apify 3.0.2b7__py3-none-any.whl → 3.0.3__py3-none-any.whl

apify/request_loaders/_apify_request_list.py

@@ -22,8 +22,8 @@ URL_NO_COMMAS_REGEX = re.compile(
 class _RequestDetails(BaseModel):
     method: HttpMethod = 'GET'
     payload: str = ''
-    headers: Annotated[dict[str, str], Field(default_factory=dict)] = {}
-    user_data: Annotated[dict[str, str], Field(default_factory=dict, alias='userData')] = {}
+    headers: Annotated[dict[str, str], Field(default_factory=dict)]
+    user_data: Annotated[dict[str, str], Field(default_factory=dict, alias='userData')]
 
 
 class _RequestsFromUrlInput(_RequestDetails):

apify/storage_clients/_apify/_dataset_client.py

@@ -1,19 +1,19 @@
 from __future__ import annotations
 
 import asyncio
+import warnings
 from logging import getLogger
 from typing import TYPE_CHECKING, Any
 
 from typing_extensions import override
 
-from apify_client import ApifyClientAsync
 from crawlee._utils.byte_size import ByteSize
 from crawlee._utils.file import json_dumps
 from crawlee.storage_clients._base import DatasetClient
 from crawlee.storage_clients.models import DatasetItemsListPage, DatasetMetadata
 from crawlee.storages import Dataset
 
-from ._utils import AliasResolver
+from ._utils import AliasResolver, create_apify_client
 
 if TYPE_CHECKING:
     from collections.abc import AsyncIterator
@@ -52,12 +52,17 @@ class ApifyDatasetClient(DatasetClient):
         self._api_client = api_client
         """The Apify dataset client for API operations."""
 
-        self._api_public_base_url = api_public_base_url
-        """The public base URL for accessing the key-value store records."""
-
         self._lock = lock
         """A lock to ensure that only one operation is performed at a time."""
 
+        if api_public_base_url:
+            # Remove in version 4.0, https://github.com/apify/apify-sdk-python/issues/635
+            warnings.warn(
+                'api_public_base_url argument is deprecated and will be removed in version 4.0.0',
+                DeprecationWarning,
+                stacklevel=2,
+            )
+
     @override
     async def get_metadata(self) -> DatasetMetadata:
         metadata = await self._api_client.get()
@@ -99,29 +104,7 @@ class ApifyDatasetClient(DatasetClient):
         if sum(1 for param in [id, name, alias] if param is not None) > 1:
             raise ValueError('Only one of "id", "name", or "alias" can be specified, not multiple.')
 
-        token = configuration.token
-        if not token:
-            raise ValueError(f'Apify storage client requires a valid token in Configuration (token={token}).')
-
-        api_url = configuration.api_base_url
-        if not api_url:
-            raise ValueError(f'Apify storage client requires a valid API URL in Configuration (api_url={api_url}).')
-
-        api_public_base_url = configuration.api_public_base_url
-        if not api_public_base_url:
-            raise ValueError(
-                'Apify storage client requires a valid API public base URL in Configuration '
-                f'(api_public_base_url={api_public_base_url}).'
-            )
-
-        # Create Apify client with the provided token and API URL.
-        apify_client_async = ApifyClientAsync(
-            token=token,
-            api_url=api_url,
-            max_retries=8,
-            min_delay_between_retries_millis=500,
-            timeout_secs=360,
-        )
+        apify_client_async = create_apify_client(configuration)
         apify_datasets_client = apify_client_async.datasets()
 
         # Normalize unnamed default storage in cases where not defined in `configuration.default_dataset_id` to unnamed
@@ -178,7 +161,7 @@ class ApifyDatasetClient(DatasetClient):
 
         return cls(
             api_client=apify_dataset_client,
-            api_public_base_url=api_public_base_url,
+            api_public_base_url='',  # Remove in version 4.0, https://github.com/apify/apify-sdk-python/issues/635
             lock=asyncio.Lock(),
         )
 

apify/storage_clients/_apify/_key_value_store_client.py

@@ -1,20 +1,18 @@
 from __future__ import annotations
 
 import asyncio
+import warnings
 from logging import getLogger
 from typing import TYPE_CHECKING, Any
 
 from typing_extensions import override
-from yarl import URL
 
-from apify_client import ApifyClientAsync
 from crawlee.storage_clients._base import KeyValueStoreClient
 from crawlee.storage_clients.models import KeyValueStoreRecord, KeyValueStoreRecordMetadata
 from crawlee.storages import KeyValueStore
 
 from ._models import ApifyKeyValueStoreMetadata, KeyValueStoreListKeysPage
-from ._utils import AliasResolver
-from apify._crypto import create_hmac_signature
+from ._utils import AliasResolver, create_apify_client
 
 if TYPE_CHECKING:
     from collections.abc import AsyncIterator
@@ -43,12 +41,17 @@ class ApifyKeyValueStoreClient(KeyValueStoreClient):
         self._api_client = api_client
         """The Apify KVS client for API operations."""
 
-        self._api_public_base_url = api_public_base_url
-        """The public base URL for accessing the key-value store records."""
-
         self._lock = lock
         """A lock to ensure that only one operation is performed at a time."""
 
+        if api_public_base_url:
+            # Remove in version 4.0, https://github.com/apify/apify-sdk-python/issues/635
+            warnings.warn(
+                'api_public_base_url argument is deprecated and will be removed in version 4.0.0',
+                DeprecationWarning,
+                stacklevel=2,
+            )
+
     @override
     async def get_metadata(self) -> ApifyKeyValueStoreMetadata:
         metadata = await self._api_client.get()
@@ -90,29 +93,7 @@ class ApifyKeyValueStoreClient(KeyValueStoreClient):
         if sum(1 for param in [id, name, alias] if param is not None) > 1:
             raise ValueError('Only one of "id", "name", or "alias" can be specified, not multiple.')
 
-        token = configuration.token
-        if not token:
-            raise ValueError(f'Apify storage client requires a valid token in Configuration (token={token}).')
-
-        api_url = configuration.api_base_url
-        if not api_url:
-            raise ValueError(f'Apify storage client requires a valid API URL in Configuration (api_url={api_url}).')
-
-        api_public_base_url = configuration.api_public_base_url
-        if not api_public_base_url:
-            raise ValueError(
-                'Apify storage client requires a valid API public base URL in Configuration '
-                f'(api_public_base_url={api_public_base_url}).'
-            )
-
-        # Create Apify client with the provided token and API URL.
-        apify_client_async = ApifyClientAsync(
-            token=token,
-            api_url=api_url,
-            max_retries=8,
-            min_delay_between_retries_millis=500,
-            timeout_secs=360,
-        )
+        apify_client_async = create_apify_client(configuration)
         apify_kvss_client = apify_client_async.key_value_stores()
 
         # Normalize unnamed default storage in cases where not defined in `configuration.default_key_value_store_id` to
@@ -170,7 +151,7 @@ class ApifyKeyValueStoreClient(KeyValueStoreClient):
 
         return cls(
             api_client=apify_kvs_client,
-            api_public_base_url=api_public_base_url,
+            api_public_base_url='',  # Remove in version 4.0, https://github.com/apify/apify-sdk-python/issues/635
             lock=asyncio.Lock(),
         )
 
@@ -251,15 +232,4 @@ class ApifyKeyValueStoreClient(KeyValueStoreClient):
         Returns:
             A public URL that can be used to access the value of the given key in the KVS.
         """
-        if self._api_client.resource_id is None:
-            raise ValueError('resource_id cannot be None when generating a public URL')
-
-        public_url = (
-            URL(self._api_public_base_url) / 'v2' / 'key-value-stores' / self._api_client.resource_id / 'records' / key
-        )
-        metadata = await self.get_metadata()
-
-        if metadata.url_signing_secret_key is not None:
-            public_url = public_url.with_query(signature=create_hmac_signature(metadata.url_signing_secret_key, key))
-
-        return str(public_url)
+        return await self._api_client.get_record_public_url(key=key)

apify/storage_clients/_apify/_models.py

@@ -94,8 +94,8 @@ class CachedRequest(BaseModel):
     Only internal structure.
     """
 
-    unique_key: str
-    """Unique key of the request."""
+    id: str
+    """Id of the request."""
 
     was_already_handled: bool
     """Whether the request was already handled."""

apify/storage_clients/_apify/_request_queue_client.py

@@ -5,7 +5,6 @@ from typing import TYPE_CHECKING, Final, Literal
 
 from typing_extensions import override
 
-from apify_client import ApifyClientAsync
 from crawlee._utils.crypto import crypto_random_object_id
 from crawlee.storage_clients._base import RequestQueueClient
 from crawlee.storage_clients.models import AddRequestsResponse, ProcessedRequest, RequestQueueMetadata
@@ -14,7 +13,7 @@ from crawlee.storages import RequestQueue
 from ._models import ApifyRequestQueueMetadata, RequestQueueStats
 from ._request_queue_shared_client import ApifyRequestQueueSharedClient
 from ._request_queue_single_client import ApifyRequestQueueSingleClient
-from ._utils import AliasResolver
+from ._utils import AliasResolver, create_apify_client
 
 if TYPE_CHECKING:
     from collections.abc import Sequence
@@ -228,29 +227,7 @@ class ApifyRequestQueueClient(RequestQueueClient):
         if sum(1 for param in [id, name, alias] if param is not None) > 1:
             raise ValueError('Only one of "id", "name", or "alias" can be specified, not multiple.')
 
-        token = configuration.token
-        if not token:
-            raise ValueError(f'Apify storage client requires a valid token in Configuration (token={token}).')
-
-        api_url = configuration.api_base_url
-        if not api_url:
-            raise ValueError(f'Apify storage client requires a valid API URL in Configuration (api_url={api_url}).')
-
-        api_public_base_url = configuration.api_public_base_url
-        if not api_public_base_url:
-            raise ValueError(
-                'Apify storage client requires a valid API public base URL in Configuration '
-                f'(api_public_base_url={api_public_base_url}).'
-            )
-
-        # Create Apify client with the provided token and API URL.
-        apify_client_async = ApifyClientAsync(
-            token=token,
-            api_url=api_url,
-            max_retries=8,
-            min_delay_between_retries_millis=500,
-            timeout_secs=360,
-        )
+        apify_client_async = create_apify_client(configuration)
         apify_rqs_client = apify_client_async.request_queues()
 
         # Normalize unnamed default storage in cases where not defined in `configuration.default_request_queue_id` to

apify/storage_clients/_apify/_request_queue_shared_client.py

@@ -54,10 +54,10 @@ class ApifyRequestQueueSharedClient:
         """The Apify request queue client for API operations."""
 
         self._queue_head = deque[str]()
-        """A deque to store request unique keys in the queue head."""
+        """A deque to store request ids in the queue head."""
 
         self._requests_cache: LRUCache[str, CachedRequest] = LRUCache(maxsize=cache_size)
-        """A cache to store request objects. Request unique key is used as the cache key."""
+        """A cache to store request objects. Request id is used as the cache key."""
 
         self._queue_has_locked_requests: bool | None = None
         """Whether the queue has requests locked by another client."""
@@ -101,12 +101,14 @@ class ApifyRequestQueueSharedClient:
         already_present_requests: list[ProcessedRequest] = []
 
         for request in requests:
-            if self._requests_cache.get(request.unique_key):
+            request_id = unique_key_to_request_id(request.unique_key)
+            if self._requests_cache.get(request_id):
                 # We are not sure if it was already handled at this point, and it is not worth calling API for it.
                 # It could have been handled by another client in the meantime, so cached information about
                 # `request.was_already_handled` is not reliable.
                 already_present_requests.append(
                     ProcessedRequest(
+                        id=request_id,
                         unique_key=request.unique_key,
                         was_already_present=True,
                         was_already_handled=request.was_already_handled,
@@ -116,12 +118,13 @@ class ApifyRequestQueueSharedClient:
             else:
                 # Add new request to the cache.
                 processed_request = ProcessedRequest(
+                    id=request_id,
                     unique_key=request.unique_key,
                     was_already_present=True,
                     was_already_handled=request.was_already_handled,
                 )
                 self._cache_request(
-                    request.unique_key,
+                    request_id,
                     processed_request,
                 )
                 new_requests.append(request)
@@ -131,7 +134,6 @@ class ApifyRequestQueueSharedClient:
             requests_dict = [
                 request.model_dump(
                     by_alias=True,
-                    exclude={'id'},  # Exclude ID fields from requests since the API doesn't accept them.
                 )
                 for request in new_requests
             ]
@@ -146,7 +148,8 @@ class ApifyRequestQueueSharedClient:
 
             # Remove unprocessed requests from the cache
             for unprocessed_request in api_response.unprocessed_requests:
-                self._requests_cache.pop(unprocessed_request.unique_key, None)
+                unprocessed_request_id = unique_key_to_request_id(unprocessed_request.unique_key)
+                self._requests_cache.pop(unprocessed_request_id, None)
 
         else:
             api_response = AddRequestsResponse.model_validate(
@@ -179,7 +182,10 @@ class ApifyRequestQueueSharedClient:
         Returns:
             The request or None if not found.
         """
-        response = await self._api_client.get_request(unique_key_to_request_id(unique_key))
+        return await self._get_request_by_id(unique_key_to_request_id(unique_key))
+
+    async def _get_request_by_id(self, request_id: str) -> Request | None:
+        response = await self._api_client.get_request(request_id)
 
         if response is None:
             return None
@@ -206,15 +212,15 @@ class ApifyRequestQueueSharedClient:
                 return None
 
             # Get the next request ID from the queue head
-            next_unique_key = self._queue_head.popleft()
+            next_request_id = self._queue_head.popleft()
 
-        request = await self._get_or_hydrate_request(next_unique_key)
+        request = await self._get_or_hydrate_request(next_request_id)
 
         # Handle potential inconsistency where request might not be in the main table yet
         if request is None:
             logger.debug(
                 'Cannot find a request from the beginning of queue, will be retried later',
-                extra={'nextRequestUniqueKey': next_unique_key},
+                extra={'next_request_id': next_request_id},
             )
             return None
 
@@ -222,16 +228,16 @@ class ApifyRequestQueueSharedClient:
         if request.handled_at is not None:
             logger.debug(
                 'Request fetched from the beginning of queue was already handled',
-                extra={'nextRequestUniqueKey': next_unique_key},
+                extra={'next_request_id': next_request_id},
             )
             return None
 
         # Use get request to ensure we have the full request object.
-        request = await self.get_request(request.unique_key)
+        request = await self._get_request_by_id(next_request_id)
         if request is None:
             logger.debug(
                 'Request fetched from the beginning of queue was not found in the RQ',
-                extra={'nextRequestUniqueKey': next_unique_key},
+                extra={'next_request_id': next_request_id},
             )
             return None
 
@@ -248,15 +254,17 @@ class ApifyRequestQueueSharedClient:
         Returns:
             Information about the queue operation. `None` if the given request was not in progress.
         """
+        request_id = unique_key_to_request_id(request.unique_key)
         # Set the handled_at timestamp if not already set
         if request.handled_at is None:
             request.handled_at = datetime.now(tz=timezone.utc)
 
-        if cached_request := self._requests_cache[request.unique_key]:
+        if cached_request := self._requests_cache[request_id]:
             cached_request.was_already_handled = request.was_already_handled
         try:
             # Update the request in the API
             processed_request = await self._update_request(request)
+            processed_request.id = request_id
             processed_request.unique_key = request.unique_key
 
             # Update assumed handled count if this wasn't already handled
@@ -265,10 +273,9 @@ class ApifyRequestQueueSharedClient:
                 self.metadata.pending_request_count -= 1
 
             # Update the cache with the handled request
-            cache_key = request.unique_key
             self._cache_request(
-                cache_key,
-                processed_request,
+                cache_key=request_id,
+                processed_request=processed_request,
                 hydrated_request=request,
             )
         except Exception as exc:
@@ -352,17 +359,17 @@ class ApifyRequestQueueSharedClient:
         # Fetch requests from the API and populate the queue head
         await self._list_head()
 
-    async def _get_or_hydrate_request(self, unique_key: str) -> Request | None:
-        """Get a request by unique key, either from cache or by fetching from API.
+    async def _get_or_hydrate_request(self, request_id: str) -> Request | None:
+        """Get a request by id, either from cache or by fetching from API.
 
         Args:
-            unique_key: Unique key of the request to get.
+            request_id: Id of the request to get.
 
         Returns:
             The request if found and valid, otherwise None.
         """
         # First check if the request is in our cache
-        cached_entry = self._requests_cache.get(unique_key)
+        cached_entry = self._requests_cache.get(request_id)
 
         if cached_entry and cached_entry.hydrated:
             # If we have the request hydrated in cache, return it
@@ -371,17 +378,17 @@ class ApifyRequestQueueSharedClient:
         # If not in cache or not hydrated, fetch the request
         try:
             # Fetch the request data
-            request = await self.get_request(unique_key)
+            request = await self._get_request_by_id(request_id)
 
             # If request is not found and return None
             if not request:
                 return None
 
             # Update cache with hydrated request
-            cache_key = request.unique_key
             self._cache_request(
-                cache_key,
-                ProcessedRequest(
+                cache_key=request_id,
+                processed_request=ProcessedRequest(
+                    id=request_id,
                     unique_key=request.unique_key,
                     was_already_present=True,
                     was_already_handled=request.handled_at is not None,
@@ -389,7 +396,7 @@ class ApifyRequestQueueSharedClient:
                 hydrated_request=request,
             )
         except Exception as exc:
-            logger.debug(f'Error fetching request {unique_key}: {exc!s}')
+            logger.debug(f'Error fetching request {request_id}: {exc!s}')
             return None
         else:
             return request
@@ -438,8 +445,8 @@ class ApifyRequestQueueSharedClient:
             logger.debug(f'Using cached queue head with {len(self._queue_head)} requests')
             # Create a list of requests from the cached queue head
             items = []
-            for unique_key in list(self._queue_head)[:limit]:
-                cached_request = self._requests_cache.get(unique_key)
+            for request_id in list(self._queue_head)[:limit]:
+                cached_request = self._requests_cache.get(request_id)
                 if cached_request and cached_request.hydrated:
                     items.append(cached_request.hydrated)
 
@@ -472,32 +479,35 @@ class ApifyRequestQueueSharedClient:
 
         for request_data in response.get('items', []):
             request = Request.model_validate(request_data)
+            request_id = request_data.get('id')
 
             # Skip requests without ID or unique key
-            if not request.unique_key:
+            if not request.unique_key or not request_id:
                 logger.debug(
-                    'Skipping request from queue head, missing unique key',
+                    'Skipping request from queue head, missing unique key or id',
                     extra={
                         'unique_key': request.unique_key,
+                        'id': request_id,
                     },
                 )
                 continue
 
             # Cache the request
             self._cache_request(
-                request.unique_key,
+                request_id,
                 ProcessedRequest(
+                    id=request_id,
                     unique_key=request.unique_key,
                     was_already_present=True,
                     was_already_handled=False,
                 ),
                 hydrated_request=request,
             )
-            self._queue_head.append(request.unique_key)
+            self._queue_head.append(request_id)
 
-        for leftover_unique_key in leftover_buffer:
+        for leftover_id in leftover_buffer:
             # After adding new requests to the forefront, any existing leftover locked request is kept in the end.
-            self._queue_head.append(leftover_unique_key)
+            self._queue_head.append(leftover_id)
         return RequestQueueHead.model_validate(response)
 
     def _cache_request(
@@ -516,7 +526,7 @@ class ApifyRequestQueueSharedClient:
             hydrated_request: The hydrated request object, if available.
         """
         self._requests_cache[cache_key] = CachedRequest(
-            unique_key=processed_request.unique_key,
+            id=processed_request.id,
             was_already_handled=processed_request.was_already_handled,
             hydrated=hydrated_request,
             lock_expires_at=None,

apify/storage_clients/_apify/_request_queue_single_client.py

@@ -56,21 +56,21 @@ class ApifyRequestQueueSingleClient:
         """The Apify request queue client for API operations."""
 
         self._requests_cache: LRUCache[str, Request] = LRUCache(maxsize=cache_size)
-        """A cache to store request objects. Request unique key is used as the cache key."""
+        """A cache to store request objects. Request id is used as the cache key."""
 
         self._head_requests: deque[str] = deque()
-        """Ordered unique keys of requests that represent queue head."""
+        """Ordered ids of requests that represent queue head."""
 
         self._requests_already_handled: set[str] = set()
         """Local estimation of requests unique keys that are already present and handled on the platform.
 
         - To enhance local deduplication.
         - To reduce the _requests_cache size. Already handled requests are most likely not going to be needed again,
-        so no need to cache more than their unique_key.
+        so no need to cache more than their id.
         """
 
         self._requests_in_progress: set[str] = set()
-        """Set of requests unique keys that are being processed locally.
+        """Set of requests ids that are being processed locally.
 
         - To help decide if the RQ is finished or not. This is the only consumer, so it can be tracked locally.
         """
@@ -105,19 +105,24 @@ class ApifyRequestQueueSingleClient:
         already_present_requests: list[ProcessedRequest] = []
 
         for request in requests:
+            # Calculate id for request
+            request_id = unique_key_to_request_id(request.unique_key)
+
             # Check if request is known to be already handled (it has to be present as well.)
-            if request.unique_key in self._requests_already_handled:
+            if request_id in self._requests_already_handled:
                 already_present_requests.append(
                     ProcessedRequest(
+                        id=request_id,
                         unique_key=request.unique_key,
                         was_already_present=True,
                         was_already_handled=True,
                     )
                 )
             # Check if request is known to be already present, but unhandled
-            elif self._requests_cache.get(request.unique_key):
+            elif self._requests_cache.get(request_id):
                 already_present_requests.append(
                     ProcessedRequest(
+                        id=request_id,
                         unique_key=request.unique_key,
                         was_already_present=True,
                         was_already_handled=request.was_already_handled,
@@ -128,11 +133,11 @@ class ApifyRequestQueueSingleClient:
                 new_requests.append(request)
 
                 # Update local caches
-                self._requests_cache[request.unique_key] = request
+                self._requests_cache[request_id] = request
                 if forefront:
-                    self._head_requests.append(request.unique_key)
+                    self._head_requests.append(request_id)
                 else:
-                    self._head_requests.appendleft(request.unique_key)
+                    self._head_requests.appendleft(request_id)
 
         if new_requests:
             # Prepare requests for API by converting to dictionaries.
@@ -151,7 +156,7 @@ class ApifyRequestQueueSingleClient:
             api_response.processed_requests.extend(already_present_requests)
             # Remove unprocessed requests from the cache
             for unprocessed_request in api_response.unprocessed_requests:
-                self._requests_cache.pop(unprocessed_request.unique_key, None)
+                self._requests_cache.pop(unique_key_to_request_id(unprocessed_request.unique_key), None)
 
         else:
             api_response = AddRequestsResponse(
@@ -178,15 +183,39 @@ class ApifyRequestQueueSingleClient:
         Returns:
             The request or None if not found.
         """
-        if unique_key in self._requests_cache:
-            return self._requests_cache[unique_key]
+        return await self._get_request(id=unique_key_to_request_id(unique_key))
+
+    async def _get_request(self, id: str) -> Request | None:
+        """Get a request by id.
+
+        Args:
+            id: Id of request to get.
+
+        Returns:
+            The request or None if not found.
+        """
+        if id in self._requests_cache:
+            return self._requests_cache[id]
 
-        response = await self._api_client.get_request(unique_key_to_request_id(unique_key))
+        # Requests that were not added by this client are not in local cache. Fetch them from platform.
+        response = await self._api_client.get_request(id)
 
         if response is None:
             return None
 
-        return Request.model_validate(response)
+        request = Request.model_validate(response)
+
+        # Updated local caches
+        if id in self._requests_in_progress:
+            # No caching of requests that are already in progress, client is already aware of them.
+            pass
+        elif request.was_already_handled:
+            # Cache only id for already handled requests
+            self._requests_already_handled.add(id)
+        else:
+            # Cache full request for unhandled requests that are not yet in progress and are not yet handled.
+            self._requests_cache[id] = request
+        return request
 
     async def fetch_next_request(self) -> Request | None:
         """Return the next request in the queue to be processed.
@@ -202,13 +231,10 @@ class ApifyRequestQueueSingleClient:
         await self._ensure_head_is_non_empty()
 
         while self._head_requests:
-            request_unique_key = self._head_requests.pop()
-            if (
-                request_unique_key not in self._requests_in_progress
-                and request_unique_key not in self._requests_already_handled
-            ):
-                self._requests_in_progress.add(request_unique_key)
-                return await self.get_request(request_unique_key)
+            request_id = self._head_requests.pop()
+            if request_id not in self._requests_in_progress and request_id not in self._requests_already_handled:
+                self._requests_in_progress.add(request_id)
+                return await self._get_request(request_id)
         # No request locally and the ones returned from the platform are already in progress.
         return None
 
@@ -233,44 +259,19 @@ class ApifyRequestQueueSingleClient:
 
         # Update the cached data
         for request_data in response.get('items', []):
-            # Due to https://github.com/apify/apify-core/blob/v0.1377.0/src/api/src/lib/request_queues/request_queue.ts#L53,
-            # the list_head endpoint may return truncated fields for long requests (e.g., long URLs or unique keys).
-            # If truncation is detected, fetch the full request data by its ID from the API.
-            # This is a temporary workaround - the caching will be refactored to use request IDs instead of unique keys.
-            # See https://github.com/apify/apify-sdk-python/issues/630 for details.
-            if '[truncated]' in request_data['uniqueKey'] or '[truncated]' in request_data['url']:
-                request_data = await self._api_client.get_request(request_id=request_data['id'])  # noqa: PLW2901
-
             request = Request.model_validate(request_data)
+            request_id = request_data['id']
 
-            if request.unique_key in self._requests_in_progress:
+            if request_id in self._requests_in_progress:
                 # Ignore requests that are already in progress, we will not process them again.
                 continue
 
             if request.was_already_handled:
-                # Do not cache fully handled requests, we do not need them. Just cache their unique_key.
-                self._requests_already_handled.add(request.unique_key)
-            else:
-                # Only fetch the request if we do not know it yet.
-                if request.unique_key not in self._requests_cache:
-                    request_id = unique_key_to_request_id(request.unique_key)
-
-                    if request_data is not None and request_id != request_data['id']:
-                        logger.warning(
-                            f'Request ID mismatch: {request_id} != {request_data["id"]}, '
-                            'this may cause unexpected behavior.'
-                        )
-
-                    # See https://github.com/apify/apify-sdk-python/issues/630 for details.
-                    if '[truncated]' not in request.unique_key:
-                        request_data = await self._api_client.get_request(request_id=request_id)  # noqa: PLW2901
-                        request = Request.model_validate(request_data)
-
-                    self._requests_cache[request.unique_key] = request
-
-                # Add new requests to the end of the head, unless already present in head
-                if request.unique_key not in self._head_requests:
-                    self._head_requests.appendleft(request.unique_key)
+                # Do not cache fully handled requests, we do not need them. Just cache their id.
+                self._requests_already_handled.add(request_id)
+            # Add new requests to the end of the head, unless already present in head
+            elif request_id not in self._head_requests:
+                self._head_requests.appendleft(request_id)
 
     async def mark_request_as_handled(self, request: Request) -> ProcessedRequest | None:
         """Mark a request as handled after successful processing.
@@ -285,12 +286,14 @@ class ApifyRequestQueueSingleClient:
         """
         # Set the handled_at timestamp if not already set
 
+        request_id = unique_key_to_request_id(request.unique_key)
+
         if request.handled_at is None:
             request.handled_at = datetime.now(tz=timezone.utc)
             self.metadata.handled_request_count += 1
             self.metadata.pending_request_count -= 1
 
-        if cached_request := self._requests_cache.get(request.unique_key):
+        if cached_request := self._requests_cache.get(request_id):
             cached_request.handled_at = request.handled_at
 
         try:
@@ -299,10 +302,10 @@ class ApifyRequestQueueSingleClient:
             # adding to the queue.)
             processed_request = await self._update_request(request)
             # Remember that we handled this request, to optimize local deduplication.
-            self._requests_already_handled.add(request.unique_key)
+            self._requests_already_handled.add(request_id)
             # Remove request from cache. It will most likely not be needed.
-            self._requests_cache.pop(request.unique_key)
-            self._requests_in_progress.discard(request.unique_key)
+            self._requests_cache.pop(request_id)
+            self._requests_in_progress.discard(request_id)
 
         except Exception as exc:
             logger.debug(f'Error marking request {request.unique_key} as handled: {exc!s}')
@@ -329,23 +332,27 @@ class ApifyRequestQueueSingleClient:
         """
         # Check if the request was marked as handled and clear it. When reclaiming,
         # we want to put the request back for processing.
+
+        request_id = unique_key_to_request_id(request.unique_key)
+
         if request.was_already_handled:
             request.handled_at = None
 
         try:
             # Make sure request is in the local cache. We might need it.
-            self._requests_cache[request.unique_key] = request
+            self._requests_cache[request_id] = request
 
             # No longer in progress
-            self._requests_in_progress.discard(request.unique_key)
+            self._requests_in_progress.discard(request_id)
             # No longer handled
-            self._requests_already_handled.discard(request.unique_key)
+            self._requests_already_handled.discard(request_id)
 
             if forefront:
                 # Append to top of the local head estimation
-                self._head_requests.append(request.unique_key)
+                self._head_requests.append(request_id)
 
             processed_request = await self._update_request(request, forefront=forefront)
+            processed_request.id = request_id
             processed_request.unique_key = request.unique_key
             # If the request was previously handled, decrement our handled count since
             # we're putting it back for processing.
@@ -407,9 +414,11 @@ class ApifyRequestQueueSingleClient:
         response = await self._api_client.list_requests(limit=10_000)
         for request_data in response.get('items', []):
             request = Request.model_validate(request_data)
+            request_id = request_data['id']
+
             if request.was_already_handled:
-                # Cache just unique_key for deduplication
-                self._requests_already_handled.add(request.unique_key)
+                # Cache just id for deduplication
+                self._requests_already_handled.add(request_id)
             else:
                 # Cache full request
-                self._requests_cache[request.unique_key] = request
+                self._requests_cache[request_id] = request

apify/storage_clients/_apify/_utils.py

@@ -192,3 +192,30 @@ def unique_key_to_request_id(unique_key: str, *, request_id_length: int = 15) ->
 
     # Truncate the key to the desired length
     return url_safe_key[:request_id_length]
+
+
+def create_apify_client(configuration: Configuration) -> ApifyClientAsync:
+    """Create and return an ApifyClientAsync instance using the provided configuration."""
+    if not configuration.token:
+        raise ValueError(f'Apify storage client requires a valid token in Configuration (token={configuration.token}).')
+
+    api_url = configuration.api_base_url
+    if not api_url:
+        raise ValueError(f'Apify storage client requires a valid API URL in Configuration (api_url={api_url}).')
+
+    api_public_base_url = configuration.api_public_base_url
+    if not api_public_base_url:
+        raise ValueError(
+            'Apify storage client requires a valid API public base URL in Configuration '
+            f'(api_public_base_url={api_public_base_url}).'
+        )
+
+    # Create Apify client with the provided token and API URL.
+    return ApifyClientAsync(
+        token=configuration.token,
+        api_url=api_url,
+        api_public_url=api_public_base_url,
+        max_retries=8,
+        min_delay_between_retries_millis=500,
+        timeout_secs=360,
+    )

{apify-3.0.2b7.dist-info → apify-3.0.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: apify
-Version: 3.0.2b7
+Version: 3.0.3
 Summary: Apify SDK for Python
 Project-URL: Apify Homepage, https://apify.com
 Project-URL: Changelog, https://docs.apify.com/sdk/python/docs/changelog
@@ -225,7 +225,7 @@ Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
 Classifier: Topic :: Software Development :: Libraries
 Requires-Python: >=3.10
-Requires-Dist: apify-client<3.0.0,>=2.0.0
+Requires-Dist: apify-client<3.0.0,>=2.2.0
 Requires-Dist: apify-shared<3.0.0,>=2.0.0
 Requires-Dist: cachetools>=5.5.0
 Requires-Dist: crawlee<2.0.0,>=1.0.2

{apify-3.0.2b7.dist-info → apify-3.0.3.dist-info}/RECORD

@@ -14,7 +14,7 @@ apify/events/_apify_event_manager.py,sha256=yArFrKa4wWDZo32iwaA3F_w36VSJf1Yaj_L1
 apify/events/_types.py,sha256=F0BHgACqnRfmdQ9GUcpnZvPxzw2bdRr8BqbGSA4cHeQ,3050
 apify/events/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 apify/request_loaders/__init__.py,sha256=SJqW0FbdZKEtAMB5kBLgqROzh3KmQc2CNEIhJpTGdPQ,356
-apify/request_loaders/_apify_request_list.py,sha256=kurCxX2jAKzHJ5N1Co6KjIgptqgVmjR0WpT8bd6uK9A,6220
+apify/request_loaders/_apify_request_list.py,sha256=jbZTHK3ACbh4YauYVJgXHXxB3rPcots5JMcr3GdIMz8,6210
 apify/request_loaders/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 apify/scrapy/__init__.py,sha256=m2a0ts_JY9xJkBy4JU5mV8PJqjA3GGKLXBFu4nl-n-A,1048
 apify/scrapy/_actor_runner.py,sha256=rXWSnlQWGskDUH8PtLCv5SkOIx4AiVa4QbCYeCett5c,938
@@ -35,14 +35,14 @@ apify/scrapy/pipelines/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSu
 apify/storage_clients/__init__.py,sha256=JheTvNpVD_luQXC1KTEgtr6yVnuMEC9ajBNLCX3HuSo,358
 apify/storage_clients/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 apify/storage_clients/_apify/__init__.py,sha256=mtbVDsxqWL3kx30elnh0kAn2kZ4s3BBsWa15Y5e7RMU,347
-apify/storage_clients/_apify/_dataset_client.py,sha256=Bb3UwOaFkyuEY7tuBf8K46R4ZP_b1EaAkDOXOqwSoW8,12498
-apify/storage_clients/_apify/_key_value_store_client.py,sha256=42dARbLX2oeOW7uYYKkDyQbEriMuh55Mxh0SqvkOEGg,10529
-apify/storage_clients/_apify/_models.py,sha256=GEaN7Got1zIg42QPH36obHRWRDVNtzOkRuOWYRf9bFU,4572
-apify/storage_clients/_apify/_request_queue_client.py,sha256=PUIVmGQxqFTkRxW9FIFWjT0OeDyAGt-ULlW-rdQDTyc,14194
-apify/storage_clients/_apify/_request_queue_shared_client.py,sha256=uxkuIG1rgCArgs6agldC9vmB2bgrIlNnm1I214Gf6WA,20550
-apify/storage_clients/_apify/_request_queue_single_client.py,sha256=EuORHJnFLC1YAT6ZfQj-ayrfSJNpU4_61r_7uDyvwgA,18092
+apify/storage_clients/_apify/_dataset_client.py,sha256=qmCJyL1MN83tYRXmc31P6yMIXVZMyRrGjr7R6-86FSE,11869
+apify/storage_clients/_apify/_key_value_store_client.py,sha256=994a5bM_BGHIeirnny6QlXjy5CzMU2I9SmMksCbHCUY,9357
+apify/storage_clients/_apify/_models.py,sha256=XxBru5XFdj0jqX6V-uVahT-pMQU3pZ501aTNzXCuoMU,4556
+apify/storage_clients/_apify/_request_queue_client.py,sha256=tAyap34gpxvPiQ0McDjX5ojq1ZIZc4EI3PrW8VQqS4k,13292
+apify/storage_clients/_apify/_request_queue_shared_client.py,sha256=pWmd6aPxM-eZ6PC1MfsfTcjD2mGGpCDS3ZZ3cG_2MEA,20971
+apify/storage_clients/_apify/_request_queue_single_client.py,sha256=d2txMwxW1nlYnvjdOH8xpxhcOYNeyc1ousGHRE7jsPg,17468
 apify/storage_clients/_apify/_storage_client.py,sha256=hFl_PuX1UgOydBD6pieZ0u2NWbDmZV-i0qygKdsuHt4,4873
-apify/storage_clients/_apify/_utils.py,sha256=ywXoSM69amRokUZcshbAvQLIcSZq4L-bpYIGyeFxCGQ,7696
+apify/storage_clients/_apify/_utils.py,sha256=375gk_TJyMWIIgRbE9SS0hQup0h6sA3mzpTG53XIjkM,8769
 apify/storage_clients/_apify/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 apify/storage_clients/_file_system/__init__.py,sha256=rDbXatXV9wHKPhKTrXDzWnexhTm7sIJQWucMi-P-SD4,130
 apify/storage_clients/_file_system/_key_value_store_client.py,sha256=gxM3ap67PnY80Rd7P3onPAf2pksYpU0LoAlJdayEMdU,4179
@@ -51,7 +51,7 @@ apify/storage_clients/_smart_apify/__init__.py,sha256=614B2AaWY-dx6RQ6mod7VVR8gF
 apify/storage_clients/_smart_apify/_storage_client.py,sha256=GCPmVe_xWAFcO2Cuej4su4i97_d33Q9Ih_Sc5xW2Wa4,4674
 apify/storages/__init__.py,sha256=-9tEYJVabVs_eRVhUehxN58GH0UG8OfuGjGwuDieP2M,122
 apify/storages/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-apify-3.0.2b7.dist-info/METADATA,sha256=nwfRWA3Q2QIjiKSLd-jAX3Y8j9ksLx3XA3NBeZ4MZnY,22582
-apify-3.0.2b7.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-apify-3.0.2b7.dist-info/licenses/LICENSE,sha256=AsFjHssKjj4LGd2ZCqXn6FBzMqcWdjQre1byPPSypVw,11355
-apify-3.0.2b7.dist-info/RECORD,,
+apify-3.0.3.dist-info/METADATA,sha256=BghvJ18oGZT-KqyoV3wlC6WdJ-rVupfzSUaWXLix2-k,22580
+apify-3.0.3.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+apify-3.0.3.dist-info/licenses/LICENSE,sha256=AsFjHssKjj4LGd2ZCqXn6FBzMqcWdjQre1byPPSypVw,11355
+apify-3.0.3.dist-info/RECORD,,