Add passive option to touch event listeners. #680
Conversation
|
I like this, but:
|
|
Understood. Is this a usual warning to get: |
|
If you have jquery UI slider, then I can see that happening. Otherwise, no. |
|
This is happening when I run the grunt test. Even on Travis. Does this look better? I found a way to fallback using a quick functionality check. |
|
I don't actually know how effectively that would work, to be perfectly honest. |
|
The detection is the same used by a few libraries (like detect-it.js) so I have confidence in it, |
| @@ -741,9 +741,21 @@ const windowIsDefined = (typeof window === "object"); | |||
| this.touchmove = this._touchmove.bind(this); | |||
|
|
|||
| if (this.touchCapable) { | |||
| // Test for passive event support | |||
| var supportsPassive = false; | |||
There was a problem hiding this comment.
could you declare this variable with let instead of var to scope to nearest enclosing block?
| // Test for passive event support | ||
| var supportsPassive = false; | ||
| try { | ||
| var opts = Object.defineProperty({}, 'passive', { |
There was a problem hiding this comment.
could you declare opts with let instead of var to ensure it is only scoped within the try {} block?
|
Great PR @crhallberg! In regards to the fallback check, I saw that Paul Irish himself recommends a similar technique, so I am cool with that. I left some feedback in the code so if you could incorporate that and ping me that would be much appreciated. Thanks! |
|
I resolved the conflict with the date change and converted my variables to |
|
Merged and published to v9.7.0. Thanks again for your contribution @crhallberg! |
|
Thanks for letting me! |
By adding a passive option to the
touchstartandtouchmoveevents we can let the browser know that we do not intend to ever cancel scroll events. I found no such cases in_touchstartnor_touchmove. This will prevent any scroll lag that may be caused by this library.More on passive event listeners
grunt testin your Terminal within the bootstrap-slider repository directoryThere is also some trailing whitespace cleanup I can remove if you'd prefer.