Transport w/ extra tests

[Martin Pool]

On Mon, 2005-07-18 at 16:16 -0500, John A Meinel wrote:

> I also started playing around with pipelining http access using
> effbot.org.http_manager
> But in my testing, it wasn't much faster (1m20 instead of 1m50).

I've identified a couple of problems that are slowing it down.  One
problem is that every socket.connect call in Python does a DNS
resolution. 

My http client based on that is working reasonably well, at least for
the one server I test it against.  It's still a bit messy.

> And now, for whatever reason, my Mac decided that "%.3f" is not a valid
> format (the 3 is not a valid format character). (I think I have system
> corruption somewhere).

That's strange.

> >>PS> It would be nice if we had a way to have "bzr selftest" only run
> >>certain tests. So that when you are doing some development, you can only
> >>run the tests you care about, until you get them working, and then go
> >>back and run everything.
> >
> > Yes, I agree.  It'd also be nice to have a way to skip tests -- I think
> > by just raising a TestSkipped exception, which would be reported
> > differently to a regular test failure.
>
> I was actually thinking more along the lines of:
> 
> bzr selftest RsyncTransportTest
> 
> Which would skip all tests that did not match the string RsyncTransportTest.

I was unclear there.  Yes, I agree about that usage.  What I meant was
that also tests should be able to say "I don't want to run", for example
because they can't be tested on Windows.  Having them just pass without
doing the tests makes me uncomfortable.

-- 
Martin