apify 3.4.2b16__tar.gz → 3.4.2b18__tar.gz

{apify-3.4.2b16 → apify-3.4.2b18}/CHANGELOG.md

@@ -5,6 +5,10 @@ All notable changes to this project will be documented in this file.
 <!-- git-cliff-unreleased-start -->
 ## 3.4.2 - **not yet released**
 
+### 🚀 Features
+
+- Expose missing platform env vars via Actor.get_env() ([#984](https://github.com/apify/apify-sdk-python/pull/984)) ([f6e9b3b](https://github.com/apify/apify-sdk-python/commit/f6e9b3b328768aefb51257a258c4b6f03980fc72)) by [@vdusek](https://github.com/vdusek)
+
 ### 🐛 Bug Fixes
 
 - **scrapy:** Correct proxy middleware exception log and import ([#953](https://github.com/apify/apify-sdk-python/pull/953)) ([5bd6eb9](https://github.com/apify/apify-sdk-python/commit/5bd6eb9843d90844cec083372e932413bceedec9)) by [@vdusek](https://github.com/vdusek)
@@ -18,6 +22,8 @@ All notable changes to this project will be documented in this file.
 - Accept arbitrary JSON userData in ApifyRequestList ([#966](https://github.com/apify/apify-sdk-python/pull/966)) ([2cfd8a5](https://github.com/apify/apify-sdk-python/commit/2cfd8a590ae52a38b7efe39bb8a3a1570437b2fc)) by [@vdusek](https://github.com/vdusek)
 - Coerce null stats in Apify request queue metadata ([#974](https://github.com/apify/apify-sdk-python/pull/974)) ([63eb771](https://github.com/apify/apify-sdk-python/commit/63eb771e0234575ae3842cba9441938c3e4116fc)) by [@vdusek](https://github.com/vdusek)
 - Allow Actor.reboot() to be retried after a failed or cancelled attempt ([#968](https://github.com/apify/apify-sdk-python/pull/968)) ([7d46ec5](https://github.com/apify/apify-sdk-python/commit/7d46ec551252e7137f605e86894511315335f488)) by [@vdusek](https://github.com/vdusek)
+- Correct reclaim_request count adjustment for already-handled requests ([#973](https://github.com/apify/apify-sdk-python/pull/973)) ([86f4cd5](https://github.com/apify/apify-sdk-python/commit/86f4cd53d774e940954faa0b80b7ca5733d74fcf)) by [@vdusek](https://github.com/vdusek)
+- **scrapy:** Drop deprecated spider arg from Scrapy proxy middleware methods ([#977](https://github.com/apify/apify-sdk-python/pull/977)) ([49dd836](https://github.com/apify/apify-sdk-python/commit/49dd836c986b0e3bda4bb6485be8ad95d286cd82)) by [@vdusek](https://github.com/vdusek)
 
 ### 🚜 Refactor
 

{apify-3.4.2b16 → apify-3.4.2b18}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: apify
-Version: 3.4.2b16
+Version: 3.4.2b18
 Summary: Apify SDK for Python
 Project-URL: Apify Homepage, https://apify.com
 Project-URL: Changelog, https://docs.apify.com/sdk/python/docs/changelog

{apify-3.4.2b16 → apify-3.4.2b18}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "apify"
-version = "3.4.2b16"
+version = "3.4.2b18"
 description = "Apify SDK for Python"
 authors = [{ name = "Apify Technologies s.r.o.", email = "support@apify.com" }]
 license = { file = "LICENSE" }

{apify-3.4.2b16 → apify-3.4.2b18}/src/apify/_actor.py

@@ -118,7 +118,7 @@ class _ActorType:
             configuration: The Actor configuration to use. If not provided, a default configuration is created.
             configure_logging: Whether to set up the default logging configuration.
             exit_process: Whether the Actor should call `sys.exit` when the context manager exits.
-                Defaults to True, except in IPython, Pytest, and Scrapy environments.
+                Defaults to True, except in IPython and Scrapy environments.
             exit_code: The exit code the Actor should use when exiting.
             status_message: Final status message to display upon Actor termination.
             event_listeners_timeout: Maximum time to wait for Actor event listeners to complete before exiting.
@@ -912,7 +912,7 @@ class _ActorType:
                 a non-zero status code.
             memory_mbytes: Memory limit for the run, in megabytes. By default, the run uses a memory limit specified
                 in the default run configuration for the Actor.
-            timeout: Optional timeout for the run, in seconds. By default, the run uses timeout specified in
+            timeout: Optional timeout for the run. By default, the run uses timeout specified in
                 the default run configuration for the Actor. Using `inherit` will set timeout of the other Actor
                 to the time remaining from this Actor timeout.
             force_permission_level: Override the Actor's permissions for this run. If not set, the Actor will run
@@ -1019,7 +1019,7 @@ class _ActorType:
                 a non-zero status code.
             memory_mbytes: Memory limit for the run, in megabytes. By default, the run uses a memory limit specified
                 in the default run configuration for the Actor.
-            timeout: Optional timeout for the run, in seconds. By default, the run uses timeout specified in
+            timeout: Optional timeout for the run. By default, the run uses timeout specified in
                 the default run configuration for the Actor. Using `inherit` will set timeout of the other Actor
                 to the time remaining from this Actor timeout.
             force_permission_level: Override the Actor's permissions for this run. If not set, the Actor will run
@@ -1088,17 +1088,16 @@ class _ActorType:
         directly rather than an Actor task, please use the `Actor.call`
 
         Args:
-            task_id: The ID of the Actor to be run.
+            task_id: The ID of the Actor task to be run.
             task_input: Overrides the input to pass to the Actor run.
             token: The Apify API token to use for this request (defaults to the `APIFY_TOKEN` environment variable).
-            content_type: The content type of the input.
             build: Specifies the Actor build to run. It can be either a build tag or build number. By default,
                 the run uses the build specified in the default run configuration for the Actor (typically latest).
             restart_on_error: If true, the Task run process will be restarted whenever it exits with
                 a non-zero status code.
             memory_mbytes: Memory limit for the run, in megabytes. By default, the run uses a memory limit specified
                 in the default run configuration for the Actor.
-            timeout: Optional timeout for the run, in seconds. By default, the run uses timeout specified in
+            timeout: Optional timeout for the run. By default, the run uses timeout specified in
                 the default run configuration for the Actor. Using `inherit` will set timeout of the other Actor to the
                 time remaining from this Actor timeout.
             webhooks: Optional webhooks (https://docs.apify.com/webhooks) associated with the Actor run, which can
@@ -1217,7 +1216,7 @@ class _ActorType:
             # Call all the listeners for the PERSIST_STATE and MIGRATING events, and wait for them to finish.
             # PERSIST_STATE listeners are called to allow the Actor to persist its state before the reboot.
             # MIGRATING listeners are called to allow the Actor to gracefully stop in-progress tasks before
-            # the reboot. Typically, crawlers are listening for the MIIGRATING event to stop processing new requests.
+            # the reboot. Typically, crawlers are listening for the MIGRATING event to stop processing new requests.
             # We can't just emit the events and wait for all listeners to finish,
             # because this method might be called from an event listener itself, and we would deadlock.
             persist_state_listeners = flatten(

{apify-3.4.2b16 → apify-3.4.2b18}/src/apify/_configuration.py

@@ -426,7 +426,7 @@ class Configuration(CrawleeConfiguration):
                 'actor_web_server_port',
                 'apify_container_port',
             ),
-            description='TCP port for the Actor to start an HTTP server on'
+            description='TCP port for the Actor to start an HTTP server on. '
             'This server can be used to receive external messages or expose monitoring and control interfaces',
         ),
     ] = 4321
@@ -458,7 +458,7 @@ class Configuration(CrawleeConfiguration):
         | None,
         Field(
             alias='apify_actor_pricing_info',
-            description='JSON string with prising info of the actor',
+            description='JSON string with pricing info of the actor',
             discriminator='pricing_model',
         ),
         BeforeValidator(_parse_actor_pricing_info),

{apify-3.4.2b16 → apify-3.4.2b18}/src/apify/events/_types.py

@@ -20,9 +20,8 @@ from apify._utils import docs_group
 ActorEventTypes = Literal['systemInfo', 'persistState', 'migrating', 'aborting']
 """Event types emitted by the Apify platform during an Actor run.
 
-This is the Apify-specific subset of [`Event`][crawlee.events.Event] — for the full set
-(including framework-level events like `SESSION_RETIRED` or `BROWSER_LAUNCHED`) use
-[`Event`][crawlee.events.Event] from `apify`.
+This is the Apify-specific subset of `Event`. For the full set (including framework-level events
+like `SESSION_RETIRED` or `BROWSER_LAUNCHED`), use `Event` from `apify`.
 """
 
 

{apify-3.4.2b16 → apify-3.4.2b18}/src/apify/scrapy/middlewares/apify_proxy.py

@@ -10,7 +10,7 @@ from apify import Actor, ProxyConfiguration
 from apify.scrapy.utils import get_basic_auth_header
 
 if TYPE_CHECKING:
-    from scrapy import Request, Spider
+    from scrapy import Request
     from scrapy.crawler import Crawler
 
 
@@ -63,17 +63,16 @@ class ApifyHttpProxyMiddleware:
 
         return cls(proxy_settings)
 
-    async def process_request(self, request: Request, spider: Spider) -> None:
+    async def process_request(self, request: Request) -> None:
         """Process a Scrapy request by assigning a new proxy.
 
         Args:
             request: Scrapy Request object.
-            spider: Scrapy Spider object.
 
         Raises:
             ValueError: If username and password are not provided in the proxy URL.
         """
-        Actor.log.debug(f'ApifyHttpProxyMiddleware.process_request: request={request}, spider={spider}')
+        Actor.log.debug(f'ApifyHttpProxyMiddleware.process_request: request={request}')
         url = await self._get_new_proxy_url()
 
         if not (url.username and url.password):
@@ -89,14 +88,12 @@ class ApifyHttpProxyMiddleware:
         self,
         request: Request,
         exception: Exception,
-        spider: Spider,
     ) -> None:
         """Process an exception that occurs during request processing.
 
         Args:
             request: Scrapy Request object.
             exception: Exception object.
-            spider: Scrapy Spider object.
 
         Returns:
             Returning None, meaning Scrapy will continue processing this exception, executing any other
@@ -104,7 +101,7 @@ class ApifyHttpProxyMiddleware:
             exception handling kicks in.
         """
         Actor.log.debug(
-            f'ApifyHttpProxyMiddleware.process_exception: request={request}, exception={exception}, spider={spider}',
+            f'ApifyHttpProxyMiddleware.process_exception: request={request}, exception={exception}',
         )
 
         if isinstance(exception, TunnelError):

{apify-3.4.2b16 → apify-3.4.2b18}/src/apify/storage_clients/_apify/_models.py

@@ -102,7 +102,13 @@ class CachedRequest(BaseModel):
     """The expiration time of the lock on the request."""
 
 
+@docs_group('Storage data')
 class ApifyRequestQueueMetadata(RequestQueueMetadata):
+    """Extended request queue metadata model for Apify platform.
+
+    Includes additional Apify-specific fields.
+    """
+
     model_config = ConfigDict(alias_generator=to_camel)
 
     stats: Annotated[

{apify-3.4.2b16 → apify-3.4.2b18}/src/apify/storage_clients/_apify/_request_queue_shared_client.py

@@ -262,9 +262,9 @@ class ApifyRequestQueueSharedClient:
                 processed_request = await self._update_request(request, forefront=forefront)
                 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.
-                if request.was_already_handled and not processed_request.was_already_handled:
+                # The platform reports the request's state before this update via `was_already_handled`. If it was
+                # handled, this update moved it from handled back to pending, so mirror that in the local metadata.
+                if processed_request.was_already_handled:
                     self.metadata.handled_request_count -= 1
                     self.metadata.pending_request_count += 1
 

{apify-3.4.2b16 → apify-3.4.2b18}/src/apify/storage_clients/_apify/_request_queue_single_client.py

@@ -263,9 +263,9 @@ class ApifyRequestQueueSingleClient:
             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.
-            if request.was_already_handled and not processed_request.was_already_handled:
+            # The platform reports the request's state before this update via `was_already_handled`. If it was
+            # handled, this update moved it from handled back to pending, so mirror that in the local metadata.
+            if processed_request.was_already_handled:
                 self.metadata.handled_request_count -= 1
                 self.metadata.pending_request_count += 1
 

{apify-3.4.2b16 → apify-3.4.2b18}/src/apify/storage_clients/_file_system/_storage_client.py

@@ -9,6 +9,7 @@ from crawlee.storage_clients import FileSystemStorageClient
 
 from ._dataset_client import ApifyFileSystemDatasetClient
 from ._key_value_store_client import ApifyFileSystemKeyValueStoreClient
+from apify._utils import docs_group
 
 if TYPE_CHECKING:
     from collections.abc import Hashable
@@ -16,6 +17,7 @@ if TYPE_CHECKING:
     from crawlee.storage_clients._file_system import FileSystemKeyValueStoreClient
 
 
+@docs_group('Storage clients')
 class ApifyFileSystemStorageClient(FileSystemStorageClient):
     """Apify-specific implementation of the file system storage client.