The error message below should state something along the lines of "f() takes at least 1 non-keyword argument (0 given)".  Regardless, this is a regression from 2.6 which would have emitted "f() takes at least 1 argument (0 given)" which while not perfect, is a more accurate description of the problem:


D:\rft\vsl\dlr\Languages\IronPython\Tests>27
Python 2.7 (r27:82525, Jul  4 2010, 09:01:59) [MSC v.1500 32 bit (Intel)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> def f(a, b=2): pass
...
>>> f(b=2)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: f() takes at least 1 argument (1 given)
>>>

Actually CPython 2.6 emits precisely what I'd expect:
D:\rft\vsl\dlr\Languages\IronPython\Tests>26
Python 2.6.2 (r262:71605, Apr 14 2009, 22:40:02) [MSC v.1500 32 bit (Intel)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> def f(a, b=2): pass
...
>>> f(b=2)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: f() takes at least 1 non-keyword argument (0 given)

Benjamin was the last one who worked on this code (in issue 6474), adding him to nosy.

FWIW, this case is tested in PyPy: http://paste.pocoo.org/show/397732/

I think some one should just rewrite this code from scratch. Every time I fix something, it breaks something else.