Add type check in get_media_requests to enforce image / file urls as a list (#6949)

scrapy/pipelines/files.py

@@ -704,6 +704,10 @@ class FilesPipeline(MediaPipeline):
         self, item: Any, info: MediaPipeline.SpiderInfo
     ) -> list[Request]:
         urls = ItemAdapter(item).get(self.files_urls_field, [])
+        if not isinstance(urls, list):
+            raise TypeError(
+                f"{self.files_urls_field} must be a list of URLs, got {type(urls).__name__}. "
+            )
         return [Request(u, callback=NO_CALLBACK) for u in urls]
 
     def file_downloaded(

scrapy/pipelines/images.py

@@ -241,6 +241,10 @@ class ImagesPipeline(FilesPipeline):
         self, item: Any, info: MediaPipeline.SpiderInfo
     ) -> list[Request]:
         urls = ItemAdapter(item).get(self.images_urls_field, [])
+        if not isinstance(urls, list):
+            raise TypeError(
+                f"{self.images_urls_field} must be a list of URLs, got {type(urls).__name__}. "
+            )
         return [Request(u, callback=NO_CALLBACK) for u in urls]
 
     def item_completed(

tests/test_pipeline_files.py

@@ -261,6 +261,26 @@ class TestFilesPipeline:
         request = Request("http://example.com")
         assert file_path(request, item=item) == "full/path-to-store-file"
 
+    @pytest.mark.parametrize(
+        "bad_type",
+        [
+            "http://example.com/file.pdf",
+            ("http://example.com/file.pdf",),
+            {"url": "http://example.com/file.pdf"},
+            123,
+            None,
+        ],
+    )
+    def test_rejects_non_list_file_urls(self, tmp_path, bad_type):
+        pipeline = FilesPipeline.from_crawler(
+            get_crawler(None, {"FILES_STORE": str(tmp_path)})
+        )
+        item = ItemWithFiles()
+        item["file_urls"] = bad_type
+
+        with pytest.raises(TypeError, match="file_urls must be a list of URLs"):
+            list(pipeline.get_media_requests(item, None))
+
 
 class TestFilesPipelineFieldsMixin(ABC):
     @property

tests/test_pipeline_images.py

@@ -209,6 +209,26 @@ class TestImagesPipeline:
         assert converted.mode == "RGB"
         assert converted.getcolors() == [(10000, (205, 230, 255))]
 
+    @pytest.mark.parametrize(
+        "bad_type",
+        [
+            "http://example.com/file.jpg",
+            ("http://example.com/file.jpg",),
+            {"url": "http://example.com/file.jpg"},
+            123,
+            None,
+        ],
+    )
+    def test_rejects_non_list_image_urls(self, tmp_path, bad_type):
+        pipeline = ImagesPipeline.from_crawler(
+            get_crawler(None, {"IMAGES_STORE": str(tmp_path)})
+        )
+        item = ImagesPipelineTestItem()
+        item["image_urls"] = bad_type
+
+        with pytest.raises(TypeError, match="image_urls must be a list of URLs"):
+            list(pipeline.get_media_requests(item, None))
+
 
 class TestImagesPipelineFieldsMixin(ABC):
     @property