[webkit-reviews] review denied: [Bug 70868] Stylesheets not created by the parser should not block scripts : [Attachment 112460] patch v1, resolved compilation error

[bugzilla-daemon at webkit.org]

Tony Gentilcore <tonyg at chromium.org> has denied Johnny(Jianning) Ding
<jnd at chromium.org>'s request for review:
Bug 70868: Stylesheets not created by the parser should not block scripts
https://bugs.webkit.org/show_bug.cgi?id=70868

Attachment 112460: patch v1, resolved compilation error
https://bugs.webkit.org/attachment.cgi?id=112460&action=review

------- Additional Comments from Tony Gentilcore <tonyg at chromium.org>
View in context: https://bugs.webkit.org/attachment.cgi?id=112460&action=review


r- for lack of a test.

As far as I can tell this is a good and safe optimization. Antti should
probably take a look too.

> Source/WebCore/ChangeLog:9
> +	   which means that that script shouldn't be dependent on style being
immediately recalculated.

It is worth linking to the spec.

> Source/WebCore/ChangeLog:11
> +	   No new tests. (OOPS!)

It is worth adding a test to demonstrate this behavior so that we don't regress
upon it in the future. This is probably easiest done with an http test that has
a slow stylesheet. See http/tests/misc/script-defer-after-slow-stylesheet.html
for an example.

Once you have the test, another key question is how other browsers behave. If
FF or IE has the same behavior of not blocking, it gives us a lot more
confidence in the spec and our interpretation of it. If we are the first to
introduce this behavior, we should proceed with more prudence.