Re: Proposed change to skel representation

Greg Hudson <ghudson@MIT.EDU> writes:

> > This part won't work, though. We need to maintain the ability
> > to have zero-length atoms.
>
> ... but this is bogus. ":0:" would be the zero-length atom.

Sorry, I misread the proposal. Was thinking you were wanting to get
rid of the counting eventually altogether. My fault.

> I'm proposing a very simple change here. There are already implicit
> atoms and explicit (counted) atoms. Right now implicit atoms start a
> letter and explicit atoms start with a number. What I'm proposing is:
>
> Phase 1: Implicit atoms start with a letter
> Explicit atoms start with a colon
> Atoms starting with a number are treated as explicit atoms
> for backward compatibility, when reading
>
> Phase 2: Implicit atoms start with a letter or number
> Explicit atoms start with a colon

Sure. Comprendo.

So make me -0 on the whole thing then. :-)

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Mon Jul 29 16:56:23 2002