We should be careful about adding non-standard APIs to the Web platform. If we want to make this available to every Web site, we should first standardized the API through W3C.
Adam 2009/5/6 Jeremy Orlow <[email protected]>: > The JavaScript bindings are (mostly) generated from .idl files found in > chromium/src/third_party/WebKit/WebCore/*/*.idl > chromium/src/third_party/WebKit/WebCore/dom/Node.cpp has a bunch > of dispatch*Event methods. > I don't know if the sandbox will get in your way, but doing this entirely in > WebKit would be easiest. If that's not possible, then you'll need to have > Chromium send an IPC to all the render processes > (http://dev.chromium.org/developers/design-documents/multi-process-architecture) > whenever the event should fire. > This is not a trivial task, but you'll learn a lot really quickly if you > choose to take this on. :-) > > 2009/5/6 PhistucK <[email protected]> >> >> I would like to give it a shot (at least the onidle event), if it is >> alright. >> But in order to do that, I need to understand (C++, but that is another >> story) where exactly you are laying the connection between the C++ events >> and the JavaScript events. >> (I am probably talking like a newbie (who writes it that way, huh?), but >> that is because I am a newbie, hehe, I never touched such code before, I >> know a little Visual Basic when it comes to actual programming and not web >> applications.) >> I am searching for events through the codesearch, but it is not coming >> along great. If you can point me to some example of firing events by C++ >> that calls up the JavaScript events, it will help me a lot. I could use an >> existing event as a syntax example. >> (Besides that, that means window.onidle will invoke the base::IdleTimer >> sort of directly with a TimeDelta::FromMilliseconds as a parameter (I am >> shooting for milliseconds, like setInterval\setTimeout). I guess this is >> really not secured, should I check that it is a positive number (with a >> limit?) first and if so, let it rock?) >> >> ☆PhistucK >> >> >> 2009/5/6 Aaron Boodman <[email protected]> >>> >>> Sure, it could be a bit more general. >>> >>> - a >>> >>> On Wed, May 6, 2009 at 10:58 AM, Amanda Walker <[email protected]> >>> wrote: >>> > Hmm. Rather than an idle event, how about a general way to notify on >>> > machine state changes? idle / busy, ac / battery, willsleep / >>> > didwake, etc. all seem to have the same usage pattern. Apps could >>> > then key off of the notification to start/stop workers or timers for >>> > idle processing, etc. >>> > >>> > --Amanda >>> > >>> > 2009/5/6 Aaron Boodman <[email protected]>: >>> >> >>> >> This also makes sense to me as something for the wider web platform. I >>> >> wonder if there could be awindow.onidle event. This seems generally >>> >> useful to all web apps. >>> >> >>> >> - a >>> >> >>> >> 2009/5/6 PhistucK <[email protected]>: >>> >>> I wish I had the disk space and the know how. >>> >>> ☆PhistucK >>> >>> >>> >>> >>> >>> On Wed, May 6, 2009 at 20:08, Evan Martin <[email protected]> wrote: >>> >>>> >>> >>>> Probably only if you do it. :) >>> >>>> >>> >>>> 2009/5/6 PhistucK <[email protected]>: >>> >>>> > Mmmm... so is that going to happen soon? :) >>> >>>> > >>> >>>> > 2009/5/6 Evan Martin <[email protected]> >>> >>>> >> >>> >>>> >> base/idle_timer implements this. The only implementation detail >>> >>>> >> would >>> >>>> >> be exposing it to JS. >>> >>> >>> >>> >>> >> >>> >> >> >>> >> >>> > >> >> >> > > > > > --~--~---------~--~----~------------~-------~--~----~ Chromium Developers mailing list: [email protected] View archives, change email options, or unsubscribe: http://groups.google.com/group/chromium-dev -~----------~----~----~----~------~----~------~--~---
