Name: blocking-elements
Owner: PolymerLabs
Description: Expose a stack of blocking elements https://github.com/whatwg/html/issues/897
Created: 2016-07-21 19:26:30.0
Updated: 2018-03-24 01:43:52.0
Pushed: 2018-03-24 01:43:50.0
Homepage: null
Size: 944
Language: JavaScript
GitHub Committers
User | Most Recent Commit | # Commits |
---|
Other Committers
User | Most Recent Commit | # Commits |
---|
Implementation of proposal https://github.com/whatwg/html/issues/897
The polyfill chooses a non-colliding name (
document.$blockingElements
instead ofdocument.blockingElements
) as the proposal is still work in progress and hasn't yet reached consensus on the semantics and functionality (see this discussion for more details).
document.$blockingElements
manages a stack of elements that inert the interaction outside them.
push(elem), remove(elem), pop()
document.$blockingElements.top
) and its subtree is the interactive part of the documenthas(elem)
returns if the element is a blocking elementThis polyfill will:
document.body
inert
to all the siblings of each parent, skipping the parents and the element's distributed content (if any)Use this polyfill together with the wicg-inert polyfill to disable interactions on the rest of the document. See the demo page as an example.
Another approach could be to listen for events that trigger focus change (e.g. focus, blur, keydown
) and prevent those if focus moves out of the blocking element.
Wrapping the focus requires to find all the focusable nodes within the top blocking element, eventually sort by tabindex, in order to find first and last focusable node.
This approach doesn't allow the focus to move outside the window (e.g. to the browser's url bar, dev console if opened, etc.), and is less robust when used with assistive technology (e.g. android talkback allows to move focus with swipe on screen, Apple Voiceover allows to move focus with special keyboard combinations).
Blocking Elements relies on the inert
attribute and uses Set
objects, so make sure to include their polyfills as needed.
install --save babel-polyfill
install --save wicg-inert
install --save blocking-elements
tml
ipt src="./node_modules/babel-polyfill/dist/polyfill.min.js"></script>
ipt src="./node_modules/wicg-inert/dist/inert.min.js"></script>
ipt src="./node_modules/blocking-elements/dist/blocking-elements.min.js"></script>
id="container">
utton onclick="makeBlocking(container)">make blocking</button>
utton onclick="undoBlocking(container)">undo blocking</button>
v>
ton>some button</button>
ipt>
nction makeBlocking(element) {
document.$blockingElements.push(element);
nction undoBlocking(element) {
document.$blockingElements.remove(element);
ript>
Install the dependencies with npm install
and serve the resources.
Run the tests locally by navigating to http://localhost:8080/test/
Performance is dependent on the inert
polyfill performance. Chrome recently landed the inert
attribute implementation behind a flag.
Let's compare the how long it takes to toggle the deepest x-trap-focus
inside nested x-b
of the demo page (http://localhost:8080/demo/ce.html?ce=v1)
.
document.$blockingElements
with native inert is ~15x faster than polyfilled inert ? ? ?
| with polyfilled inert (M58) | with native inert (M60) | |———-|——–| | | | | | | | | | | | |