Code coverage is pretty good without any new tests (see <http://styx.livinglogic.de/~walter/python/unicodeobject.c.gcov>) The only function uncovered is unicodeiter_len (striter_len). I've added a test for this to the new version of the patch (diff2.txt). File Added: diff2.txt
What is the point of this patch for Py2.6? There is nothing broken here. Not every sequence needs it's own __iter__. That is provided automatically. I intentionally left-out strings from when I added __iter__ to tuples, lists, and dicts. There was no compelling speed benefit as there was for the other objects.