sep base url and loaded url in sub link extraction (#10895)

libs/langchain/langchain/document_loaders/recursive_url_loader.py

@@ -145,7 +145,8 @@ class RecursiveUrlLoader(BaseLoader):
         # Store the visited links and recursively visit the children
         sub_links = extract_sub_links(
             response.text,
-            self.url,
+            url,
+            base_url=self.url,
             pattern=self.link_regex,
             prevent_outside=self.prevent_outside,
         )
@@ -224,7 +225,8 @@ class RecursiveUrlLoader(BaseLoader):
         if depth < self.max_depth - 1:
             sub_links = extract_sub_links(
                 text,
-                self.url,
+                url,
+                base_url=self.url,
                 pattern=self.link_regex,
                 prevent_outside=self.prevent_outside,
             )

libs/langchain/langchain/utils/html.py

@@ -1,5 +1,5 @@
 import re
-from typing import List, Union
+from typing import List, Optional, Union
 from urllib.parse import urljoin, urlparse
 
 PREFIXES_TO_IGNORE = ("javascript:", "mailto:", "#")
@@ -37,16 +37,18 @@ def find_all_links(
 
 def extract_sub_links(
     raw_html: str,
-    base_url: str,
+    url: str,
     *,
+    base_url: Optional[str] = None,
     pattern: Union[str, re.Pattern, None] = None,
     prevent_outside: bool = True,
 ) -> List[str]:
     """Extract all links from a raw html string and convert into absolute paths.
 
     Args:
-        raw_html: original html
-        base_url: the base url of the html
+        raw_html: original html.
+        url: the url of the html.
+        base_url: the base url to check for outside links against.
         pattern: Regex to use for extracting links from raw html.
         prevent_outside: If True, ignore external links which are not children
             of the base url.
@@ -54,6 +56,7 @@ def extract_sub_links(
     Returns:
         List[str]: sub links
     """
+    base_url = base_url if base_url is not None else url
     all_links = find_all_links(raw_html, pattern=pattern)
     absolute_paths = set()
     for link in all_links:
@@ -62,9 +65,9 @@ def extract_sub_links(
             absolute_paths.add(link)
         # Some may have omitted the protocol like //to/path
         elif link.startswith("//"):
-            absolute_paths.add(f"{urlparse(base_url).scheme}:{link}")
+            absolute_paths.add(f"{urlparse(url).scheme}:{link}")
         else:
-            absolute_paths.add(urljoin(base_url, link))
+            absolute_paths.add(urljoin(url, link))
     if prevent_outside:
         return [p for p in absolute_paths if p.startswith(base_url)]
     return list(absolute_paths)

libs/langchain/tests/unit_tests/utils/test_html.py

@@ -102,3 +102,28 @@ def test_extract_sub_links() -> None:
         ]
     )
     assert actual == expected
+
+
+def test_extract_sub_links_base() -> None:
+    html = (
+        '<a href="https://foobar.com">one</a>'
+        '<a href="http://baz.net">two</a>'
+        '<a href="//foobar.com/hello">three</a>'
+        '<a href="/how/are/you/doing">four</a>'
+        '<a href="alexis.html"</a>'
+    )
+
+    expected = sorted(
+        [
+            "https://foobar.com",
+            "https://foobar.com/hello",
+            "https://foobar.com/how/are/you/doing",
+            "https://foobar.com/hello/alexis.html",
+        ]
+    )
+    actual = sorted(
+        extract_sub_links(
+            html, "https://foobar.com/hello/bill.html", base_url="https://foobar.com"
+        )
+    )
+    assert actual == expected