incubator-callback-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Joe Bowser <bows...@gmail.com>
Subject Re: [Android] - Merging CordovaWebView into Master this week?
Date Thu, 10 May 2012 22:11:49 GMT
OK, we can do that.

BTW: Turns out the File callback issues were caused by a timing bug with
the Callback server.  When you re-start the callback server on the loadUrl
method, this can cause certain callbacks to be dropped, and for slower APIs
like the File API to not get through.  We just managed to resolve it by
putting the callback server start/stop code in the CordovaWebViewClient,
although I'm concerned as to what this will do to iFrames.  File tests are
running now.

Joe

On Thu, May 10, 2012 at 2:57 PM, Bryce Curtis <curtis.bryce@gmail.com>wrote:

> I just got back from vacation today - been disconnected from the
> internet for 2 weeks, if you can imagine that :-)
>
> If you could, please hold off merge until I've had time to go through
> it on our end.  I will work with Simon and get back with any comments.
>
> Thanks.
>
> On Wed, May 9, 2012 at 6:40 PM, Joe Bowser <bowserj@gmail.com> wrote:
> > OK, I found the issue.  It exists in 1.7, but in 1.8, since I had to move
> > where the Callback Server lives, it makes the situation worse:
> > https://issues.apache.org/jira/browse/CB-680
> >
> > It's pretty critical that we figure this one out, but so far this is the
> > main issue that is preventing this from going into the repository.
> >
> > On Wed, May 9, 2012 at 3:34 PM, Joe Bowser <bowserj@gmail.com> wrote:
> >
> >> OK, Tested with the latest from cordova-js repo.  I'm still not getting
> >> any failures on the File tests.  I'm going to test it on Android 2.3
> since
> >> all my tests so far have been Android 4.x.
> >>
> >>
> >> On Wed, May 9, 2012 at 3:23 PM, Joe Bowser <bowserj@gmail.com> wrote:
> >>
> >>> Just tested the File tests, when I manage to get them running, I don't
> >>> see any errors.  However, there does appear to be issues with Jasmine
> >>> spinning up the tests on the latest code, and I can use some help with
> it.
> >>>  I am using the cordova.android.js that we used on 1.7, and not the
> latest
> >>> cordova-js.  I'll test against that next.
> >>>
> >>> Joe
> >>>
> >>>
> >>> On Wed, May 9, 2012 at 2:48 PM, Joe Bowser <bowserj@gmail.com> wrote:
> >>>
> >>>> Hey
> >>>>
> >>>> I just reverted back to the to the other branch and I'm getting this
> as
> >>>> well on Cordova:
> >>>> D/CordovaLog( 2840): JSCallback Server Closed: Stopping callbacks.
> >>>> D/CordovaLog( 2840): file:///android_asset/www/cordova-1.7.0.js: Line
> >>>> 3560 : JSCallback Server Closed: Stopping callbacks.
> >>>> I/Web Console( 2840): JSCallback Server Closed: Stopping callbacks.
at
> >>>> file:///android_asset/www/cordova-1.7.0.js:3560
> >>>>
> >>>> I'm suspecting that the issue is the fact that we shut down and
> re-start
> >>>> the server every time we go from one page to another.  I do notice
> that
> >>>> certain Jasmine specs aren't working on the latest version of the
> code, and
> >>>> I would definitely appreciate any help figuring out why.
> >>>>
> >>>> Still, I really think we should merge this into 1.8 before we release
> an
> >>>> RC1.
> >>>>
> >>>> Joe
> >>>>
> >>>> On Wed, May 9, 2012 at 1:56 PM, Joe Bowser <bowserj@gmail.com>
wrote:
> >>>>
> >>>>> On Wed, May 9, 2012 at 1:42 PM, Simon MacDonald <
> >>>>> simon.macdonald@gmail.com> wrote:
> >>>>>
> >>>>>> Should I be using a particular cordova.android.js file when
doing my
> >>>>>> testing? I'm using the latest from the js project. When I do
I get a
> >>>>>> few errors:
> >>>>>>
> >>>>>> 1) Trying to get a persistent FS fails so most of the automatic
file
> >>>>>> tests fail.
> >>>>>>
> >>>>>
> >>>>> I'll take a look at it now.  Last time I ran the tests it passed.
> >>>>>
> >>>>>
> >>>>>> 2) Every call returns a variation of:
> >>>>>>
> >>>>>> I/Web Console( 1103): JSCallback Error: Request failed. at
> >>>>>> file:///android_asset/www/cordova.android.js:3588
> >>>>>>
> >>>>>> but it appears that he call that I've made succeeds.
> >>>>>>
> >>>>>
> >>>>> Strange!
> >>>>>
> >>>>>
> >>>>>>
> >>>>>> Other comments:
> >>>>>>
> >>>>>> 1) Plugins will break again. Every plugin I've tried (Barcode,
> >>>>>> ChildBrowser, TTS, VideoPlayer, etc.) has required changes in
the
> Java
> >>>>>> code in order to compile.
> >>>>>>
> >>>>>
> >>>>> I'm OK with this breakage if we document it. There's no way we can
> get
> >>>>> around not breaking plugins with this change, but It doesn't mean
> that we
> >>>>> shouldn't do it.
> >>>>>
> >>>>>
> >>>>>> 2) I don't like how "ctx" is a CordovaInterface but is casted
to
> >>>>>> Activity or Context depending on what is required.
> >>>>>
> >>>>>
> >>>>> ctx is NOT a CordovaInterface.  ctx is now a Context,like it was
> before
> >>>>> we started overly relying on DroidGap. CordovaInterface is something
> that
> >>>>> we recommend that people implement if they want their random
> activity to
> >>>>> work with certain plugins that require parts of the Android Life
> Cycle.
> >>>>>  This is because we can't guarantee that we're using an Activity,
a
> >>>>> DroidGap or a CordovaInterface, and we have to account for some
> people not
> >>>>> listening to us.  That's why we wrote CordovaException to be thrown
> for
> >>>>> these plugins.
> >>>>>
> >>>>>
> >>>>>
> >>>>
> >>>
> >>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message