[MERGE] Handling frozen set inputs in mpdiff generation when a key is missing

[Andrew Bennetts]

John Arbash Meinel wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Robert Collins wrote:
> | Just some minor tweaks found during b+tree index development.
> |
> | One interns GraphIndex strings, the other handles error raising when the
> | input to mpdiff generation was a frozenset.
> |
> | -Rob
> |
>
> I'm not 100% sure about interning the strings, but the other is
> certainly good.
>
> Have you tested the intern() with a large repository and what happens
> with memory consumption and speed?

The intern() call makes sense to me, but then I think I was the person that
originally suggested it :)

That said, it's a fairly subtle tweak purely for performance reasons, and so it
probably deserves a brief comment explaining why it's there.

-Andrew.