- From: Rick Byers <rbyers@google.com>
- Date: Tue, 26 Feb 2013 10:10:03 -0500
- To: Arthur Barstow <art.barstow@nokia.com>
- Cc: "public-pointer-events@w3.org" <public-pointer-events@w3.org>
On Mon, Feb 25, 2013 at 6:37 PM, Arthur Barstow <art.barstow@nokia.com> wrote: > Re the testing agenda topic for the Feb 26 call, below are some of my > thoughts. > > -AB > > On 2/25/13 7:43 AM, ext Arthur Barstow wrote: >> >> 3. Testing Pointer Events v1 spec: testharness.js >> <http://www.w3.org/2008/webapps/wiki/Harness>; use hg for test repository >> (<https://dvcs.w3.org/hg/pointerevents/tests/>) ; test assertions - >> something like <http://www.w3.org/2010/webevents/wiki/TestAssertions>? > > > We already agreed to use testharness.js, the same harness used by many other > WGs to test APIs like Pointer Events. > > Regarding the test repository, I propose using a directory structure similar > to the one used by WebEvents and that would mean directories like the > following: > > .../hg/pointerevents/tests/pointer-events-v1/ > > #Submissions from WG members go in their own directory under the submissions > dir: > .../hg/pointerevents/tests/pointer-events-v1/submissions/ > .../hg/pointerevents/tests/pointer-events-v1/submissions/Google/ > .../hg/pointerevents/tests/pointer-events-v1/submissions/Microsoft/ > .../hg/pointerevents/tests/pointer-events-v1/submissions/Mozilla/ > .../hg/pointerevents/tests/pointer-events-v1/submissions/Nokia/ > > #Tests approved by the group are copied to the following directory: > .../hg/pointerevents/tests/pointer-events-v1/approved/ Sounds good. > Since some tests will require some type of user interaction (i.e. not > totally automated), it could be useful to clearly "tag" the related file as > such. For example, name the file something like > "manual_feature_blah_blah.html" or put all manual tests in a separate > directory (perhaps labeled something like "manual_tests"). Of course, > metadata within the file could also be used. Any suggestions or preferences > here? I like what Matt did in WebEvents where the manual test cases were broken up by the interaction required. I.e. one test case that involves touch one div and dragging to another, then we put all the tests we can into that file. I think minimizing the manual input required to run the suite is essential. > Lastly, Cathy created a list of Test Assertions for the Touch Events spec > and that was a valuable resource, especially with trying to scope the set of > tests needed <http://www.w3.org/2010/webevents/wiki/TestAssertions>. Do > people think something like this would be useful? Is anyone willing to lead > such an effort or contribute to it? Yes, I think it's useful. I don't have the cycles to lead it though.
Received on Tuesday, 26 February 2013 15:10:51 UTC