- From: Gregg Kellogg <gregg@greggkellogg.net>
- Date: Wed, 9 Oct 2013 16:19:13 -0700
- To: Peter F. Patel-Schneider <pfpschneider@gmail.com>
- Cc: 'RDF WG' <public-rdf-wg@w3.org>
On Oct 9, 2013, at 3:30 PM, Peter F. Patel-Schneider <pfpschneider@gmail.com> wrote: > Hi: > > Did people who ran the various test suites just do them by hand, or is there some well-known test harness that can be used? I think most developers have implemented their own test harnesses for running tests. In my case, this is done using Ruby RSpec helpers <https://github.com/ruby-rdf/rdf-turtle/blob/master/spec/suite_helper.rb> and <https://github.com/ruby-rdf/rdf-turtle/blob/master/spec/suite_spec.rb> I have another stand-alone version I use for generating EARL reports at <https://github.com/ruby-rdf/rdf-turtle/blob/master/script/tc> Basically, these load the manifest, and create an object for each test entry defined in the manifest and use these to create a call, in this case to my Turtle reader, which then typically compares the result of the Turtle parse with the result of the N-Triples parse using an Isomorphism library. In principle, doing this for any kind of test is much the same, but it makes sense to run it in an environment appropriate for the software being tested. Of course, this could be adapted to call any running any an external tool too. > peter > >
Received on Wednesday, 9 October 2013 23:19:43 UTC