Mailing-List: contact dev-help@cordova.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@cordova.apache.org
Received-SPF: pass (athena.apache.org: domain of mhweiner234@gmail.com
 designates 74.125.82.178 as permitted sender)
MIME-Version: 1.0
In-Reply-To: 
 <CAEeF2TfpeOyssRWTMeFYAObchiuoGE8kqxiuWb+AsNGS2UBASQ@mail.gmail.com>
References: 
 <CABiQX1WF4BNSsw1ibY8uPE1QGRcTi1jiPeC2OekBBogd1oXLuA@mail.gmail.com>
 <CAK_TSX+7q7O-5U2jHAwQMeZ6+VKnfji_3Sta1PsrcnPTbvLtUQ@mail.gmail.com>
 <CAEeF2TfpeOyssRWTMeFYAObchiuoGE8kqxiuWb+AsNGS2UBASQ@mail.gmail.com>
From: Marc Weiner <mhweiner234@gmail.com>
Date: Mon, 26 May 2014 13:36:34 -0400
Message-ID: 
 <CANV2e=HJwbjsYScA2AWDefzUdnF7e2KqbPsDruRW6dDVtcUsTA@mail.gmail.com>
Subject: Re: Android: Deprecate WebView.sendJavascript()?
To: dev <dev@cordova.apache.org>
Content-Type: multipart/alternative; boundary=f46d043be224797ebb04fa51068a

--f46d043be224797ebb04fa51068a
Content-Type: text/plain; charset=UTF-8

+1 for deprecation sooner and removal in 4.0. Might as well break
everything at once. Doesn't sound like it's in the spirit of Cordova.


On Mon, May 26, 2014 at 1:16 PM, Michal Mocny <mmocny@chromium.org> wrote:

> Should we write a plugin that must be installed along with cordova-android
> that does the suggested "sendJavascript via PluginResults" and swap the
> implementation on native-side?
>
> -Michal
>
>
> On Mon, May 26, 2014 at 1:02 PM, Ian Clelland <iclelland@chromium.org
> >wrote:
>
> > It's probably a good candidate for removal in the 4.0 branch, if we
> really
> > don't want to support it
> >
> >
> > On Mon, May 26, 2014 at 12:59 PM, Andrew Grieve <agrieve@chromium.org>
> > wrote:
> >
> > > From: https://issues.apache.org/jira/browse/CB-6746
> > >
> > > It's a fairly common anti-pattern to use sendJavascript() to
> communicate
> > > with native side. Not sure if deprecating the call is the right move,
> but
> > > wanted to pitch the idea.
> > >
> > > There are multiple pitfalls:
> > > 1. As in the bug, it doesn't work if cordova.js isn't loaded yet
> > > 2. It's ripe for escaping bugs
> > > 3. It's less performant than using PluginResults
> > >
> > > Given that you can implement sendJavascript via PluginResults by just
> > > eval()ing the results, maybe we could just deprecate the function?
> > >
> >
>

--f46d043be224797ebb04fa51068a--