The following code works differently in Python 2.5 than
Python 2.4:

x = [(1,), (2,), (3,)]
for y, in x:
    print y

In Python 2.4, this code produces the following:
1
2
3

In Python 2.5, this code produces the following:
(1,)
(2,)
(3,)

Interestingly enough the following code:

x = (1,)
y, = x
print y

produces the output
1

in both Python 2.4 and Python 2.5. I'm thinking this is
not intentional. :-)

Logged In: YES 
user_id=80475

Ouch.  This is bad.  The disassembly shows that the compiler
 isn't generating the unpack_sequence opcode.

Logged In: YES 
user_id=33168

Awww come on, can't we change the language just to make your
life difficult? ;-)

Thanks a lot for catching this!

Committed revision 50597.

Logged In: YES 
user_id=26911

I have the impression that the bug has only been partially
solved, i.e. in 2.5c1, singleton tuples don't get unfoled in
list comprehensions.

Python 2.5c1 (r25c1:51305, Sep  4 2006, 10:15:09)
[GCC 4.1.1 20060525 (Red Hat 4.1.1-1)] on linux2
Type "help", "copyright", "credits" or "license" for more
information.
>>> l = [(1,), (2,)]
>>> print [x for x, in l]
[(1,), (2,)]

Same example in Python 2.4.3:

Python 2.4.3 (#1, Jul 25 2006, 11:53:03)
[GCC 4.1.1 20060525 (Red Hat 4.1.1-1)] on linux2
Type "help", "copyright", "credits" or "license" for more
information.
>>> l = [(1,), (2,)]
>>> print [x for x, in l]
[1, 2]

Logged In: YES 
user_id=849994

You're right. Attached patch fixes this.

Logged In: YES 
user_id=33168

Thanks for spotting this.  Let us know if you find any more
bugs.

Well, I'm glad Georg and I came up with the same basic patch
(thanks Georg!).  I had worked it up prior to seeing his
version here.  It's also the same fix for the for loops.

Committed revision 51729. (2.6)
Committed revision 51730. (2.5)