[TIP] Revising subunit's wire protocol - feedback solicited

Robert Collins robertc at robertcollins.net
Thu Feb 14 03:35:37 PST 2013


Apologies for the cross post - please follow either on the blog post
or on the subunit-dev list, to prevent too much fragmentation in the
discussion.

http://rbtcollins.wordpress.com/2013/02/14/time-to-revise-the-subunit-protocol/

If you are not interested in the gory details of supporting concurrent
test execution on multiple heterogeneous machines in awkward
configurations with scewed clocks..... skip this. Otherwise, I'd love
your input :). (Actually there are many simpler cases that this post
is relevant to.. - but solving the nasty cases is on my hit list right
now).

I had intended to also include a subtests thing like Python is growing
upstream, which will also serve adapted TAP streams, and trivial C
test runners that don't have individually addressable test cases
within a single binary, but I forgot that until after I posted the
blog post. I think a sub test id field is probably enough, and
document that that cannot be passed when select tests to run.

Cheers,
Rob

-- 
Robert Collins <rbtcollins at hp.com>
Distinguished Technologist
HP Cloud Services



More information about the testing-in-python mailing list