wip

libs/community/langchain_community/document_loaders/recursive_url_loader.py

@@ -199,7 +199,8 @@ class RecursiveUrlLoader(BaseLoader):
         if depth + 1 < self.max_depth:
             for link in self._extract_sub_links(text, url):
                 if link not in visited:
-                    yield from self._lazy_load_recursive(link, visited, depth=depth + 1)
+                    for doc in self._lazy_load_recursive(link, visited, depth=depth + 1):
+                        yield doc
                 if link not in visited:
                     raise ValueError
 

libs/community/tests/integration_tests/document_loaders/test_recursive_url_loader.py

@@ -32,7 +32,7 @@ def test_async_recursive_url_loader_deterministic() -> None:
 
 
 def test_sync_recursive_url_loader() -> None:
-    url = "https://docs.python.org/3.9/"
+    url = "https://python.langchain.com/"
     loader = RecursiveUrlLoader(
         url,
         extractor=lambda _: "placeholder",

libs/core/langchain_core/utils/html.py

@@ -121,4 +121,4 @@ def extract_sub_links(
                 continue
 
         results.append(path)
-    return results
+    return sorted(results)