Associcate idle periods with similar-origin window agent #1
Conversation
The idle callback lists and idle periods should be per-event loop rather than per-window. This change moves the lists and the starting of idle periods to be on a per-event loop basis. Addresses w3c#82
index.html
Outdated
| <li>An <dfn>idle callback identifier</dfn>, which is a number which MUST | ||
| initially be zero.</li> | ||
| </ul> | ||
|
|
||
| <p>In addition, each <a>event loop</a> has:</p> |
There was a problem hiding this comment.
I think it would be cleaner to use the similar-origin window agent here as the ordering is only supposed to be observable within such an agent. That's also what we do for mutation observers.
There was a problem hiding this comment.
Agreed, that makes sense. Is there an easy way to get the all the similar-origin window agents associated with an event loop (i.e., for step 11 of the event loop processing model)?
There was a problem hiding this comment.
No concrete <dfn> you can link to. But IMO something like
all similar-origin window agents that use event loop
is well-defined, given the definition at https://html.spec.whatwg.org/#window-event-loop
There was a problem hiding this comment.
Great, I've switch this over to similar-origin window agent.
rmcilroy
changed the title
| <li>A <dfn>list of idle request callbacks</dfn>. The list MUST be | ||
| initially empty and each entry in this list is identified by <a>Window</a> | ||
| and a number, the combination of which MUST be unique within the list for | ||
| the lifetime of the <a>similar-origin window agent</a>.</li> |
There was a problem hiding this comment.
The second requirement should probably fall out of how you mutate the list and not be redundantly required here (a note indicating the uniqueness might be nice though). Same below.
The idle callback lists and idle periods should be per-similar-origin window
agent, rather than per-window. This change moves the lists and the starting
of idle periods to be on a per-event loop basis.
Addresses w3c#82