w3 :: #webdriver

17 Mar 2017
06:18MikeSmithAutomatedTester: thanks much
06:18MikeSmithAll those issues have now been labeled. The bulk of them are now labeled
06:19MikeSmithclarification, which includes both editorial clarifications and also the
06:19MikeSmithunderspecified cases I described earlier.
06:19MikeSmithFive of them have been labeled bug. One of those is a change to a non-
06:19MikeSmithnormative example. The rest are refinements to normative text but are
06:19MikeSmithhousekeeping changes in natureeither the result of a reviewer catching
06:19MikeSmithsome language that should have already been dropped but got overlooked, or
06:19MikeSmithotherwise catching something where its already know that no implementations
06:19MikeSmithmatch the existing language, so it needs to be tweaked to match what is
06:19MikeSmithknown to be actually implemented.
06:19MikeSmith> Our assessment of those open issues remains that none of them will require
06:19MikeSmithimplementations to change any behavior and none invalidate previous
06:19MikeSmithreviews, and so wed be confident in going ahead with CR transition and
06:19MikeSmithresolving all of those issues during the CR period.
14:34AutomatedTesterMikeSmith: I saw the email, thanks for being our advocate
14:41MikeSmithAutomatedTester: cheers
14:42MikeSmithyeah, been pasting those quotes in here just for the benefit of others on the channel who might want to know what the status is
14:42MikeSmithwhich is also why I Cced www-archive
14:43MikeSmithW3C process doesnt require the transition request e-mail discussions to be on a public list
14:43MikeSmithbut I think they should be
16:03jgrahamTests written: 1. Bugs found: 1
16:03jgraham(Well OK I actually wrote 4 tests, but the first one found a bug that caused them all to fail)
16:12AutomatedTesterjgraham: nice
16:12AutomatedTesterwhat was the test?
16:13jgrahamStarting a new session :p
16:14jgrahamApparently the response has a "capabilities" key, which is not what we were returning
16:16jimevansno, i believe geckodriver returned the capabilities wrapped in another value property? or am i misremembering?
16:17jgrahamSomething like that
17:43jgrahamWhat's the intent in the spec if I try to create a new session with something like {"capabilties": {"alwaysMatch": {"browserName": 2}}}
17:43jgrahamIn the processing capabilites section it says
17:44jgrahamIf value is not a string return an error with error code invalid argument. Otherwise, let deserialized be set to value.
17:44jgrahamAnd at the top level it says
17:44jgrahamLet capabilities be the result of trying to process capabilities with parameters as an argument.
17:45jgrahamBut it still has old text that says
17:45jgrahamIf capabilities result is an error, return error with error code session not created and the error data from capabilities result.
17:46jgrahamWhich suggests that at some point in the past it converted all errors to Session Not Created (which is a bizzare thing to do in that way vs creating an error of the actual type you want at the point where it happens)
17:46jgrahamAutomatedTester: ^
17:46jgrahamHmm, no simonstewart and AutomatedTester won't be back for a while I guess these tests aren't getting finished today
17:47AutomatedTesterjgraham: I dont have time to read it right now, will do it in a bit (sorry)
17:48* AutomatedTester is being distracted with other tasks
17:48jgrahamAutomatedTester: I wasn't expecting you to be here at all for at least an hour and a half
18 Mar 2017
No messages
Last message: 11 days and 13 hours ago