scrapely-python 0.1.0__py3-none-any.whl

scrapely/__init__.py

@@ -0,0 +1,246 @@
+"""
+Scrapely Python SDK - A toolkit for building Actors on the Scrapely platform.
+
+Usage:
+    from scrapely import Actor, Dataset
+
+    async def main():
+        async with Actor:
+            actor_input = await Actor.get_input()
+            await Actor.push_data({'url': actor_input['url']})
+"""
+
+# Side-effect: patch Crawlee Playwright browser pool to use remote CDP (CDP_URL).
+# Must run on package import so `from scrapely import Actor` enables the patch
+# before PlaywrightCrawler launches a browser (Node SDK parity).
+from .patch_browser_pool import set_cdp_url  # noqa: F401
+
+from crawlee import Request
+
+from crawlee.request_loaders import (
+    RequestLoader,
+    RequestManager,
+    RequestManagerTandem,
+    SitemapRequestLoader,
+)
+from crawlee.storages import Dataset, KeyValueStore, RequestQueue
+from crawlee.storage_clients.models import DatasetItemsListPage
+
+from .actor import Actor
+from .configuration import Configuration
+from .charging_client import (
+    USES_PUSH_DATA_INTERCEPTION,
+    create_charging_client,
+    push_data_charging_context,
+)
+from .storage import (
+    ApiStorageClient,
+    FileSystemStorageClient,
+    MemoryStorageClient,
+    ScrapelyStorageClient,
+    SmartScrapelyStorageClient,
+    StorageClient,
+    set_api_storage,
+)
+from .request_loaders import ScrapelyRequestList
+from crawlee.events import EventListener
+from .events import (
+    AbortingEvent,
+    ActorEventTypes,
+    CrawleeEventManager,
+    DeprecatedEvent,
+    Event,
+    EventAbortingData,
+    EventExitData,
+    EventManager,
+    EventMessage,
+    EventMigratingData,
+    EventPersistStateData,
+    EventSystemInfoData,
+    EventWithoutData,
+    ExitEvent,
+    LocalEventManager,
+    MigratingEvent,
+    PersistStateEvent,
+    ScrapelyEventManager,
+    SystemInfoEvent,
+    SystemInfoEventData,
+    UnknownEvent,
+)
+from .proxy_configuration import ProxyConfiguration
+from .browser_configuration import BrowserConfiguration
+from .charging import (
+    ChargingManager,
+    ChargeUncertainError,
+    ChargeRejectedError,
+    has_default_dataset_item_event_in_pricing,
+    is_default_dataset_item_event,
+    merge_charge_results,
+    push_data_and_charge,
+    resolve_default_dataset_item_event_name,
+    APIFY_DEFAULT_DATASET_ITEM_EVENT,
+    SCRAPELY_DEFAULT_DATASET_ITEM_EVENT,
+    DEFAULT_DATASET_ITEM_EVENT,
+)
+
+from .types import (
+    ActorInput,
+    ActorPermissionLevel,
+    ActorRun,
+    ApifyEnv,
+    AbortOptions,
+    CallOptions,
+    ExitOptions,
+    InitOptions,
+    MetamorphOptions,
+    OpenStorageOptions,
+    ProxyInfo,
+    RebootOptions,
+    RecordOptions,
+    SetStatusMessageOptions,
+    StartOptions,
+    WebhookOptions,
+)
+
+from .webhook import (
+    Webhook,
+    WebhookEventType,
+)
+from .pricing_types import (
+    PricingModel,
+    FreeActorPricingInfo,
+    FlatPricePerMonthActorPricingInfo,
+    PricePerDatasetItemActorPricingInfo,
+    PayPerEventActorPricingInfo,
+    PricingPerEvent,
+    ActorPricingInfo,
+    ActorChargeEvent,
+    ChargeResult,
+    ActorPricingInfoResult,
+)
+from .errors import (
+    ScrapelyClientError,
+    ScrapelyApiError,
+    InvalidResponseBodyError,
+    InvalidRequestError,
+    UnauthorizedError,
+    ForbiddenError,
+    NotFoundError,
+    ConflictError,
+    RateLimitError,
+    ServerError,
+    raise_for_status,
+)
+
+__version__ = '0.1.0'
+__all__ = [
+    # Core
+    'Actor',
+    'Configuration',
+    'set_cdp_url',
+    # ApifyClient
+    'create_charging_client',
+    'USES_PUSH_DATA_INTERCEPTION',
+    'push_data_charging_context',
+
+    # Storage
+    'Dataset',
+    'KeyValueStore',
+    'RequestQueue',
+    'Request',
+    'DatasetItemsListPage',
+    # Storage clients
+    'ApiStorageClient',
+    'ScrapelyStorageClient',
+    'SmartScrapelyStorageClient',
+    'FileSystemStorageClient',
+    'MemoryStorageClient',
+    'StorageClient',
+    'set_api_storage',
+    # Request loaders
+    'ScrapelyRequestList',
+    'RequestLoader',
+    'RequestManager',
+    'RequestManagerTandem',
+    'SitemapRequestLoader',
+    # Events
+    'Event',
+    'EventManager',
+    'ScrapelyEventManager',
+    'CrawleeEventManager',
+    'LocalEventManager',
+    # Event data types
+    'SystemInfoEventData',
+    'EventAbortingData',
+    'EventExitData',
+    'EventListener',
+    'EventMigratingData',
+    'EventPersistStateData',
+    'EventSystemInfoData',
+    # Event types
+    'PersistStateEvent',
+    'SystemInfoEvent',
+    'MigratingEvent',
+    'AbortingEvent',
+    'ExitEvent',
+    'EventWithoutData',
+    'DeprecatedEvent',
+    'UnknownEvent',
+    'EventMessage',
+    'ActorEventTypes',
+    # Proxy
+    'ProxyConfiguration',
+    'ProxyInfo',
+    # Remote browser
+    'BrowserConfiguration',
+    # Charging
+    'ChargingManager',
+    'ChargeResult',
+    'ChargeUncertainError',
+    'ChargeRejectedError',
+    'merge_charge_results',
+    'push_data_and_charge',
+
+    # Types
+    'ActorInput',
+    'ActorPermissionLevel',
+    'ActorRun',
+    'ApifyEnv',
+    'AbortOptions',
+    'CallOptions',
+    'ExitOptions',
+    'InitOptions',
+    'MetamorphOptions',
+    'OpenStorageOptions',
+    'RebootOptions',
+    'RecordOptions',
+    'SetStatusMessageOptions',
+    'StartOptions',
+    'WebhookOptions',
+
+    # Webhook
+    'Webhook',
+    'WebhookEventType',
+    # Pricing
+    'PricingModel',
+    'FreeActorPricingInfo',
+    'FlatPricePerMonthActorPricingInfo',
+    'PricePerDatasetItemActorPricingInfo',
+    'PayPerEventActorPricingInfo',
+    'PricingPerEvent',
+    'ActorPricingInfo',
+    'ActorChargeEvent',
+    'ActorPricingInfoResult',
+    # Errors
+    'ScrapelyClientError',
+    'ScrapelyApiError',
+    'InvalidResponseBodyError',
+    'InvalidRequestError',
+    'UnauthorizedError',
+    'ForbiddenError',
+    'NotFoundError',
+    'ConflictError',
+    'RateLimitError',
+    'ServerError',
+    'raise_for_status',
+]