Test that resources fetched by cross-origin CSS are not in the timeline

[yoavweiss]

Closes w3c/resource-timing#70

This is a long standing missing test. Seems like all browsers are doing the wrong thing here, which leads me to think we may need to change the spec...

[w3c-bots]

Build PENDING

Started: None
Finished: None

View more information about this build on:

• WPT PR Status
• TravisCI

[igrigorik]

This is a long standing missing test. Seems like all browsers are doing the wrong thing here, which leads me to think we may need to change the spec...

Mozilla claims to have address this in https://bugzilla.mozilla.org/show_bug.cgi?id=1180145 — is the test still failing for you? On Chrome side, see: https://bugs.chromium.org/p/chromium/issues/detail?id=532399#c32 — it seems like we still don't have agreement on w3c/ServiceWorker#719.

/cc @annevk @toddreifsteck @rniwa

[yoavweiss]

I see this test failing in the latest Firefox nightly.

I agree that if this information is exposed through Service Workers, there's no point in blocking this info from Resource Timing.

[annevk]

@jwatt and @wanderview can maybe help out here. I think our position is that we'd rather block it where possible, even if it's inconsistent.

[yoavweiss]

I've rebased and updated the test to include an @import test.

Safari Tech Preview seems to pass the test.
Firefox nightly still fails because fonts are present in the timeline. There's an open issue on that front.

[yoavweiss]

Commented on the flakiness in the related bug