Add a variant test for script scheduling #47018
Conversation
| <div id="log"></div> | ||
| <p>This test is similar to <a href="128.html">128.html</a>, but here the parent script is non-parser-inserted all along.</p> | ||
| <script> | ||
| var t = async_test(); |
There was a problem hiding this comment.
We don't need an async test here right? Can we just do this all in a sync test?
| var t = async_test(); | ||
| t.step(function() { | ||
| log("inline script #1"); | ||
| var script = document.createElement("script"); |
There was a problem hiding this comment.
Something other than var here and below?
| @@ -0,0 +1,29 @@ | |||
| <!DOCTYPE html> | |||
There was a problem hiding this comment.
Hmm, you might want to see https://github.com/web-platform-tests/wpt/blob/master/dom/nodes/insertion-removing-steps/Node-appendChild-script-in-script.tentative.html. It's related to whatwg/dom#1261 and whatwg/html#10188, and once the latter lands, I think we'll need a different set of expectations corresponding with the simpler spec.
In fact, I'm wondering if we don't even need the test in this PR once we make the existing one I linked to above, non-tentative. Thoughts?
There was a problem hiding this comment.
The one you linked to is basically the same as 128, right, in that it relies on the parser-inserted script s1 being unflagged as parser-inserted? So I think this variant still has some small but nonzero value compared to either of them.
There was a problem hiding this comment.
So I think this variant still has some small but nonzero value compared to either of them.
Yep, that may well address the point about whether we should remove this test or not. But I still think it could be good to update the expectations to match what we're pursuing in HTML regarding script scheduling + children changed events.
It took me a while to figure out why 128 worked at all (the script I thought was parser-inserted wasn't anymore at the point of test), and by that time I'd already added 151.
Note that the order of execution of scripts 2 and 3 is in flux (ref whatwg/dom#1261), but it should at least be consistent between the 128 and 151.