MediaPipeline: log media_to_download errors before stripping them (#5068)

scrapy/pipelines/media.py

@@ -118,14 +118,14 @@ class MediaPipeline:
         info.downloading.add(fp)
         dfd = mustbe_deferred(self.media_to_download, request, info, item=item)
         dfd.addCallback(self._check_media_to_download, request, info, item=item)
+        dfd.addErrback(self._log_exception)
         dfd.addBoth(self._cache_result_and_execute_waiters, fp, info)
-        dfd.addErrback(
-            lambda f: logger.error(
-                f.value, exc_info=failure_to_exc_info(f), extra={"spider": info.spider}
-            )
-        )
         return dfd.addBoth(lambda _: wad)  # it must return wad at last
 
+    def _log_exception(self, result):
+        logger.exception(result)
+        return result
+
     def _make_compatible(self):
         """Make overridable methods of MediaPipeline and subclasses backwards compatible"""
         methods = [

tests/test_pipeline_crawl.py

@@ -8,6 +8,7 @@ from w3lib.url import add_or_replace_parameter
 
 from scrapy import signals
 from scrapy.crawler import CrawlerRunner
+from scrapy.utils.misc import load_object
 from tests.mockserver import MockServer
 from tests.spiders import SimpleSpider
 
@@ -192,6 +193,29 @@ class FileDownloadCrawlTestCase(TestCase):
             crawler.stats.get_value("downloader/response_status_count/302"), 3
         )
 
+    @defer.inlineCallbacks
+    def test_download_media_file_path_error(self):
+        cls = load_object(self.pipeline_class)
+
+        class ExceptionRaisingMediaPipeline(cls):
+            def file_path(self, request, response=None, info=None, *, item=None):
+                return 1 / 0
+
+        settings = {
+            **self.settings,
+            "ITEM_PIPELINES": {ExceptionRaisingMediaPipeline: 1},
+        }
+        runner = CrawlerRunner(settings)
+        crawler = self._create_crawler(MediaDownloadSpider, runner=runner)
+        with LogCapture() as log:
+            yield crawler.crawl(
+                self.mockserver.url("/files/images/"),
+                media_key=self.media_key,
+                media_urls_key=self.media_urls_key,
+                mockserver=self.mockserver,
+            )
+        self.assertIn("ZeroDivisionError", str(log))
+
 
 try:
     from PIL import Image  # noqa: imported just to check for the import error