Within NamedTuple, izip is faster on most cases, and equally fast on others

diff --git a/lib/sqlalchemy/util.py b/lib/sqlalchemy/util.py
index da426cbd80bf7bf1ff29755efd92992a3272febb..13f2c63c1a3a0fc93e60a94a610692fbff00a7c0 100644 (file)
--- a/lib/sqlalchemy/util.py
+++ b/lib/sqlalchemy/util.py
@@ -645,7 +645,7 @@ class NamedTuple(tuple):
     def __new__(cls, labels, vals):
         vals = list(vals)
         t = tuple.__new__(cls, vals)
-        t.__dict__ = dict(zip(labels, vals))
+        t.__dict__ = dict(itertools.izip(labels, vals))
         t._labels = labels
         return t