Roll back the update_settings() changes.

scrapy/crawler.py

@@ -71,6 +71,7 @@ class Crawler:
 
         self.spidercls: Type[Spider] = spidercls
         self.settings: Settings = settings.copy()
+        self.spidercls.update_settings(self.settings)
 
         self.addons: AddonManager = AddonManager(self)
         self.addons.load_settings(self.settings)
@@ -104,10 +105,10 @@ class Crawler:
             crawler=self,
         )
 
-        self._init_reactor = init_reactor
-
+        self._init_reactor: bool = init_reactor
         self.crawling: bool = False
         self._started: bool = False
+        self.extensions: Optional[ExtensionManager] = None
         self.spider: Optional[Spider] = None
         self.engine: Optional[ExecutionEngine] = None
 
@@ -125,7 +126,6 @@ class Crawler:
 
         try:
             self.spider = self._create_spider(*args, **kwargs)
-            self.spider.update_settings(self.settings)
 
             reactor_class: str = self.settings["TWISTED_REACTOR"]
             event_loop: str = self.settings["ASYNCIO_EVENT_LOOP"]
@@ -142,7 +142,7 @@ class Crawler:
                 if is_asyncio_reactor_installed() and event_loop:
                     verify_installed_asyncio_event_loop(event_loop)
 
-            self.extensions: ExtensionManager = ExtensionManager.from_crawler(self)
+            self.extensions = ExtensionManager.from_crawler(self)
             self.settings.freeze()
 
             self.engine = self._create_engine()

scrapy/spiders/__init__.py

@@ -83,8 +83,9 @@ class Spider(object_ref):
             f"{self.__class__.__name__}.parse callback is not defined"
         )
 
-    def update_settings(self, settings: BaseSettings) -> None:
-        settings.setdict(self.custom_settings or {}, priority="spider")
+    @classmethod
+    def update_settings(cls, settings: BaseSettings) -> None:
+        settings.setdict(cls.custom_settings or {}, priority="spider")
 
     @classmethod
     def handles_request(cls, request: Request) -> bool:

tests/test_spider.py

@@ -2,13 +2,16 @@ import gzip
 import inspect
 import warnings
 from io import BytesIO
+from typing import Any
 from unittest import mock
 
 from testfixtures import LogCapture
+from twisted.internet.defer import inlineCallbacks
 from twisted.trial import unittest
 from w3lib.url import safe_url_string
 
 from scrapy import signals
+from scrapy.crawler import Crawler
 from scrapy.http import HtmlResponse, Request, Response, TextResponse, XmlResponse
 from scrapy.linkextractors import LinkExtractor
 from scrapy.settings import Settings
@@ -91,16 +94,32 @@ class SpiderTest(unittest.TestCase):
         self.spider_class.custom_settings = spider_settings
         settings = Settings(project_settings, priority="project")
 
-        spider = self.spider_class("example.com")
-        spider.update_settings(settings)
+        self.spider_class.update_settings(settings)
         self.assertEqual(settings.get("TEST1"), "spider")
         self.assertEqual(settings.get("TEST2"), "spider")
         self.assertEqual(settings.get("TEST3"), "project")
 
-        spider_instance_settings = {"TEST1": "spider_instance"}
-        spider.custom_settings = spider_instance_settings
-        spider.update_settings(settings)
-        self.assertEqual(settings.get("TEST1"), "spider_instance")
+    @inlineCallbacks
+    def test_settings_in_from_crawler(self):
+        spider_settings = {"TEST1": "spider", "TEST2": "spider"}
+        project_settings = {"TEST1": "project", "TEST3": "project"}
+
+        class TestSpider(self.spider_class):
+            name = "test"
+            custom_settings = spider_settings
+
+            @classmethod
+            def from_crawler(cls, crawler: Crawler, *args: Any, **kwargs: Any):
+                spider = super().from_crawler(crawler, *args, **kwargs)
+                spider.settings.set("TEST1", "spider_instance", priority="spider")
+                return spider
+
+        crawler = get_crawler(TestSpider, settings_dict=project_settings)
+        self.assertEqual(crawler.settings.get("TEST1"), "spider")
+        self.assertEqual(crawler.settings.get("TEST2"), "spider")
+        self.assertEqual(crawler.settings.get("TEST3"), "project")
+        yield crawler.crawl()
+        self.assertEqual(crawler.settings.get("TEST1"), "spider_instance")
 
     def test_logger(self):
         spider = self.spider_class("example.com")