- From: James Graham <james@hoppipolla.co.uk>
- Date: Fri, 11 Oct 2013 17:31:28 +0100
- To: public-test-infra@w3.org
On 11/10/13 17:20, David Burns wrote: > I honestly don't see that there should be any difference between 1 and 3. I think as a practical matter that's kind of horrible. If I am working on a testsuite I might reasonably want to try it out on a bunch of browsers. Having to start the test environment seperately for each browser and track which browser instances are test-compatible and which are not doesn't seem like much fun. Plus, depending on the exact design, it might not work at all on browsers without explicit support. That's a problem if it's IE certainly, but also if it's Servo or some research project based on an existing browser engine with just enough changes that launching the browser directly won't work. Also I would like to avoid, as far as possible, requiring something like virtualenv + pip to run the tests. Importing a large part of mozbase, and equivalents - if indeed they exist - without this is going to be ugly at best.
Received on Friday, 11 October 2013 16:31:51 UTC