AW: Newbie question--how to clone hg itself?

[Dirkjan Ochtman]

Kauker, Hubert <Hubert.Kauker <at> travelbasys.de> writes:
> I tried the following alternatives to compare them.
> Is there any explanation for the vast timing differences?
> Are there any other feasible variants?
> Which is the most preferable one?

Did you run each one just once? There could be statistical noise (e.g., other
users pulling at just the same time, increasing the server load).

In general, -r tip good be a bit faster than bare clone/pull, since there should
be less work to do on the server. It also does less: it only pulls the parts of
the revision graph of which tip is a descendant (so no unmerged branches).
Personally, I usually just do a bare hg clone. Takes a while for larger
repositories (like mozilla-central), but at least you get everything. If there's
a significant, reproducible difference between init/pull and clone, that is
probably a bug.

Cheers,

Dirkjan