For now this only contains a simple "netstrings" benchmark with just a few test cases. The interesting case is long netstrings, because in that setup, parsley is terribly slow.

To run the benchmarks, simply do python -m benchmark.runbenchmark

Example output:

parsley short and simple took 0:00:00.032414 parsley many short messages took 0:00:00.478769 parsley long messages took 0:00:04.700344 parsley many long messages took 0:00:20.034139 interp2 short and simple took 0:00:00.038897 interp2 many short messages took 0:00:00.383198 interp2 long messages took 0:00:00.040608 interp2 many long messages took 0:00:00.064451