mozilla :: #jsapi

9 Sep 2017
00:01tcampbellmccr8: I put the link in the bug to the other tests in the same suite. You should probably give them a look over too
00:01shutcampbell: i am happy to see you have taken up the scopes mantle
00:01tcampbellshu: I didn't realize the email from mccr8 was a trap
00:02shuthat's usually how those things start
00:02tcampbellindeed
00:18jorendorff:D
00:35John-Galtmccr8, tcampbell: So, it looks like doing the Right Thing for Cu.import when an NSVO is on the scope chain actually breaks some frame scripts that expect us to do the Wrong Thing.
00:35tcampbellExample?
00:35John-GaltI just ran into one that threw an error because it was accidentally depending on the Services.jsm import from another frame script.
00:36John-GaltIt used to be that Cu.import without a target would define the properties on the global rather than the NSVO. Now it defines them on the NSVO. I think.
00:36* tcampbell thinking
00:37John-GaltWhich I think is the right behavior. We just may run across a few bugs where people are depending on the wrong behavior until this stabilizes.
00:49tcampbellJohn-Galt: I'm off for the evening. I can take a peek at it at some point tomorrow, otherwise Monday we can go through the issues
00:50John-Galttcampbell: I think we should probably be able to ignore it in the end. I just thought I should mention it in case any issues come up. Have a good weekend :)
00:52tcampbellI'd like to know all mismatches even if we can fix the JS instead of the engine.
00:54* John-Galt nods
15:49mrgigglesthe mozilla-inbound tree is now closed (bustage from bug 1360334 )
15:49firebotBug https://bugzil.la/1360334 is not accessible
16:06mrgigglesthe mozilla-inbound tree is now open
17:38tcampbellmccr8: ping
17:39tcampbell(well, if you are around, otherwise we can do asynchronously on bugzilla)
10 Sep 2017
No messages
   
Last message: 11 days and 10 hours ago