Fix image store ACL settings for S3 and GCS (#7614)

* Fix image store ACL settings for S3 and GCS

* Fix typing issues in ImagesPipeline ACL settings

* Fix typing issues in ImagesPipeline ACL settings

* Call parent _update_stores in ImagesPipeline

scrapy/pipelines/images.py

@@ -11,14 +11,20 @@ import hashlib
 import warnings
 from contextlib import suppress
 from io import BytesIO
-from typing import TYPE_CHECKING, Any, ClassVar
+from typing import TYPE_CHECKING, Any, ClassVar, cast
 
 from itemadapter import ItemAdapter
 
 from scrapy.exceptions import NotConfigured, ScrapyDeprecationWarning
 from scrapy.http import Request, Response
 from scrapy.http.request import NO_CALLBACK
-from scrapy.pipelines.files import FileException, FilesPipeline, _md5sum
+from scrapy.pipelines.files import (
+    FileException,
+    FilesPipeline,
+    GCSFilesStore,
+    S3FilesStore,
+    _md5sum,
+)
 from scrapy.utils.defer import ensure_awaitable
 from scrapy.utils.python import to_bytes
 
@@ -33,6 +39,7 @@ if TYPE_CHECKING:
 
     from scrapy.crawler import Crawler
     from scrapy.pipelines.media import FileInfoOrError, MediaPipeline
+    from scrapy.settings import BaseSettings
 
 
 class ImageException(FileException):
@@ -126,6 +133,20 @@ class ImagesPipeline(FilesPipeline):
     ) -> str:
         return await self.image_downloaded(response, request, info, item=item)
 
+    @classmethod
+    def _update_stores(cls, settings: BaseSettings) -> None:
+        super()._update_stores(settings)
+
+        s3store: type[S3FilesStore] = cast(
+            "type[S3FilesStore]", cls.STORE_SCHEMES["s3"]
+        )
+        s3store.POLICY = settings["IMAGES_STORE_S3_ACL"]
+
+        gcs_store: type[GCSFilesStore] = cast(
+            "type[GCSFilesStore]", cls.STORE_SCHEMES["gs"]
+        )
+        gcs_store.POLICY = settings["IMAGES_STORE_GCS_ACL"] or None
+
     async def image_downloaded(
         self,
         response: Response,

tests/test_pipeline_images.py

@@ -14,6 +14,7 @@ from itemadapter import ItemAdapter
 
 from scrapy.http import Request, Response
 from scrapy.item import Field, Item
+from scrapy.pipelines.files import GCSFilesStore, S3FilesStore
 from scrapy.pipelines.images import ImageException, ImagesPipeline
 from scrapy.utils.test import get_crawler
 
@@ -541,6 +542,44 @@ class TestImagesPipelineCustomSettings:
             expected_value = settings.get(settings_attr)
             assert getattr(pipeline_cls, pipe_attr.lower()) == expected_value
 
+    def test_images_store_s3_acl_setting_used(self, tmp_path):
+        old_policy = S3FilesStore.POLICY
+
+        try:
+            crawler = get_crawler(
+                None,
+                {
+                    "IMAGES_STORE": tmp_path,
+                    "IMAGES_STORE_S3_ACL": "public-read",
+                    "FILES_STORE_S3_ACL": "private",
+                },
+            )
+
+            ImagesPipeline.from_crawler(crawler)
+
+            assert S3FilesStore.POLICY == "public-read"
+        finally:
+            S3FilesStore.POLICY = old_policy
+
+    def test_images_store_gcs_acl_setting_used(self, tmp_path):
+        old_policy = GCSFilesStore.POLICY
+
+        try:
+            crawler = get_crawler(
+                None,
+                {
+                    "IMAGES_STORE": tmp_path,
+                    "IMAGES_STORE_GCS_ACL": "authenticatedRead",
+                    "FILES_STORE_GCS_ACL": "",
+                },
+            )
+
+            ImagesPipeline.from_crawler(crawler)
+
+            assert GCSFilesStore.POLICY == "authenticatedRead"
+        finally:
+            GCSFilesStore.POLICY = old_policy
+
 
 def _create_image(format_, *a, **kw):
     buf = io.BytesIO()