Folk, You might be interested in my modification to Ruby to support code mutation testing. See the blog posting here to get started: <http://dataconstellation.com/blog/2012/06/22/a_new_twist_on_testing_ruby> This work was almost all done at the recent RailsCamp in Queensland, so it doesn't represent a large development effort yet.
I've started on changes to RSpec to make use of the new Twister hook, see <https://github.com/cjheath/rspec-core>. I'm currently paused trying to figure out how to modify the formatters (or whether to simply replace them) during mutation runs. I'd appreciate some of your thoughts and perhaps some pull requests to show how you think mutation testing should be done. There are limitations to mutation testing by re-loading code. I undefine any new constants and reload. That doesn't work if your code has applied any monkey patches, or if instances of classes it defined are still extant. Perhaps you have some ideas on how to make this part smoother? Anyhow, take a look, tell me what you think. Clifford Heath, Data Constellation, http://dataconstellation.com Agile Information Management and Design Skype: cjheath, Ph: (+61/0)401-533-540 _______________________________________________ rspec-users mailing list rspec-users@rubyforge.org http://rubyforge.org/mailman/listinfo/rspec-users
