apify 3.0.4b1__tar.gz → 3.0.4b2__tar.gz

{apify-3.0.4b1 → apify-3.0.4b2}/CHANGELOG.md

@@ -8,6 +8,11 @@ All notable changes to this project will be documented in this file.
 ### 🐛 Bug Fixes
 
 - Fix type of `cloud_storage_client` in `SmartApifyStorageClient` ([#642](https://github.com/apify/apify-sdk-python/pull/642)) ([3bf285d](https://github.com/apify/apify-sdk-python/commit/3bf285d60f507730954986a80c19ed2e27a38f9c)) by [@vdusek](https://github.com/vdusek)
+- Fix local charging log dataset name ([#649](https://github.com/apify/apify-sdk-python/pull/649)) ([fdb1276](https://github.com/apify/apify-sdk-python/commit/fdb1276264aee2687596d87c96d19033fe915823)) by [@vdusek](https://github.com/vdusek), closes [#648](https://github.com/apify/apify-sdk-python/issues/648)
+
+### ⚡ Performance
+
+- Use Apify-provided environment variables to obtain PPE pricing information ([#644](https://github.com/apify/apify-sdk-python/pull/644)) ([0c32f29](https://github.com/apify/apify-sdk-python/commit/0c32f29d6a316f5bacc931595d694f262c925b2b)) by [@Mantisus](https://github.com/Mantisus), closes [#614](https://github.com/apify/apify-sdk-python/issues/614)
 
 
 <!-- git-cliff-unreleased-end -->

{apify-3.0.4b1 → apify-3.0.4b2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: apify
-Version: 3.0.4b1
+Version: 3.0.4b2
 Summary: Apify SDK for Python
 Project-URL: Apify Homepage, https://apify.com
 Project-URL: Changelog, https://docs.apify.com/sdk/python/docs/changelog

{apify-3.0.4b1 → apify-3.0.4b2}/docs/02_concepts/12_pay_per_event.mdx

@@ -47,6 +47,6 @@ It is encouraged to test your monetization code on your machine before releasing
 ACTOR_TEST_PAY_PER_EVENT=true python -m youractor
 ```
 
-If you also wish to see a log of all the events charged throughout the run, the Apify SDK keeps a log of charged events in a so called charging dataset. Your charging dataset can be found under the `charging_log` name (unless you change your storage settings, this dataset is stored in `storage/datasets/charging_log/`). Please note that this log is not available when running the Actor in production on the Apify platform.
+If you also wish to see a log of all the events charged throughout the run, the Apify SDK keeps a log of charged events in a so called charging dataset. Your charging dataset can be found under the `charging-log` name (unless you change your storage settings, this dataset is stored in `storage/datasets/charging-log/`). Please note that this log is not available when running the Actor in production on the Apify platform.
 
 Because pricing configuration is stored by the Apify platform, all events will have a default price of $1.

{apify-3.0.4b1 → apify-3.0.4b2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "apify"
-version = "3.0.4b1"
+version = "3.0.4b2"
 description = "Apify SDK for Python"
 authors = [{ name = "Apify Technologies s.r.o.", email = "support@apify.com" }]
 license = { file = "LICENSE" }

{apify-3.0.4b1 → apify-3.0.4b2}/src/apify/_charging.py

@@ -4,13 +4,20 @@ import math
 from dataclasses import dataclass
 from datetime import datetime, timezone
 from decimal import Decimal
-from typing import TYPE_CHECKING, Protocol
+from typing import TYPE_CHECKING, Protocol, TypedDict
 
 from pydantic import TypeAdapter
 
 from crawlee._utils.context import ensure_context
 
-from apify._models import ActorRun, PricingModel
+from apify._models import (
+    ActorRun,
+    FlatPricePerMonthActorPricingInfo,
+    FreeActorPricingInfo,
+    PayPerEventActorPricingInfo,
+    PricePerDatasetItemActorPricingInfo,
+    PricingModel,
+)
 from apify._utils import docs_group
 from apify.log import logger
 from apify.storages import Dataset
@@ -111,24 +118,16 @@ class ActorPricingInfo:
 class ChargingManagerImplementation(ChargingManager):
     """Implementation of the `ChargingManager` Protocol - this is only meant to be instantiated internally."""
 
-    LOCAL_CHARGING_LOG_DATASET_NAME = 'charging_log'
+    LOCAL_CHARGING_LOG_DATASET_NAME = 'charging-log'
 
     def __init__(self, configuration: Configuration, client: ApifyClientAsync) -> None:
         self._max_total_charge_usd = configuration.max_total_charge_usd or Decimal('inf')
+        self._configuration = configuration
         self._is_at_home = configuration.is_at_home
         self._actor_run_id = configuration.actor_run_id
         self._purge_charging_log_dataset = configuration.purge_on_start
         self._pricing_model: PricingModel | None = None
 
-        if configuration.test_pay_per_event:
-            if self._is_at_home:
-                raise ValueError(
-                    'Using the ACTOR_TEST_PAY_PER_EVENT environment variable is only supported '
-                    'in a local development environment'
-                )
-
-            self._pricing_model = 'PAY_PER_EVENT'
-
         self._client = client
         self._charging_log_dataset: Dataset | None = None
 
@@ -140,37 +139,46 @@ class ChargingManagerImplementation(ChargingManager):
 
     async def __aenter__(self) -> None:
         """Initialize the charging manager - this is called by the `Actor` class and shouldn't be invoked manually."""
-        self.active = True
-
-        if self._is_at_home:
-            # Running on the Apify platform - fetch pricing info for the current run.
-
-            if self._actor_run_id is None:
-                raise RuntimeError('Actor run ID not found even though the Actor is running on Apify')
+        # Validate config
+        if self._configuration.test_pay_per_event and self._is_at_home:
+            raise ValueError(
+                'Using the ACTOR_TEST_PAY_PER_EVENT environment variable is only supported '
+                'in a local development environment'
+            )
 
-            run = run_validator.validate_python(await self._client.run(self._actor_run_id).get())
-            if run is None:
-                raise RuntimeError('Actor run not found')
+        self.active = True
 
-            if run.pricing_info is not None:
-                self._pricing_model = run.pricing_info.pricing_model
+        # Retrieve pricing information from env vars or API
+        pricing_data = await self._fetch_pricing_info()
+        pricing_info = pricing_data['pricing_info']
+        charged_event_counts = pricing_data['charged_event_counts']
+        max_total_charge_usd = pricing_data['max_total_charge_usd']
 
-                if run.pricing_info.pricing_model == 'PAY_PER_EVENT':
-                    for event_name, event_pricing in run.pricing_info.pricing_per_event.actor_charge_events.items():
-                        self._pricing_info[event_name] = PricingInfoItem(
-                            price=event_pricing.event_price_usd,
-                            title=event_pricing.event_title,
-                        )
+        # Set pricing model
+        if self._configuration.test_pay_per_event:
+            self._pricing_model = 'PAY_PER_EVENT'
+        else:
+            self._pricing_model = pricing_info.pricing_model if pricing_info else None
+
+        # Load per-event pricing information
+        if pricing_info and pricing_info.pricing_model == 'PAY_PER_EVENT':
+            for event_name, event_pricing in pricing_info.pricing_per_event.actor_charge_events.items():
+                self._pricing_info[event_name] = PricingInfoItem(
+                    price=event_pricing.event_price_usd,
+                    title=event_pricing.event_title,
+                )
 
-                    self._max_total_charge_usd = run.options.max_total_charge_usd or self._max_total_charge_usd
+            self._max_total_charge_usd = max_total_charge_usd
 
-            for event_name, count in (run.charged_event_counts or {}).items():
-                price = self._pricing_info.get(event_name, PricingInfoItem(Decimal(), title='')).price
-                self._charging_state[event_name] = ChargingStateItem(
-                    charge_count=count,
-                    total_charged_amount=count * price,
-                )
+        # Load charged event counts
+        for event_name, count in charged_event_counts.items():
+            price = self._pricing_info.get(event_name, PricingInfoItem(Decimal(), title='')).price
+            self._charging_state[event_name] = ChargingStateItem(
+                charge_count=count,
+                total_charged_amount=count * price,
+            )
 
+        # Set up charging log dataset for local development
         if not self._is_at_home and self._pricing_model == 'PAY_PER_EVENT':
             # We are not running on the Apify platform, but PPE is enabled for testing - open a dataset that
             # will contain a log of all charge calls for debugging purposes.
@@ -328,6 +336,38 @@ class ChargingManagerImplementation(ChargingManager):
     def get_max_total_charge_usd(self) -> Decimal:
         return self._max_total_charge_usd
 
+    async def _fetch_pricing_info(self) -> _FetchedPricingInfoDict:
+        """Fetch pricing information from environment variables or API."""
+        # Check if pricing info is available via environment variables
+        if self._configuration.actor_pricing_info is not None and self._configuration.charged_event_counts is not None:
+            return _FetchedPricingInfoDict(
+                pricing_info=self._configuration.actor_pricing_info,
+                charged_event_counts=self._configuration.charged_event_counts,
+                max_total_charge_usd=self._configuration.max_total_charge_usd or Decimal('inf'),
+            )
+
+        # Fall back to API call
+        if self._is_at_home:
+            if self._actor_run_id is None:
+                raise RuntimeError('Actor run ID not found even though the Actor is running on Apify')
+
+            run = run_validator.validate_python(await self._client.run(self._actor_run_id).get())
+            if run is None:
+                raise RuntimeError('Actor run not found')
+
+            return _FetchedPricingInfoDict(
+                pricing_info=run.pricing_info,
+                charged_event_counts=run.charged_event_counts or {},
+                max_total_charge_usd=run.options.max_total_charge_usd or Decimal('inf'),
+            )
+
+        # Local development without environment variables
+        return _FetchedPricingInfoDict(
+            pricing_info=None,
+            charged_event_counts={},
+            max_total_charge_usd=self._configuration.max_total_charge_usd or Decimal('inf'),
+        )
+
 
 @dataclass
 class ChargingStateItem:
@@ -339,3 +379,15 @@ class ChargingStateItem:
 class PricingInfoItem:
     price: Decimal
     title: str
+
+
+class _FetchedPricingInfoDict(TypedDict):
+    pricing_info: (
+        FreeActorPricingInfo
+        | FlatPricePerMonthActorPricingInfo
+        | PricePerDatasetItemActorPricingInfo
+        | PayPerEventActorPricingInfo
+        | None
+    )
+    charged_event_counts: dict[str, int]
+    max_total_charge_usd: Decimal

{apify-3.0.4b1 → apify-3.0.4b2}/src/apify/_configuration.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+import json
 from datetime import datetime, timedelta
 from decimal import Decimal
 from logging import getLogger
@@ -14,6 +15,12 @@ from crawlee._utils.models import timedelta_ms
 from crawlee._utils.urls import validate_http_url
 from crawlee.configuration import Configuration as CrawleeConfiguration
 
+from apify._models import (
+    FlatPricePerMonthActorPricingInfo,
+    FreeActorPricingInfo,
+    PayPerEventActorPricingInfo,
+    PricePerDatasetItemActorPricingInfo,
+)
 from apify._utils import docs_group
 
 logger = getLogger(__name__)
@@ -409,6 +416,29 @@ class Configuration(CrawleeConfiguration):
         ),
     ] = None
 
+    actor_pricing_info: Annotated[
+        FreeActorPricingInfo
+        | FlatPricePerMonthActorPricingInfo
+        | PricePerDatasetItemActorPricingInfo
+        | PayPerEventActorPricingInfo
+        | None,
+        Field(
+            alias='apify_actor_pricing_info',
+            description='JSON string with prising info of the actor',
+            discriminator='pricing_model',
+        ),
+        BeforeValidator(lambda data: json.loads(data) if isinstance(data, str) else data if data else None),
+    ] = None
+
+    charged_event_counts: Annotated[
+        dict[str, int] | None,
+        Field(
+            alias='apify_charged_actor_event_counts',
+            description='Counts of events that were charged for the actor',
+        ),
+        BeforeValidator(lambda data: json.loads(data) if isinstance(data, str) else data if data else None),
+    ] = None
+
     @model_validator(mode='after')
     def disable_browser_sandbox_on_platform(self) -> Self:
         """Disable the browser sandbox mode when running on the Apify platform.

{apify-3.0.4b1 → apify-3.0.4b2}/tests/integration/test_actor_lifecycle.py

@@ -2,6 +2,8 @@ from __future__ import annotations
 
 from typing import TYPE_CHECKING
 
+import pytest
+
 from apify import Actor
 
 if TYPE_CHECKING:
@@ -119,6 +121,7 @@ async def test_actor_fails_correctly_with_exception(
     assert run_result.status == 'FAILED'
 
 
+@pytest.mark.skip(reason='Skipped due to known instability, see issue #650.')
 async def test_actor_with_crawler_reboot(make_actor: MakeActorFunction, run_actor: RunActorFunction) -> None:
     """Test that crawler in actor works as expected after reboot.
 

{apify-3.0.4b1 → apify-3.0.4b2}/tests/integration/test_actor_log.py

@@ -12,7 +12,7 @@ if TYPE_CHECKING:
 
 # TODO: What to do with the `browserforge` output?
 # https://github.com/apify/apify-sdk-python/issues/423
-@pytest.mark.skip
+@pytest.mark.skip(reason='Skipped due to known instability, see issue #423.')
 async def test_actor_logging(
     make_actor: MakeActorFunction,
     run_actor: RunActorFunction,