq_name` or `rq_alias`
+                can be provided.
+            rq_alias: Alias of the `RequestQueue` to add the requests to. Only one of `rq_id`, `rq_name` or `rq_alias`
+                can be provided.
             **kwargs: Additional keyword arguments.
         """
 
@@ -424,12 +471,14 @@ class GetKeyValueStoreFunction(Protocol):
         *,
         id: str | None = None,
         name: str | None = None,
+        alias: str | None = None,
     ) -> Coroutine[None, None, KeyValueStore]:
         """Call dunder method.
 
         Args:
             id: The ID of the `KeyValueStore` to get.
-            name: The name of the `KeyValueStore` to get.
+            name: The name of the `KeyValueStore` to get (global scope, named storage).
+            alias: The alias of the `KeyValueStore` to get (run scope, unnamed storage).
         """
 
 
@@ -444,12 +493,14 @@ class GetKeyValueStoreFromRequestHandlerFunction(Protocol):
         *,
         id: str | None = None,
         name: str | None = None,
+        alias: str | None = None,
     ) -> Coroutine[None, None, KeyValueStoreInterface]:
         """Call dunder method.
 
         Args:
             id: The ID of the `KeyValueStore` to get.
-            name: The name of the `KeyValueStore` to get.
+            name: The name of the `KeyValueStore` to get (global scope, named storage).
+            alias: The alias of the `KeyValueStore` to get (run scope, unnamed storage).
         """
 
 
@@ -466,6 +517,7 @@ class PushDataFunction(Protocol):
         data: list[dict[str, Any]] | dict[str, Any],
         dataset_id: str | None = None,
         dataset_name: str | None = None,
+        dataset_alias: str | None = None,
         **kwargs: Unpack[PushDataKwargs],
     ) -> Coroutine[None, None, None]:
         """Call dunder method.
@@ -473,7 +525,8 @@ class PushDataFunction(Protocol):
         Args:
             data: The data to push to the `Dataset`.
             dataset_id: The ID of the `Dataset` to push the data to.
-            dataset_name: The name of the `Dataset` to push the data to.
+            dataset_name: The name of the `Dataset` to push the data to (global scope, named storage).
+            dataset_alias: The alias of the `Dataset` to push the data to (run scope, unnamed storage).
             **kwargs: Additional keyword arguments.
         """
 
@@ -590,6 +643,25 @@ class BasicCrawlingContext:
         """Return hash of the context. Each context is considered unique."""
         return id(self)
 
+    def create_modified_copy(
+        self,
+        push_data: PushDataFunction | None = None,
+        add_requests: AddRequestsFunction | None = None,
+        get_key_value_store: GetKeyValueStoreFromRequestHandlerFunction | None = None,
+    ) -> Self:
+        """Create a modified copy of the crawling context with specified changes."""
+        original_fields = {field.name: getattr(self, field.name) for field in dataclasses.fields(self)}
+        modified_fields = {
+            key: value
+            for key, value in {
+                'push_data': push_data,
+                'add_requests': add_requests,
+                'get_key_value_store': get_key_value_store,
+            }.items()
+            if value
+        }
+        return self.__class__(**{**original_fields, **modified_fields})
+
 
 class GetDataKwargs(TypedDict):
     """Keyword arguments for dataset's `get_data` method."""

crawlee/_utils/context.py

@@ -1,6 +1,6 @@
 from __future__ import annotations
 
-import asyncio
+import inspect
 from collections.abc import Callable
 from functools import wraps
 from typing import Any, TypeVar
@@ -44,4 +44,4 @@ def ensure_context(method: T) -> T:
 
         return await method(self, *args, **kwargs)
 
-    return async_wrapper if asyncio.iscoroutinefunction(method) else sync_wrapper  # type: ignore[return-value]
+    return async_wrapper if inspect.iscoroutinefunction(method) else sync_wrapper  # type: ignore[return-value]

crawlee/_utils/file.py

@@ -163,6 +163,13 @@ async def export_csv_to_stream(
     dst: TextIO,
     **kwargs: Unpack[ExportDataCsvKwargs],
 ) -> None:
+    # Set lineterminator to '\n' if not explicitly provided. This prevents double line endings on Windows.
+    # The csv.writer default is '\r\n', which when written to a file in text mode on Windows gets converted
+    # to '\r\r\n' due to newline translation. By using '\n', we let the platform handle the line ending
+    # conversion: '\n' stays as '\n' on Unix, and becomes '\r\n' on Windows.
+    if 'lineterminator' not in kwargs:
+        kwargs['lineterminator'] = '\n'
+
     writer = csv.writer(dst, **kwargs)  # type: ignore[arg-type]
     write_header = True
 

crawlee/_utils/raise_if_too_many_kwargs.py

@@ -0,0 +1,12 @@
+from typing import Any
+
+
+def raise_if_too_many_kwargs(max_kwargs: int = 1, **kwargs: Any) -> None:
+    """Raise ValueError if there are more non-None kwargs then max_kwargs."""
+    none_kwargs_names = [f'"{kwarg_name}"' for kwarg_name, value in kwargs.items() if value is not None]
+    if len(none_kwargs_names) > max_kwargs:
+        all_kwargs_names = [f'"{kwarg_name}"' for kwarg_name in kwargs]
+        raise ValueError(
+            f'Only one of {", ".join(all_kwargs_names)} can be specified, but following arguments were '
+            f'specified: {", ".join(none_kwargs_names)}.'
+        )