crawlee 1.0.3b9__tar.gz → 1.0.4__tar.gz

{crawlee-1.0.3b9 → crawlee-1.0.4}/CHANGELOG.md

@@ -2,8 +2,18 @@
 
 All notable changes to this project will be documented in this file.
 
-<!-- git-cliff-unreleased-start -->
-## 1.0.3 - **not yet released**
+## [1.0.4](https://github.com/apify/crawlee-python/releases/tag/v1.0.4) (2025-10-24)
+
+### 🐛 Bug Fixes
+
+- Respect `enqueue_strategy` in `enqueue_links` ([#1505](https://github.com/apify/crawlee-python/pull/1505)) ([6ee04bc](https://github.com/apify/crawlee-python/commit/6ee04bc08c50a70f2e956a79d4ce5072a726c3a8)) by [@Mantisus](https://github.com/Mantisus), closes [#1504](https://github.com/apify/crawlee-python/issues/1504)
+- Exclude incorrect links before checking `robots.txt` ([#1502](https://github.com/apify/crawlee-python/pull/1502)) ([3273da5](https://github.com/apify/crawlee-python/commit/3273da5fee62ec9254666b376f382474c3532a56)) by [@Mantisus](https://github.com/Mantisus), closes [#1499](https://github.com/apify/crawlee-python/issues/1499)
+- Resolve compatibility issue between `SqlStorageClient` and `AdaptivePlaywrightCrawler` ([#1496](https://github.com/apify/crawlee-python/pull/1496)) ([ce172c4](https://github.com/apify/crawlee-python/commit/ce172c425a8643a1d4c919db4f5e5a6e47e91deb)) by [@Mantisus](https://github.com/Mantisus), closes [#1495](https://github.com/apify/crawlee-python/issues/1495)
+- Fix `BasicCrawler` statistics persistence ([#1490](https://github.com/apify/crawlee-python/pull/1490)) ([1eb1c19](https://github.com/apify/crawlee-python/commit/1eb1c19aa6f9dda4a0e3f7eda23f77a554f95076)) by [@Pijukatel](https://github.com/Pijukatel), closes [#1501](https://github.com/apify/crawlee-python/issues/1501)
+- Save context state in result for `AdaptivePlaywrightCrawler` after isolated processing in `SubCrawler` ([#1488](https://github.com/apify/crawlee-python/pull/1488)) ([62b7c70](https://github.com/apify/crawlee-python/commit/62b7c70b54085fc65a660062028014f4502beba9)) by [@Mantisus](https://github.com/Mantisus), closes [#1483](https://github.com/apify/crawlee-python/issues/1483)
+
+
+## [1.0.3](https://github.com/apify/crawlee-python/releases/tag/v1.0.3) (2025-10-17)
 
 ### 🐛 Bug Fixes
 
@@ -13,7 +23,6 @@ All notable changes to this project will be documented in this file.
 - Fix `KeyValueStore.auto_saved_value` failing in some scenarios ([#1438](https://github.com/apify/crawlee-python/pull/1438)) ([b35dee7](https://github.com/apify/crawlee-python/commit/b35dee78180e57161b826641d45a61b8d8f6ef51)) by [@Pijukatel](https://github.com/Pijukatel), closes [#1354](https://github.com/apify/crawlee-python/issues/1354)
 
 
-<!-- git-cliff-unreleased-end -->
 ## [1.0.2](https://github.com/apify/crawlee-python/releases/tag/v1.0.2) (2025-10-08)
 
 ### 🐛 Bug Fixes

{crawlee-1.0.3b9 → crawlee-1.0.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: crawlee
-Version: 1.0.3b9
+Version: 1.0.4
 Summary: Crawlee for Python
 Project-URL: Apify Homepage, https://apify.com
 Project-URL: Changelog, https://crawlee.dev/python/docs/changelog

{crawlee-1.0.3b9 → crawlee-1.0.4}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "crawlee"
-version = "1.0.3b9"
+version = "1.0.4"
 description = "Crawlee for Python"
 authors = [{ name = "Apify Technologies s.r.o.", email = "support@apify.com" }]
 license = { file = "LICENSE" }

{crawlee-1.0.3b9 → crawlee-1.0.4}/src/crawlee/_utils/recurring_task.py

@@ -7,6 +7,9 @@ from typing import TYPE_CHECKING
 if TYPE_CHECKING:
     from collections.abc import Callable
     from datetime import timedelta
+    from types import TracebackType
+
+    from typing_extensions import Self
 
 logger = getLogger(__name__)
 
@@ -26,6 +29,18 @@ class RecurringTask:
         self.delay = delay
         self.task: asyncio.Task | None = None
 
+    async def __aenter__(self) -> Self:
+        self.start()
+        return self
+
+    async def __aexit__(
+        self,
+        exc_type: type[BaseException] | None,
+        exc_value: BaseException | None,
+        exc_traceback: TracebackType | None,
+    ) -> None:
+        await self.stop()
+
     async def _wrapper(self) -> None:
         """Continuously execute the provided function with the specified delay.
 

{crawlee-1.0.3b9 → crawlee-1.0.4}/src/crawlee/_utils/urls.py

@@ -7,6 +7,7 @@ from yarl import URL
 
 if TYPE_CHECKING:
     from collections.abc import Iterator
+    from logging import Logger
 
 
 def is_url_absolute(url: str) -> bool:
@@ -22,13 +23,19 @@ def convert_to_absolute_url(base_url: str, relative_url: str) -> str:
     return str(URL(base_url).join(URL(relative_url)))
 
 
-def to_absolute_url_iterator(base_url: str, urls: Iterator[str]) -> Iterator[str]:
+def to_absolute_url_iterator(base_url: str, urls: Iterator[str], logger: Logger | None = None) -> Iterator[str]:
     """Convert an iterator of relative URLs to absolute URLs using a base URL."""
     for url in urls:
         if is_url_absolute(url):
             yield url
         else:
-            yield convert_to_absolute_url(base_url, url)
+            converted_url = convert_to_absolute_url(base_url, url)
+            # Skip the URL if conversion fails, probably due to an incorrect format, such as 'mailto:'.
+            if not is_url_absolute(converted_url):
+                if logger:
+                    logger.debug(f'Could not convert URL "{url}" to absolute using base URL "{base_url}". Skipping it.')
+                continue
+            yield converted_url
 
 
 _http_url_adapter = TypeAdapter(AnyHttpUrl)

{crawlee-1.0.3b9 → crawlee-1.0.4}/src/crawlee/crawlers/_abstract_http/_abstract_http_crawler.py

@@ -167,7 +167,9 @@ class AbstractHttpCrawler(
             kwargs.setdefault('strategy', 'same-hostname')
 
             links_iterator: Iterator[str] = iter(self._parser.find_links(parsed_content, selector=selector))
-            links_iterator = to_absolute_url_iterator(context.request.loaded_url or context.request.url, links_iterator)
+            links_iterator = to_absolute_url_iterator(
+                context.request.loaded_url or context.request.url, links_iterator, logger=context.log
+            )
 
             if robots_txt_file:
                 skipped, links_iterator = partition(lambda url: robots_txt_file.is_allowed(url), links_iterator)

{crawlee-1.0.3b9 → crawlee-1.0.4}/src/crawlee/crawlers/_adaptive_playwright/_adaptive_playwright_crawler.py

@@ -149,10 +149,6 @@ class AdaptivePlaywrightCrawler(
                 non-default configuration.
             kwargs: Additional keyword arguments to pass to the underlying `BasicCrawler`.
         """
-        # Some sub crawler kwargs are internally modified. Prepare copies.
-        basic_crawler_kwargs_for_static_crawler = deepcopy(kwargs)
-        basic_crawler_kwargs_for_pw_crawler = deepcopy(kwargs)
-
         # Adaptive crawling related.
         self.rendering_type_predictor = rendering_type_predictor or DefaultRenderingTypePredictor()
         self.result_checker = result_checker or (lambda _: True)
@@ -170,11 +166,11 @@ class AdaptivePlaywrightCrawler(
         # Each sub crawler will use custom logger .
         static_logger = getLogger('Subcrawler_static')
         static_logger.setLevel(logging.ERROR)
-        basic_crawler_kwargs_for_static_crawler['_logger'] = static_logger
+        basic_crawler_kwargs_for_static_crawler: _BasicCrawlerOptions = {'_logger': static_logger, **kwargs}
 
         pw_logger = getLogger('Subcrawler_playwright')
         pw_logger.setLevel(logging.ERROR)
-        basic_crawler_kwargs_for_pw_crawler['_logger'] = pw_logger
+        basic_crawler_kwargs_for_pw_crawler: _BasicCrawlerOptions = {'_logger': pw_logger, **kwargs}
 
         # Initialize sub crawlers to create their pipelines.
         static_crawler_class = AbstractHttpCrawler.create_parsed_http_crawler_class(static_parser=static_parser)
@@ -319,7 +315,7 @@ class AdaptivePlaywrightCrawler(
                 ),
                 logger=self._logger,
             )
-            return SubCrawlerRun(result=result)
+            return SubCrawlerRun(result=result, run_context=context_linked_to_result)
         except Exception as e:
             return SubCrawlerRun(exception=e)
 
@@ -375,7 +371,8 @@ class AdaptivePlaywrightCrawler(
                 self.track_http_only_request_handler_runs()
 
                 static_run = await self._crawl_one(rendering_type='static', context=context)
-                if static_run.result and self.result_checker(static_run.result):
+                if static_run.result and static_run.run_context and self.result_checker(static_run.result):
+                    self._update_context_from_copy(context, static_run.run_context)
                     self._context_result_map[context] = static_run.result
                     return
                 if static_run.exception:
@@ -406,13 +403,10 @@ class AdaptivePlaywrightCrawler(
         if pw_run.exception is not None:
             raise pw_run.exception
 
-        if pw_run.result:
-            self._context_result_map[context] = pw_run.result
-
+        if pw_run.result and pw_run.run_context:
             if should_detect_rendering_type:
                 detection_result: RenderingType
                 static_run = await self._crawl_one('static', context=context, state=old_state_copy)
-
                 if static_run.result and self.result_comparator(static_run.result, pw_run.result):
                     detection_result = 'static'
                 else:
@@ -421,6 +415,9 @@ class AdaptivePlaywrightCrawler(
                 context.log.debug(f'Detected rendering type {detection_result} for {context.request.url}')
                 self.rendering_type_predictor.store_result(context.request, detection_result)
 
+            self._update_context_from_copy(context, pw_run.run_context)
+            self._context_result_map[context] = pw_run.result
+
     def pre_navigation_hook(
         self,
         hook: Callable[[AdaptivePlaywrightPreNavCrawlingContext], Awaitable[None]] | None = None,
@@ -455,8 +452,32 @@ class AdaptivePlaywrightCrawler(
     def track_rendering_type_mispredictions(self) -> None:
         self.statistics.state.rendering_type_mispredictions += 1
 
+    def _update_context_from_copy(self, context: BasicCrawlingContext, context_copy: BasicCrawlingContext) -> None:
+        """Update mutable fields of `context` from `context_copy`.
+
+        Uses object.__setattr__ to bypass frozen dataclass restrictions,
+        allowing state synchronization after isolated crawler execution.
+        """
+        updating_attributes = {
+            'request': ('headers', 'user_data'),
+            'session': ('_user_data', '_usage_count', '_error_score', '_cookies'),
+        }
+
+        for attr, sub_attrs in updating_attributes.items():
+            original_sub_obj = getattr(context, attr)
+            copy_sub_obj = getattr(context_copy, attr)
+
+            # Check that both sub objects are not None
+            if original_sub_obj is None or copy_sub_obj is None:
+                continue
+
+            for sub_attr in sub_attrs:
+                new_value = getattr(copy_sub_obj, sub_attr)
+                object.__setattr__(original_sub_obj, sub_attr, new_value)
+
 
 @dataclass(frozen=True)
 class SubCrawlerRun:
     result: RequestHandlerRunResult | None = None
     exception: Exception | None = None
+    run_context: BasicCrawlingContext | None = None

{crawlee-1.0.3b9 → crawlee-1.0.4}/src/crawlee/crawlers/_basic/_basic_crawler.py

@@ -56,7 +56,7 @@ from crawlee.errors import (
     SessionError,
     UserDefinedErrorHandlerError,
 )
-from crawlee.events._types import Event, EventCrawlerStatusData
+from crawlee.events._types import Event, EventCrawlerStatusData, EventPersistStateData
 from crawlee.http_clients import ImpitHttpClient
 from crawlee.router import Router
 from crawlee.sessions import SessionPool
@@ -437,14 +437,23 @@ class BasicCrawler(Generic[TCrawlingContext, TStatisticsState]):
         self._statistics_log_format = statistics_log_format
 
         # Statistics
-        self._statistics = statistics or cast(
-            'Statistics[TStatisticsState]',
-            Statistics.with_default_state(
-                periodic_message_logger=self._logger,
-                statistics_log_format=self._statistics_log_format,
-                log_message='Current request statistics:',
-            ),
-        )
+        if statistics:
+            self._statistics = statistics
+        else:
+
+            async def persist_state_factory() -> KeyValueStore:
+                return await self.get_key_value_store()
+
+            self._statistics = cast(
+                'Statistics[TStatisticsState]',
+                Statistics.with_default_state(
+                    persistence_enabled=True,
+                    periodic_message_logger=self._logger,
+                    statistics_log_format=self._statistics_log_format,
+                    log_message='Current request statistics:',
+                    persist_state_kvs_factory=persist_state_factory,
+                ),
+            )
 
         # Additional context managers to enter and exit
         self._additional_context_managers = _additional_context_managers or []
@@ -689,7 +698,6 @@ class BasicCrawler(Generic[TCrawlingContext, TStatisticsState]):
         except CancelledError:
             pass
         finally:
-            await self._crawler_state_rec_task.stop()
             if threading.current_thread() is threading.main_thread():
                 with suppress(NotImplementedError):
                     asyncio.get_running_loop().remove_signal_handler(signal.SIGINT)
@@ -721,8 +729,6 @@ class BasicCrawler(Generic[TCrawlingContext, TStatisticsState]):
     async def _run_crawler(self) -> None:
         event_manager = self._service_locator.get_event_manager()
 
-        self._crawler_state_rec_task.start()
-
         # Collect the context managers to be entered. Context managers that are already active are excluded,
         # as they were likely entered by the caller, who will also be responsible for exiting them.
         contexts_to_enter = [
@@ -733,6 +739,7 @@ class BasicCrawler(Generic[TCrawlingContext, TStatisticsState]):
                 self._statistics,
                 self._session_pool if self._use_session_pool else None,
                 self._http_client,
+                self._crawler_state_rec_task,
                 *self._additional_context_managers,
             )
             if cm and getattr(cm, 'active', False) is False
@@ -744,6 +751,9 @@ class BasicCrawler(Generic[TCrawlingContext, TStatisticsState]):
 
             await self._autoscaled_pool.run()
 
+            # Emit PERSIST_STATE event when crawler is finishing to allow listeners to persist their state if needed
+            event_manager.emit(event=Event.PERSIST_STATE, event_data=EventPersistStateData(is_migrating=False))
+
     async def add_requests(
         self,
         requests: Sequence[str | Request],
@@ -972,6 +982,7 @@ class BasicCrawler(Generic[TCrawlingContext, TStatisticsState]):
                         label=label,
                         user_data=user_data,
                         transform_request_function=transform_request_function,
+                        **kwargs,
                     ),
                     rq_id=rq_id,
                     rq_name=rq_name,

{crawlee-1.0.3b9 → crawlee-1.0.4}/src/crawlee/crawlers/_playwright/_playwright_crawler.py

@@ -366,7 +366,9 @@ class PlaywrightCrawler(BasicCrawler[PlaywrightCrawlingContext, StatisticsState]
             links_iterator: Iterator[str] = iter(
                 [url for element in elements if (url := await element.get_attribute('href')) is not None]
             )
-            links_iterator = to_absolute_url_iterator(context.request.loaded_url or context.request.url, links_iterator)
+            links_iterator = to_absolute_url_iterator(
+                context.request.loaded_url or context.request.url, links_iterator, logger=context.log
+            )
 
             if robots_txt_file:
                 skipped, links_iterator = partition(lambda url: robots_txt_file.is_allowed(url), links_iterator)

{crawlee-1.0.3b9 → crawlee-1.0.4}/src/crawlee/statistics/_statistics.py

@@ -96,7 +96,7 @@ class Statistics(Generic[TStatisticsState]):
 
         self._state = RecoverableState(
             default_state=state_model(stats_id=self._id),
-            persist_state_key=persist_state_key or f'SDK_CRAWLER_STATISTICS_{self._id}',
+            persist_state_key=persist_state_key or f'__CRAWLER_STATISTICS_{self._id}',
             persistence_enabled=persistence_enabled,
             persist_state_kvs_name=persist_state_kvs_name,
             persist_state_kvs_factory=persist_state_kvs_factory,
@@ -130,6 +130,7 @@ class Statistics(Generic[TStatisticsState]):
         persistence_enabled: bool = False,
         persist_state_kvs_name: str | None = None,
         persist_state_key: str | None = None,
+        persist_state_kvs_factory: Callable[[], Coroutine[None, None, KeyValueStore]] | None = None,
         log_message: str = 'Statistics',
         periodic_message_logger: Logger | None = None,
         log_interval: timedelta = timedelta(minutes=1),
@@ -141,6 +142,7 @@ class Statistics(Generic[TStatisticsState]):
             persistence_enabled=persistence_enabled,
             persist_state_kvs_name=persist_state_kvs_name,
             persist_state_key=persist_state_key,
+            persist_state_kvs_factory=persist_state_kvs_factory,
             log_message=log_message,
             periodic_message_logger=periodic_message_logger,
             log_interval=log_interval,
@@ -187,7 +189,10 @@ class Statistics(Generic[TStatisticsState]):
         if not self._active:
             raise RuntimeError(f'The {self.__class__.__name__} is not active.')
 
-        self._state.current_value.crawler_finished_at = datetime.now(timezone.utc)
+        if not self.state.crawler_last_started_at:
+            raise RuntimeError('Statistics.state.crawler_last_started_at not set.')
+        self.state.crawler_finished_at = datetime.now(timezone.utc)
+        self.state.crawler_runtime += self.state.crawler_finished_at - self.state.crawler_last_started_at
 
         await self._state.teardown()
 
@@ -255,8 +260,7 @@ class Statistics(Generic[TStatisticsState]):
         if self._instance_start is None:
             raise RuntimeError('The Statistics object is not initialized')
 
-        crawler_runtime = datetime.now(timezone.utc) - self._instance_start
-        total_minutes = crawler_runtime.total_seconds() / 60
+        total_minutes = self.state.crawler_runtime.total_seconds() / 60
         state = self._state.current_value
         serialized_state = state.model_dump(by_alias=False)
 
@@ -267,7 +271,7 @@ class Statistics(Generic[TStatisticsState]):
             requests_failed_per_minute=math.floor(state.requests_failed / total_minutes) if total_minutes else 0,
             request_total_duration=state.request_total_finished_duration + state.request_total_failed_duration,
             requests_total=state.requests_failed + state.requests_finished,
-            crawler_runtime=crawler_runtime,
+            crawler_runtime=state.crawler_runtime,
             requests_finished=state.requests_finished,
             requests_failed=state.requests_failed,
             retry_histogram=serialized_state['request_retry_histogram'],