On Fri, Apr 18, 2014 at 11:13 AM, Rick Waldron <waldron.rick@gmail.com>wrote: > On Fri, Apr 18, 2014 at 1:58 PM, Adam Klein <adamk@chromium.org> wrote: > >> On Fri, Apr 18, 2014 at 10:08 AM, Rick Waldron <waldron.rick@gmail.com>wrote: >>> >>> On Thu, Apr 17, 2014 at 8:14 PM, Boris Zbarsky <bzbarsky@mit.edu> wrote: >>> >>>> On 4/17/14 6:35 PM, Rick Waldron wrote: >>>> >>>>> Will this actually matter? Will `w` ever have a _different_ identity >>>>> such that it will no longer be usable get the value from the map? >>>>> >>>> >>>> In Chrome, yes. Did you not try my testcase at >>>> http://jsfiddle.net/zVqVX/1/ in Chrome? >>>> >>> >>> I don't know where the disconnect was, but yes I see this result >>> now—thanks for your patience. >>> >>> @Mark Miller, have you seen this example? Run it in Chrome with >>> experimental JS on—the object apparently violates its own identity. >>> >> >> To be clear, I'm actively working on the Map/Set implementations, >> > > Excellent, thanks for the heads up. > > >> and they're flag-guarded, so changing Chrome's behavior is very much on >> the table (and it's why I sent this message in the first place). >> > > Yes, as my previous message explicitly stated: "with experimental JS on". > > To recap then, the behavior will match the expectation that a WindowProxy > preserves identity across browsering contexts; is this still correct? > Yes, that's what Toon originally suggested to me (prompting me to start this thread), and I think I've been convinced that that's the right behavior.
Received on Friday, 18 April 2014 18:21:53 UTC