Set spider attribute on log records

scrapy_playwright/handler.py

@@ -39,7 +39,8 @@
 PlaywrightHandler = TypeVar("PlaywrightHandler", bound="ScrapyPlaywrightDownloadHandler")
 
 
-logger = logging.getLogger("scrapy-playwright")
+LOGGER_NAME = "scrapy-playwright"
+logger = logging.getLogger(LOGGER_NAME)
 
 
 DEFAULT_BROWSER_TYPE = "chromium"
@@ -60,6 +61,7 @@ def __init__(self, crawler: Crawler) -> None:
         super().__init__(settings=settings, crawler=crawler)
         verify_installed_reactor("twisted.internet.asyncioreactor.AsyncioSelectorReactor")
         crawler.signals.connect(self._engine_started, signals.engine_started)
+        crawler.signals.connect(self._spider_opened, signals.spider_opened)
         self.stats = crawler.stats
 
         # browser
@@ -109,6 +111,17 @@ def _engine_started(self) -> Deferred:
         """Launch the browser. Use the engine_started signal as it supports returning deferreds."""
         return deferred_from_coro(self._launch())
 
+    def _spider_opened(self, spider: Spider) -> None:
+        old_factory = logging.getLogRecordFactory()
+
+        def record_factory(name: str, *args, **kwargs):
+            record = old_factory(name, *args, **kwargs)
+            if name == LOGGER_NAME:
+                record.spider = spider
+            return record
+
+        logging.setLogRecordFactory(record_factory)
+
     async def _launch(self) -> None:
         """Launch Playwright manager and configured startup context(s)."""
         logger.info("Starting download handler")

tests/test_playwright_requests.py

@@ -295,7 +295,7 @@ def should_abort_request_sync(request):
                     assert handler.stats.get_value(f"{req_prefix}/aborted") == 3
 
     @pytest.mark.asyncio
-    async def test_page_initialization_ok(self, caplog):
+    async def test_page_initialization_ok(self):
         async def init_page(page, request):
             await page.set_extra_http_headers({"Extra-Header": "Qwerty"})
 
@@ -360,6 +360,21 @@ async def test_redirect(self):
             server.urljoin("/redirect"),
         ]
 
+    @pytest.mark.asyncio
+    async def test_logging_record_spider(self, caplog):
+        """Make sure at least one log record has the spider as an attribute
+        (records sent before opening the spider will not have it).
+        """
+        caplog.set_level(logging.INFO)
+        spider = Spider("spider_name")
+        async with make_handler({"PLAYWRIGHT_BROWSER_TYPE": self.browser_type}) as handler:
+            handler._spider_opened(spider)
+            with MockServer() as server:
+                req = Request(url=server.urljoin("/index.html"), meta={"playwright": True})
+                await handler._download_request(req, spider)
+
+        assert any(getattr(rec, "spider", None) is spider for rec in caplog.records)
+
 
 class TestCaseChromium(MixinTestCase):
     browser_type = "chromium"