From: Raymond Hettinger <python@rcn.com>
Date: Sat, 14 May 2005 18:08:25 +0000 (+0000)
Subject: SF patch #1200051:  Small optimization for PyDict_Merge()
X-Git-Tag: v2.5a0~1786
X-Git-Url: http://git.ipfire.org/gitweb.cgi?a=commitdiff_plain;h=186e739d2996270c6a30d76e94193921fccfc2f5;p=thirdparty%2FPython%2Fcpython.git

SF patch #1200051:  Small optimization for PyDict_Merge()
(Contributed by Barry Warsaw and Matt Messier.)
---

diff --git a/Objects/dictobject.c b/Objects/dictobject.c
index 1be34774e3e3..aaca5aa4c1da 100644
--- a/Objects/dictobject.c
+++ b/Objects/dictobject.c
@@ -1203,6 +1203,12 @@ PyDict_Merge(PyObject *a, PyObject *b, int override)
 		if (other == mp || other->ma_used == 0)
 			/* a.update(a) or a.update({}); nothing to do */
 			return 0;
+		if (mp->ma_used == 0)
+			/* Since the target dict is empty, PyDict_GetItem()
+			 * always returns NULL.  Setting override to 1
+			 * skips the unnecessary test.
+			 */
+			override = 1;
 		/* Do one big resize at the start, rather than
 		 * incrementally resizing as we insert new items.  Expect
 		 * that there will be no (or few) overlapping keys.