new allow_offsite parameter in OffsiteMiddleware

docs/topics/downloader-middleware.rst

@@ -797,9 +797,10 @@ OffsiteMiddleware
    :attr:`~scrapy.Spider.allowed_domains` attribute, or the
    attribute is empty, the offsite middleware will allow all requests.
 
-   If the request has the :attr:`~scrapy.Request.dont_filter` attribute
-   set, the offsite middleware will allow the request even if its domain is not
-   listed in allowed domains.
+   If the request has the :attr:`~scrapy.Request.dont_filter` attribute set to
+   ``True`` or :attr:`Request.meta` has ``allow_offsite`` set to ``True``, then
+   the OffsiteMiddleware will allow the request even if its domain is not listed
+   in allowed domains.
 
 RedirectMiddleware
 ------------------

docs/topics/request-response.rst

@@ -145,9 +145,9 @@ Request objects
     :type priority: int
 
     :param dont_filter: indicates that this request should not be filtered by
-       the scheduler. This is used when you want to perform an identical
-       request multiple times, to ignore the duplicates filter. Use it with
-       care, or you will get into crawling loops. Default to ``False``.
+       the scheduler or some middlewares. This is used when you want to perform
+       an identical request multiple times, to ignore the duplicates filter.
+       Use it with care, or you will get into crawling loops. Default to ``False``.
     :type dont_filter: bool
 
     :param errback: a function that will be called if any exception was

scrapy/downloadermiddlewares/offsite.py

@@ -28,7 +28,11 @@ def request_scheduled(self, request, spider):
         self.process_request(request, spider)
 
     def process_request(self, request, spider):
-        if request.dont_filter or self.should_follow(request, spider):
+        if (
+            request.dont_filter
+            or request.meta.get("allow_offsite")
+            or self.should_follow(request, spider)
+        ):
             return None
         domain = urlparse_cached(request).hostname
         if domain and domain not in self.domains_seen:

scrapy/spidermiddlewares/offsite.py

@@ -57,7 +57,11 @@ async def process_spider_output_async(
     def _filter(self, request: Any, spider: Spider) -> bool:
         if not isinstance(request, Request):
             return True
-        if request.dont_filter or self.should_follow(request, spider):
+        if (
+            request.dont_filter
+            or request.meta.get("allow_offsite")
+            or self.should_follow(request, spider)
+        ):
             return True
         domain = urlparse_cached(request).hostname
         if domain and domain not in self.domains_seen:

tests/test_downloadermiddleware_offsite.py

@@ -62,6 +62,37 @@ def test_process_request_dont_filter(value, filtered):
         assert mw.process_request(request, spider) is None
 
 
+@pytest.mark.parametrize(
+    ("allow_offsite", "dont_filter", "filtered"),
+    (
+        (True, UNSET, False),
+        (True, None, False),
+        (True, False, False),
+        (True, True, False),
+        (False, UNSET, True),
+        (False, None, True),
+        (False, False, True),
+        (False, True, False),
+    ),
+)
+def test_process_request_allow_offsite(allow_offsite, dont_filter, filtered):
+    crawler = get_crawler(Spider)
+    spider = crawler._create_spider(name="a", allowed_domains=["a.example"])
+    mw = OffsiteMiddleware.from_crawler(crawler)
+    mw.spider_opened(spider)
+    kwargs = {"meta": {}}
+    if allow_offsite is not UNSET:
+        kwargs["meta"]["allow_offsite"] = allow_offsite
+    if dont_filter is not UNSET:
+        kwargs["dont_filter"] = dont_filter
+    request = Request("https://b.example", **kwargs)
+    if filtered:
+        with pytest.raises(IgnoreRequest):
+            mw.process_request(request, spider)
+    else:
+        assert mw.process_request(request, spider) is None
+
+
 @pytest.mark.parametrize(
     "value",
     (

tests/test_spidermiddleware_offsite.py

@@ -29,6 +29,7 @@ def test_process_spider_output(self):
             Request("http://scrapy.org/1"),
             Request("http://sub.scrapy.org/1"),
             Request("http://offsite.tld/letmepass", dont_filter=True),
+            Request("http://offsite-2.tld/allow", meta={"allow_offsite": True}),
             Request("http://scrapy.test.org/"),
             Request("http://scrapy.test.org:8000/"),
         ]