Re: [PerformanceTimeline] Returning Navigation Timing attributes in High Resolution Time

I'm not clear on the final recommendation here. Are we saying that we will
add all navigation timing attributes (which as part of PerformanceTiming
interface) to PerformanceEntry interface (which would mean removing some
attributes from PerformanceResourceTiming interface)?

Or would we create another object called PerformanceNavigationTiming
(similar to PerformanceResourceTiming) which extends from PerformanceEntry?

Arvind

On Fri, Apr 13, 2012 at 2:42 PM, James Simonsen <simonjam@chromium.org>wrote:

> On Wed, Apr 11, 2012 at 1:06 PM, Andy Davies <dajdavies@gmail.com> wrote:
>
>> On 11 April 2012 19:24, Jatinder Mann <jmann@microsoft.com> wrote:
>> >
>> > You can still compare Navigation Timing and Resource Timing attributes.
>> E.g., if you want the difference between the connectStart of the document
>> and the first resource, you can do the following:
>> >
>> > var resourceList = window.performance.getEntriesByType("resource");
>> > var delta = resourceList[0].connectStart -
>> (window.performance.timing.connectStart -
>> window.performance.timing.navigationStart);
>> >
>>
>> Don't know whether I'm missing something here but is resourceList[0]
>> actually the HTML page (I couldn't work out from the spec whether it's
>> included in the resource list or not)?
>>
>
> In this example, no. It'll be the first subresource. The main document
> isn't a "resource" entry, it'd be a "navigation" entry.
>
> The order is still a good question though. The spec says getEntries()
> should return in chronological order. Is it chronological by startTime? Or
> chronological by when the entries are added to the buffer? For resources,
> the latter will be sorted by finish time.
>
> James
>