bpo-43882 - urllib.parse should sanitize urls containing ASCII newline and tabs. (GH-25595)

* issue43882 - urllib.parse should sanitize urls containing ASCII newline and tabs.

Co-authored-by: Gregory P. Smith <greg@krypto.org>
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
diff --git a/Lib/urllib/parse.py b/Lib/urllib/parse.py
index 21cae47..c11c695 100644
--- a/Lib/urllib/parse.py
+++ b/Lib/urllib/parse.py
@@ -78,6 +78,9 @@
                 '0123456789'
                 '+-.')
 
+# Unsafe bytes to be removed per WHATWG spec
+_UNSAFE_URL_BYTES_TO_REMOVE = ['\t', '\r', '\n']
+
 # XXX: Consider replacing with functools.lru_cache
 MAX_CACHE_SIZE = 20
 _parse_cache = {}
@@ -469,6 +472,9 @@ def urlsplit(url, scheme='', allow_fragments=True):
         else:
             scheme, url = url[:i].lower(), url[i+1:]
 
+    for b in _UNSAFE_URL_BYTES_TO_REMOVE:
+        url = url.replace(b, "")
+
     if url[:2] == '//':
         netloc, url = _splitnetloc(url, 2)
         if (('[' in netloc and ']' not in netloc) or