Why is it that *dbm modules don't provide an iterator? (Languagedesign question)

[Akira Kitada]

Hi,

I was wondering why *dbm modules in Python do not give us an iterable interface?
Take a look at an example below

"""
# Python 2.6

import gdbm
d = gdbm.open("spam.db", "n")
d["key1"] = "ham"
d["key2"] = "spam"

for k in d:

.... print k
....
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
TypeError: 'gdbm.gdbm' object is not iterable
"""

The loop has to be:
""".... print k
.... k = d.nextkey(k)
key2
key1
"""

I would like to know the background of this design decision.
Some research on this led me to this ticket:

http://bugs.python.org/issue662923

It looks I'm not the only one who wondered on this.

Thanks in advance.