-
Notifications
You must be signed in to change notification settings - Fork 24
Define interactions with contextmenu
event?
#68
Comments
I concur that we should not be firing synthentic mouse events for touch-derived |
I agree that for |
I'm assuming that by "whole touch be cancelled" you mean sending the |
Sorry, I was wrong. Chrome on Android does actually send the |
There's a dicussion currently going on for Firefox on Android similar to this issue |
This stackoverflow tries to list the type of events being sent by different browsers. |
Using https://codepen.io/webcompat/pen/gOmRKee
Firefox
Chrome
Safari
|
Part of the issue here is also that Apple/WebKit are not engaging with this spec (so are unlikely to change their implementation based on what is decided here), and for the most part other browsers are now focusing more on Pointer Events. So I'm not sure if there's any more momentum here other than somehow document divergence and warning that "there be dragons" |
Just a sidenote: The issue in FF was solved. |
Now that the spec is relatively explicit about the mouse events to expect on a tap, perhaps we should also say something about
contextmenu
?In particular, Mobile Safari and Firefox don't generate any
mouse*
events prior to a touch-derivedcontextmenu
but Chrome does. IIRC we did this in Chrome just to be consistent with the approach used forclick
- simulating the full event sequence you'd get with a real mouse. But perhaps this isn't worth it forcontextmenu
(which also isn't preceded by mouse events when triggered by the keyboard). If there's reason to believe it's sufficiently web compatible, then I'd support removing generation of themouse*
events from Chrome. @dtapuska WDYT?The text was updated successfully, but these errors were encountered: