[RFC] making the null revision become real

[Denys Duchier]

John A Meinel <john at arbash-meinel.com> writes:

> I would like to get rid of revid=None being a valid revision, versus 'no
> revision supplied'. I think your arguments have some nice elegance. I'd
> like to hear what Aaron has to say, as he is one who has worked a lot
> with those identifiers.

I should have mentioned that I posted this proposal after talking with Aaron on
IRC.  Unless I misinterpreted, he seemed favorable to the general idea.  Perhaps
he disagrees on details, though.  Let's see.

> (It also would require a branch format upgrade).

I'd need a bit of help with this because I don't know what it entails

Cheers,

--Denys