Fixes

Author: vdusek

src/apify_client/_apify_client.py

@@ -12,12 +12,7 @@
     DEFAULT_TIMEOUT,
 )
 from apify_client._docs import docs_group
-from apify_client._http_clients import (
-    HttpClient,
-    HttpClientAsync,
-    ImpitHttpClient,
-    ImpitHttpClientAsync,
-)
+from apify_client._http_clients import HttpClient, HttpClientAsync, ImpitHttpClient, ImpitHttpClientAsync
 from apify_client._resource_clients import (
     ActorClient,
     ActorClientAsync,
@@ -75,6 +70,7 @@
     WebhookDispatchCollectionClientAsync,
 )
 from apify_client._statistics import ClientStatistics
+from apify_client._utils import check_custom_headers
 
 if TYPE_CHECKING:
     from datetime import timedelta
@@ -143,6 +139,9 @@ def __init__(
         api_url = DEFAULT_API_URL if api_url is None else api_url
         api_public_url = DEFAULT_API_URL if api_public_url is None else api_public_url
 
+        if headers:
+            check_custom_headers(self.__class__.__name__, headers)
+
         self._token = token
         """Apify API token for authentication."""
 
@@ -481,6 +480,9 @@ def __init__(
         api_url = DEFAULT_API_URL if api_url is None else api_url
         api_public_url = DEFAULT_API_URL if api_public_url is None else api_public_url
 
+        if headers:
+            check_custom_headers(self.__class__.__name__, headers)
+
         self._token = token
         """Apify API token for authentication."""
 
@@ -573,16 +575,6 @@ async def call(self, *, method, url, **kwargs) -> HttpResponse:
         instance._http_client = http_client
         return instance
 
-    @cached_property
-    def _base_kwargs(self) -> dict:
-        """Base keyword arguments for resource client construction."""
-        return {
-            'base_url': self._base_url,
-            'public_base_url': self._public_base_url,
-            'http_client': self.http_client,
-            'client_registry': self._client_registry,
-        }
-
     @property
     def token(self) -> str | None:
         """The Apify API token used by the client."""
@@ -606,6 +598,16 @@ def http_client(self) -> HttpClientAsync:
             )
         return self._http_client
 
+    @cached_property
+    def _base_kwargs(self) -> dict:
+        """Base keyword arguments for resource client construction."""
+        return {
+            'base_url': self._base_url,
+            'public_base_url': self._public_base_url,
+            'http_client': self.http_client,
+            'client_registry': self._client_registry,
+        }
+
     def actor(self, actor_id: str) -> ActorClientAsync:
         """Get the sub-client for a specific Actor.
 

src/apify_client/_consts.py

@@ -47,3 +47,6 @@
     }
 )
 """Set of terminal Actor job statuses that indicate the job has finished."""
+
+OVERRIDABLE_DEFAULT_HEADERS = {'Accept', 'Authorization', 'Accept-Encoding', 'User-Agent'}
+"""Headers that can be overridden by users, but will trigger a warning if they do so, as it may lead to API errors."""

src/apify_client/_utils.py

@@ -6,13 +6,15 @@
 import json
 import string
 import time
+import warnings
 from base64 import b64encode, urlsafe_b64encode
 from enum import Enum
 from http import HTTPStatus
 from typing import TYPE_CHECKING, Any, Literal, TypeVar, overload
 
 import impit
 
+from apify_client._consts import OVERRIDABLE_DEFAULT_HEADERS
 from apify_client.errors import InvalidResponseBodyError
 
 if TYPE_CHECKING:
@@ -346,3 +348,17 @@ def create_storage_content_signature(
 
     base64url_encoded_payload = urlsafe_b64encode(f'{version}.{expires_at}.{hmac_sig}'.encode())
     return base64url_encoded_payload.decode('utf-8')
+
+
+def check_custom_headers(class_name: str, headers: dict[str, str]) -> None:
+    """Warn if custom headers override important default headers."""
+    overwrite_headers = [key for key in headers if key.title() in OVERRIDABLE_DEFAULT_HEADERS]
+
+    if overwrite_headers:
+        warnings.warn(
+            f'{", ".join(overwrite_headers)} headers of {class_name} was overridden with an '
+            'explicit value. A wrong header value can lead to API errors, it is recommended to use the default '
+            f'value for following headers: {", ".join(OVERRIDABLE_DEFAULT_HEADERS)}.',
+            category=UserWarning,
+            stacklevel=3,
+        )