fix(url): normalize base url path without corrupting query and fragment (#45647)

api/libs/url_utils.py

@@ -3,7 +3,10 @@ from urllib.parse import urlsplit, urlunsplit
 
 def normalize_api_base_url(base_url: str) -> str:
     """Normalize a base URL to always end with /v1, avoiding double /v1 suffixes."""
-    return base_url.rstrip("/").removesuffix("/v1").rstrip("/") + "/v1"
+    base_url = base_url.strip()
+    parsed = urlsplit(base_url)
+    path = parsed.path.rstrip("/").removesuffix("/v1").rstrip("/") + "/v1"
+    return urlunsplit(parsed._replace(path=path))
 
 
 def github_blob_to_raw_url(url: str) -> str:

api/tests/unit_tests/libs/test_url_utils.py

@@ -0,0 +1,18 @@
+from libs.url_utils import normalize_api_base_url
+
+
+def test_normalize_api_base_url_appends_v1() -> None:
+    assert normalize_api_base_url("https://api.example.com") == "https://api.example.com/v1"
+    assert normalize_api_base_url("https://api.example.com/") == "https://api.example.com/v1"
+    assert normalize_api_base_url("https://api.example.com/v1") == "https://api.example.com/v1"
+    assert normalize_api_base_url("https://api.example.com/v1/") == "https://api.example.com/v1"
+
+
+def test_normalize_api_base_url_preserves_query_and_fragment() -> None:
+    assert normalize_api_base_url("https://api.example.com/v1?timeout=30") == "https://api.example.com/v1?timeout=30"
+    assert normalize_api_base_url("https://api.example.com?timeout=30") == "https://api.example.com/v1?timeout=30"
+    assert normalize_api_base_url("https://api.example.com/v1#section") == "https://api.example.com/v1#section"
+
+
+def test_normalize_api_base_url_strips_surrounding_whitespace() -> None:
+    assert normalize_api_base_url("  https://api.example.com/v1  ") == "https://api.example.com/v1"