Fixing issue with normalize and empty query items

test/canonicalize_url_test.py

@@ -41,6 +41,7 @@
  ("http://example.com?test", "http://example.com/?test"),
  ("http://example.com#test", "http://example.com/#test"),
  ("http://example.com?test#test", "http://example.com/?test#test"),
+ ('http://lemonde.fr/?test&test=&test=value', 'http://lemonde.fr/?test&test=&test=value')
 ]
 
 TESTS_ADVANCED = [

test/normalize_url_test.py

@@ -200,6 +200,7 @@
  ),
  ("http://lemonde.fr?%3d=value", "lemonde.fr?%3D=value"),
  ("http://lemonde.fr/default.asp", "lemonde.fr"),
+ ('http://lemonde.fr/?test&test=&test=value', 'lemonde.fr?test&test=&test=value')
 ]
 
 
@@ -232,7 +233,7 @@
  ),
  (
  "https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=4&cad=rja&uact=8&ved=2ahUKEwjp8Lih_LnmAhWQlxQKHVTmCJYQFjADegQIARAB&url=http%3A%2F%2Fwww.mon-ip.com%2F&usg=AOvVaw0sfeZJyVtUS2smoyMlJmes",
- "google.com/url?cad=rja&cd=4&esrc=s&q&rct=j&sa=t&source=web&uact=8&url=http://www.mon-ip.com/&usg=AOvVaw0sfeZJyVtUS2smoyMlJmes&ved=2ahUKEwjp8Lih_LnmAhWQlxQKHVTmCJYQFjADegQIARAB",
+ "google.com/url?cad=rja&cd=4&esrc=s&q=&rct=j&sa=t&source=web&uact=8&url=http://www.mon-ip.com/&usg=AOvVaw0sfeZJyVtUS2smoyMlJmes&ved=2ahUKEwjp8Lih_LnmAhWQlxQKHVTmCJYQFjADegQIARAB",
  {"infer_redirection": False},
  ),
  (

ural/normalize_url.py

@@ -102,11 +102,8 @@
 ]
 
 
-def coerce_empty_query_item(item):
- if item[1].strip() == "":
- return item[0], None
-
- return item
+def qsl_sort_key(item):
+ return item[0], item[1] or '', 0 if item[1] is None else 1
 
 
 def should_strip_query_item(
@@ -321,8 +318,10 @@ def normalize_url(
  None,
  )
 
+ # TODO: what to do of empty query items vs. no valued
+ # TODO: should be dedupe query items?
  qsl = [
- coerce_empty_query_item(item)
+ item
  for item in safe_qsl_iter(query)
  if not should_strip_query_item(
  item,
@@ -333,7 +332,7 @@ def normalize_url(
  ]
 
  if sort_query:
- qsl = sorted(qsl)
+ qsl = sorted(qsl, key=qsl_sort_key)
 
  # Dropping fragment if it's not routing
  if fragment and strip_fragment: