monotone autotags

Tim Ringenbach tim.ringenbach at
Sat May 5 15:07:04 EDT 2007

On 5/5/07, Ethan Blanton <elb at> wrote:
> See monotone-devel, this has been hashed over numerous times.

Oops, I should have realized that would be the case and done so before

I don't see how arbitrary tags are particularly better than c4af5d
> (which is enough to disambiguate this revision).  They do have the
> benefit of letting you know what revisions are (possibly) children of
> others, I guess, if done as you suggest below ... but mtn automate
> ancestors will tell you this with certainty (instead of a heuristic
> broken by divergent branches, etc.), and could be wrapped up nicely to
> do so.

Maybe it's not as big of a deal as I think it is. I think arbitrary tags are
better because they contain some actual meaning, combined with some
chronology. It's a little like having a mnemonic for an opcode.

Having seen the monotone-devel threads on this, I'm not a big fan.
> There is a set of wrapper scripts out there which uses custom certs to
> accomplish this sort of thing.

Yeah, I'll read the archives. Feel free to provide me with a link to any
particular thread you think I ought to see.

For your particular scheme here, I think a good plan would be to
> simply put those revisions on the im.pidgin.marv.whitespace branch --
> I think you would achieve most of what you wanted, using existing and
> powerful mechanisms.

I am somewhat impressed that monotone has ways to archive this sort of thing
without fundamentally altering it.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <>

More information about the Devel mailing list