<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40"><head><meta http-equiv=Content-Type content="text/html; charset=iso-8859-1"><meta name=Generator content="Microsoft Word 15 (filtered medium)"><style></style></head><body lang=FR link=blue vlink=purple><div class=WordSection1><div><div>> I think it would be easier to evaluate if someone wrote down a proposal for the js API :)<o:p></o:p><o:p> </o:p>Yes, I’m going to work of that.<o:p></o:p>In a meantime I’ve drawn a global schema architecture: <a href="http://wiki.sugarlabs.org/go/File:SugarWebApp_Architecture.svg">http://wiki.sugarlabs.org/go/File:SugarWebApp_Architecture.svg</a> <o:p></o:p>It’s more the “step 1” architecture but it’s easy to adapt for the “step 2”. BTW, feel free to comment/upgrade.<o:p></o:p><o:p> </o:p></div><div> > Well, it's OS specific, but generally quite hacky. Slightly better than console-message though :) > Maybe the iOS approach could work for us. They load a gap_exec url inside an iframe, when the WebView notices that, it evaluates some kind of fetchMessage() method and handle the result. >Or if the iframe thing doesn't work you could poll but ugggh. They are actually doing that for some version of android, for the opposite direction though.<o:p></o:p>Oooch, really ? Very hacky indeed. But yes, we could try the iFrame hack.<o:p></o:p> >In WebkitGtk2 you can actually get the result of execute_script (though probably only from C code, you would need to wrap that up in sugar-toolkit or something).<o:p></o:p>Nice. But it require WebkitGtk2.<o:p></o:p> Lionel.<o:p></o:p></div><div> <o:p></o:p></div></div></div></body></html>