W3C home > Mailing lists > Public > public-web-perf@w3.org > July 2014

Re: Where are instances of PerformanceNavigationTiming created?

From: Boris Zbarsky <bzbarsky@MIT.EDU>
Date: Thu, 10 Jul 2014 20:38:03 -0400
Message-ID: <53BF31EB.9070805@mit.edu>
To: Philippe Le Hegaret <plh@w3.org>
CC: Arvind Jain <arvind@google.com>, "public-web-perf@w3.org" <public-web-perf@w3.org>
On 7/10/14, 3:28 PM, Philippe Le Hegaret wrote:
> Btw, I believe we need to be clearer on nested browsing contexts. In
> Firefox, only the top-level browsing context takes the previous document
> into account. Any nested browsing context will always have 0 for
> unloadEventStart and unloadEventEnd. In Chrome however, it will take
> into account any navigation within the nested context. See simple test
> at [1]. I didn't try with IE but it seems to me that the behavior of
> Chrome is the expected one for proper user perception expectation.
>
> Philippe
>
> [1] http://jay.w3.org/~plehegar/navigation.html

When I just tried that testcase in a Firefox nightly, I got:

   topContext.unloadEventStart= 1405038943592
   frameContext.unloadEventStart= 0

when initially loading the URL in a tab that had some other page loaded 
before it.  Then when I click the link I get

   frameContextAfterNavigation.unloadEventStart= 1405038988608

I get the same behavior in a Chrome dev build (with slightly different 
timestamps, because time is passing and all).

-Boris
Received on Friday, 11 July 2014 00:38:36 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:04:39 UTC