- From: Stefan Håkansson LK <stefan.lk.hakansson@ericsson.com>
- Date: Mon, 21 Oct 2013 14:10:04 +0000
- To: Anne van Kesteren <annevk@annevk.nl>
- CC: "public-media-capture@w3.org" <public-media-capture@w3.org>
On 2013-10-21 15:01, Anne van Kesteren wrote: > On Mon, Oct 21, 2013 at 8:28 AM, Stefan Håkansson LK > <stefan.lk.hakansson@ericsson.com> wrote: >> Section 4.2 "MediaStream", add paragraph: >> "When a user agent (e.g. as result of postMessage() being called with a >> MediaStream as argument) is required to obtain a structured clone of a >> MediaStream, follow the algorithm outlined in >> [http://www.w3.org/html/wg/drafts/html/master/infrastructure.html#safe-passing-of-structured-data]; >> when the 'internal structured cloning algorithm' step is reached, carry >> out the same processing as if the 'clone()' method had been called for >> that MediaStream, and let streamClone be output." >> ================================= >> >> Does this look reasonable? > > We'll move this to JavaScript eventually, see > https://github.com/dslomov-chromium/ecmascript-structured-clone for > work in progress, and at that point you'll need to make it clear it > ends up in a different global as well (realm in ES6-speak). So you > can't just use clone() as that only works same-global Thanks for feedback. I see from the github doc that InternalStructuredClone will take a "targetRealm" input, and I assume you mean that we need have that covered in the InternalStructuredClone algorithm for MediaStream. Is that right? If so I guess we need to device a InternalStructuredClone algorithm that takes a targetRealm input. Do you see "targetRealm" coming soon, or would something like I proposed be sufficient for now? (Anne, sorry for spamming you, by mistake a missed the public- list in my first response) > >
Received on Monday, 21 October 2013 14:10:29 UTC