Why id() for same Unicode string literals gives different result?

Question

Why Unicode string literals show different id's ? I was hoping the same behavior as that of String literals.

>>> p = 'abcd'
>>> q = 'abcd'
>>> id(p) == id(q)
True
>>> p = u'abcd'
>>> q = u'abcd'
>>> id(p) == id(q)
False

Please provide some pointers on this.

Answer 1

For the same reason two dicts with the same contents would have different ids: they are distinct objects. I suspect that the non-Unicode string literals being the same object is something of an optimization.