On Thu, 7 Mar 2013, Oleg Pudeyev wrote:

>> http://curl.haxx.se/dev/runtests.html, the test case file format is
>> documented here:
>> https://github.com/bagder/curl/blob/master/tests/FILEFORMAT
>
> I investigated that some more and the biggest question presently is
> this:
>
> It looks like the test runner launches both the test server and the test
> client (which the test runner assumes to be curl). Is it possible to launch
> just the server component?

That would be really simple to make it support, but I'd say that's a bad idea.
The point of the test infrastructure is to make sure the correct server(s)
run, run the test case and verify that the test did fine (in several different
aspects). Then do the next test - without having to restart servers that are
already running. When the last test is ran, it shuts down all the test servers
it started.

The test script runs 'curl' by default, yes. But you can tell it to run
something else instead, which is used already in a bunch of tests that run
separate binaries - see the 500, 1300 and 1500 series for example.

> While I am not opposed to this idea, I am also not very eager to implement
> it right now

I understand completely. I just wanted to let you know what exists already and
how it works. I leave it to you (all) to decide the best way forward.

-- 
  / daniel.haxx.se
_______________________________________________
http://cool.haxx.se/cgi-bin/mailman/listinfo/curl-and-python