Improve name and comment for _get_redirect_url().

python · nascheme · Jun 16, 2022 · Jun 16, 2022 · Jun 16, 2022 · Jun 16, 2022
commit f99e80b935613e802ade6f9d85e8e8c3595456a5
diff --git a/Lib/http/server.py b/Lib/http/server.py
@@ -678,7 +678,7 @@ def send_head(self):
        path = self.translate_path(self.path)
        f = None
        if os.path.isdir(path):
-            new_url = _add_trailing_slash(self.path)
+            new_url = _get_redirect_url(self.path)
            if new_url:
                # redirect browser - doing basically what apache does
                self.send_response(HTTPStatus.MOVED_PERMANENTLY)
@@ -878,10 +878,16 @@ def guess_type(self, path):
        return 'application/octet-stream'


-def _add_trailing_slash(path):
+def _get_redirect_url(path):
    """Returns URL with trailing slash on path, if required.  If not required,
    returns None.
    """
+    # It would be simpler to use urllib.parse.urlsplit() here.  However, the
+    # 'path' is not truly a URI in that it can't have a scheme or netloc.  We
+    # need to avoid parsing it incorrectly.  For example, as reported in
+    # gh-87389, a path starting with a double slash should not be treated as a
+    # relative URI.  Also, a path with a colon in the first component could
+    # also be parsed wrongly.
    path, _, fragment = path.partition('#')
    path, _, query = path.partition('?')
    if path.endswith('/'):