-
Notifications
You must be signed in to change notification settings - Fork 378
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Synthetic click activation steps should state whether the event is composed or not #578
Comments
Note that the last test case in |
"initialize" is used to define I agree that we should clarify "click" however. And I think it should be composed because it's user input. We mainly considered UI Events and forgot HTML dispatches some of those too. |
Well, UI events already spec that user-initiated click is composed. The question is whether synthetic click event fired by HTMLElement.prototype.click should be composed or not. |
We also fire a synthetic click for keyboard-initiated "clicks", no? In any event, I think they should be composed. |
An UI Event which is not dispatched via In other words, if users do not have any chance to specify whether the event is composed or not via an explicit parameter, it should be considered composed. |
That's a pretty good easy-to-follow principle! |
But that only applies to non-synthetic UI events, right? |
Ah, yeah. I have found that We have a kind of contradiction. :( |
That's okay. The principle you stated in #578 (comment) doesn't concern trusted-ness at all. The only question is whether author script can specify |
How do you tell something is a UI event? But yeah, given what @smaug---- said elsewhere about his views on |
Not sure how my comment elsewhere about click() applies here. Hadn't thought about .composed. I don't have strong opinion on this. #578 (comment) doesn't sound bad, so click() would dispatch .composed set to true. |
Do we still want to flip this? What do implementations do and do we have tests? HTML still ends up setting composed to false by not talking about it. |
Ah okay, we have a test at https://github.com/w3c/web-platform-tests/blob/master/shadow-dom/event-composed.html#L76 as pointed out earlier. |
WebKit fails that test and Chrome passes it. It seems that everyone thinks it should be composed so I'll fix that. |
Tests: see shadow-dom/event-composed.html in web-platform-tests. Fixes WICG/webcomponents#578 and part of #1922.
The implementation bug in WebKit is tracked as https://bugs.webkit.org/show_bug.cgi?id=170211. |
Tests: see shadow-dom/event-composed.html in web-platform-tests. Fixes WICG/webcomponents#578 and part of #1922.
Tests: see shadow-dom/event-composed.html in web-platform-tests. Fixes WICG/webcomponents#578 and part of whatwg#1922.
To run synthetic click activation steps or more broadly the definition of firing a
click
event should clarify when if ever the event is fired withcomposed
flag set to true.Also, the definition of initializing an event should take composed as an argument.
As things stand, the specifications are extremely unclear about when & how this flag is set.
The text was updated successfully, but these errors were encountered: