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: CordovaWebView meeting
Date Tue, 15 May 2012 22:05:01 GMT
BTW: An activity similar to this is in the test package, but isn't
connected to the manual tests.

On Tue, May 15, 2012 at 3:04 PM, Joe Bowser <bowserj@gmail.com> wrote:

> OK, since this looks hard, here's what I'm using to test:
>
> https://github.com/infil00p/CordovaViewTest
>
> (Note: I didn't delete or ignore the jar in the libs directory)
>
> I should be able to see mobile spec start up on this one and give me the
> basic device info, however I don't get that.  I also don't see any errors
> in Logcat, which makes this even more frustrating.  At least the DroidGap
> method is still working.
>
> On another note, since the CordovaWebView method of building apps is so
> different than the DroidGap method, where would documentation go? While
> this might affect plugin developers, this branch shouldn't affect regular
> users of Cordova at all.
>
> On Tue, May 15, 2012 at 2:20 PM, Joe Bowser <bowserj@gmail.com> wrote:
>
>> If I use DroidGap, mobile-spec works fine.  If I use CordovaWebView as a
>> component, it doesn't work.  If I don't get it working by EOD today, I'll
>> just commit a test example on my github.
>>
>>
>> On Tue, May 15, 2012 at 2:11 PM, Bryce Curtis <curtis.bryce@gmail.com>wrote:
>>
>>> Not sure what you are seeing in mobile-spec, but for Automated Test,
>>> my testing shows:
>>>
>>> CordovaWebView branch: 224 specs, 4 failing
>>>  - navigator.contacts save (due to timeout)
>>>  - window.resolveLocalFileSystemURI
>>>  - FileTransfer download method (due to timeout)
>>>  - Media position
>>>
>>> Master branch:  224 specs, 4 failing
>>>  - Same 4 tests failed
>>>
>>> Galaxy Nexus 4G with 4.0.2.
>>>
>>> On Tue, May 15, 2012 at 3:44 PM, Joe Bowser <bowserj@gmail.com> wrote:
>>> > We can still remove it.  I used console.log instead, because I don't
>>> want a
>>> > weird error when using the view in the XML.  BTW: mobile-spec no longer
>>> > runs in CordovaWebView.  I'm looking at it now, but I don't know if I
>>> can
>>> > get it fixed by the EOD today.
>>> >
>>> > On Tue, May 15, 2012 at 1:19 PM, Filip Maj <fil@adobe.com> wrote:
>>> >
>>> >> OK, well, the reason we said "lets remove the exception" is because
>>> it was
>>> >> (thought to be) unused.
>>> >>
>>> >> Now it apparently is required. So let's keep it :)
>>> >>
>>> >> On 5/15/12 11:52 AM, "Joe Bowser" <bowserj@gmail.com> wrote:
>>> >>
>>> >> >Hey
>>> >> >
>>> >> >We DO need CordovaException.  The constructors were changed, and
that
>>> >> >breaks CordovaWebView, since it requires that we override the
>>> existing
>>> >> >constructors for views that are inflated from the XML, which means
>>> we need
>>> >> >to throw an exception if we're using a Context that's not a
>>> >> >CordovaInterface.
>>> >> >
>>> >> >Putting it on this thread because JIRA is down.
>>> >> >
>>> >> >Joe
>>> >> >
>>> >> >On Tue, May 15, 2012 at 10:02 AM, Filip Maj <fil@adobe.com>
wrote:
>>> >> >
>>> >> >> Joe, Simon, Bryce,
>>> >> >>
>>> >> >> Thanks for taking the time to talk about outstanding issues
with
>>> >> >> CordovaWebView. To recap, here are new todos introduced by
our
>>> meeting:
>>> >> >>
>>> >> >> - add cordovainterface to tests that are now failing due to
>>> introducing
>>> >> >> requirement of cordovainterface
>>> >> >> - remove preferencenode/set and cordovaexception classes
>>> >> >> - add any missing tests (Joe/Bryce please elaborate on this
once
>>> the
>>> >> >>issue
>>> >> >> lands)
>>> >> >> - document changes, esp. running tests (Joe, please prioritize
>>> this one
>>> >> >> once it lands)
>>> >> >> - double check fullscreen and background color preferences
>>> >> >>
>>> >> >> I will drop these as issues into JIRA under the parent
>>> CordovaWebView
>>> >> >> issue [1].
>>> >> >>
>>> >> >> Another concern Bryce and Simon had was overriding the show/hide
>>> web
>>> >> >>view
>>> >> >> functionality. I mentioned on the call that iOS already does
>>> something
>>> >> >> similar (Shaz if you can provide details here that would be
>>> >> >>appreciated).
>>> >> >> Essentially: giving power back to user on when the web view
should
>>> be
>>> >> >> unhidden (or hiding the splash screen) as there may be other
assets
>>> >> >> (jquery mobile, for example) that add to the web view
>>> initialization
>>> >> >>time.
>>> >> >> I think this is a cross-platform concern as well. Coming up
with a
>>> >> >> solution for this issue on Android is a first step, but I think
we
>>> >> >>should
>>> >> >> also loop back with the iOS implementation, and possibly research
>>> >> >>how/what
>>> >> >> other platforms offer for this functionality, and abstract
away the
>>> >> >> platform specifics for our users.
>>> >> >>
>>> >> >> Finally, from Adobe's side, we would love to see this land
in 1.8.
>>> >> >> Realistically for this to happen, to me this means it must
get
>>> merged in
>>> >> >> before we tag the RC for 1.8.
>>> >> >>
>>> >> >> Please add anything I've missed.
>>> >> >>
>>> >> >> Cheers,
>>> >> >> Fil
>>> >> >>
>>> >> >> [1] https://issues.apache.org/jira/browse/CB-190
>>> >> >>
>>> >> >>
>>> >> >> On 5/15/12 8:56 AM, "Filip Maj" <fil@adobe.com> wrote:
>>> >> >>
>>> >> >> >I'm reviewing the changes between current master and the
>>> CordovaWebView
>>> >> >> >branch on github [1]:
>>> >> >> >
>>> >> >> >I did a grep for "TODO" in the java code and found a few
comments
>>> >> >>dropped
>>> >> >> >in by Bryce:
>>> >> >> >
>>> >> >> >framework/src/org/apache/cordova/App.java:51:         
  else if
>>> >> >> >(action.equals("show")) { // TODO @bc - Not in master branch.
>>>  When
>>> >> >>should
>>> >> >> >this be called?
>>> >> >> >framework/src/org/apache/cordova/DroidGap.java:254:   
    //
>>> TODO @bc
>>> >> >>-
>>> >> >> >What about fullscreen?
>>> >> >> >framework/src/org/apache/cordova/DroidGap.java:340:   
    //
>>> TODO @bc
>>> >> >>-
>>> >> >> >background color doesn't work
>>> >> >> >
>>> >> >> >
>>> >> >> >I can talk about the top one (the App plugin's "show" action).
>>> >> >> >
>>> >> >> >Essentially, it was to fix a specific issue, during the
JavaScript
>>> >> >>boot up
>>> >> >> >for Android, there was a call to prompt() that would tell
the
>>> native
>>> >> >> >framework to show the web view. I have instead changed
that to
>>> route
>>> >> >>via
>>> >> >> >exec(), and thus added an additional action to the App
plugin to
>>> >> >> >encapsulate that functionality. A commit in the js [2]
introduced
>>> >> >>this. We
>>> >> >> >just have to get the latest cordova-js dropped in and that
should
>>> work
>>> >> >> >out. The reason for this is procedure, we want to keep
reusing
>>> the same
>>> >> >> >mechanism to talk to native (I.e. Exec()) instead of using
exec
>>> for
>>> >> >> >certain cases and the underlying mechanism for exec() in
others.
>>> Also,
>>> >> >>for
>>> >> >> >emulation tools such as Ripple, changing this to an exec()
call
>>> gets
>>> >> >>rid
>>> >> >> >of an annoying prompt if you are emulating Android.
>>> >> >> >
>>> >> >> >Other than that I think everything is OK. Pull in latest
>>> cordova-js
>>> >> >>and we
>>> >> >> >should be golden.
>>> >> >> >
>>> >> >> >Bryce/Simon are there any other questions other than the
three
>>> I've
>>> >> >> >identified above?
>>> >> >> >
>>> >> >> >[1]
>>> >> >> >
>>> >> >>
>>> >> >>
>>> >>
>>> https://github.com/apache/incubator-cordova-android/compare/master...Cord
>>> >> >>o
>>> >> >> >v
>>> >> >> >aWebView
>>> >> >> >[2]
>>> >> >> >
>>> >> >>
>>> >> >>
>>> >>
>>> http://git-wip-us.apache.org/repos/asf?p=incubator-cordova-js.git;a=blobd
>>> >> >>i
>>> >> >> >f
>>> >> >>
>>> >>
>>> >>>f;f=lib/android/platform.js;h=afab8e11899eb1f560a630539d470535e24f0914;h
>>> >> >>>p=
>>> >> >> >7
>>> >> >>
>>> >>
>>> >>>c38206f2f7d1be755236859df23e6b57a61c7e7;hb=9a080cc2704171a4169739b2c94ca
>>> >> >>>55
>>> >> >> >4
>>> >> >> >27b92f93;hpb=71d806e61e5240ccf9f6299500741b68ea9c5d3d
>>> >> >> >
>>> >> >> >On 5/14/12 9:58 PM, "Joe Bowser" <bowserj@gmail.com>
wrote:
>>> >> >> >
>>> >> >> >>I think we have a different idea of what merge nightmare
is,
>>> since it
>>> >> >> >>didn't look that bad.  That being said, there's a lot
of code
>>> that was
>>> >> >> >>changed to the old pre-cordova-js way of doing things.
 Are you
>>> sure
>>> >> >>your
>>> >> >> >>repo is up to date with the master?  I've been doing
merge
>>> commits on
>>> >> >>the
>>> >> >> >>CordovaWebView branch the whole time, so there shouldn't
be a
>>> huge
>>> >> >> >>difference between master and the branch.  Also, I'm
seeing
>>> stuff in
>>> >> >>your
>>> >> >> >>tree that I haven't seen since before the cordovajs
change.
>>>  Fil, can
>>> >> >>you
>>> >> >> >>look at the diffs and make sure that I'm not just seeing
things?
>>> >> >> >>
>>> >> >> >>Finally, can we please decide on a format? Are we using
tabs or
>>> >> >>spaces in
>>> >> >> >>Java files, and how many spaces.  I thought we settled
this a
>>> while
>>> >> >>ago
>>> >> >> >>with spaces, but I'm not sure, and these commits look
weird.  I'm
>>> >> >>certain
>>> >> >> >>that I'm doing something different than other people
again with
>>> my
>>> >> >>"Death
>>> >> >> >>to Tabs" kick.
>>> >> >> >>
>>> >> >> >>On Mon, May 14, 2012 at 9:41 PM, Bryce Curtis
>>> >> >> >><curtis.bryce@gmail.com>wrote:
>>> >> >> >>
>>> >> >> >>> I've pushed changes to CordovaWebView branch.
 Hopefully I
>>> didn't
>>> >> >>mess
>>> >> >> >>> things up too bad, but merge was messy.
>>> >> >> >>>
>>> >> >> >>> On Mon, May 14, 2012 at 3:41 PM, Simon MacDonald
>>> >> >> >>> <simon.macdonald@gmail.com> wrote:
>>> >> >> >>> > Just wanted to make sure. I'll do the call
before heading
>>> over the
>>> >> >> >>> > conference.
>>> >> >> >>> >
>>> >> >> >>> > Simon Mac Donald
>>> >> >> >>> > http://hi.im/simonmacdonald
>>> >> >> >>> >
>>> >> >> >>> >
>>> >> >> >>> > On Mon, May 14, 2012 at 4:37 PM, Joe Bowser
<
>>> bowserj@gmail.com>
>>> >> >> >>>wrote:
>>> >> >> >>> >
>>> >> >> >>> >> I'm cool with a 9 AM call this once.
I don't really sleep
>>> anymore
>>> >> >> >>> anyway.
>>> >> >> >>> >> :P
>>> >> >> >>> >>
>>> >> >> >>> >> On Mon, May 14, 2012 at 1:33 PM, Simon
MacDonald
>>> >> >> >>> >> <simon.macdonald@gmail.com>wrote:
>>> >> >> >>> >>
>>> >> >> >>> >> > Just double checking, we are doing
the call at 9am pacific
>>> >> >>time on
>>> >> >> >>> >> Tuesday?
>>> >> >> >>> >> >
>>> >> >> >>> >> > Simon Mac Donald
>>> >> >> >>> >> > http://hi.im/simonmacdonald
>>> >> >> >>> >> >
>>> >> >> >>> >> >
>>> >> >> >>> >> > On Mon, May 14, 2012 at 1:09 PM,
Joe Bowser <
>>> bowserj@gmail.com
>>> >> >
>>> >> >> >>> wrote:
>>> >> >> >>> >> >
>>> >> >> >>> >> > > Hey
>>> >> >> >>> >> > >
>>> >> >> >>> >> > > Here's the number for the call:
(855) 870-5454
>>> >> >> >>> >> > >
>>> >> >> >>> >> > > Conference Code is: 5328695080
>>> >> >> >>> >> > >
>>> >> >> >>> >> > > If you want, we can set up
a Connect room and we can
>>> walk
>>> >> >> >>>through
>>> >> >> >>> all
>>> >> >> >>> >> the
>>> >> >> >>> >> > > code, or we can just talk about
it from a High Level.  I
>>> >> >>really
>>> >> >> >>> want to
>>> >> >> >>> >> > see
>>> >> >> >>> >> > > this get in 1.8.  That being
said, any commits that are
>>> done
>>> >> >>to
>>> >> >> >>>the
>>> >> >> >>> >> > branch
>>> >> >> >>> >> > > now are welcome, which is why
I put it in on the Apache
>>> >> >> >>>repository,
>>> >> >> >>> and
>>> >> >> >>> >> > we
>>> >> >> >>> >> > > should probably use the list
to talk about it as well
>>> as the
>>> >> >> >>>call.
>>> >> >> >>> >> > >
>>> >> >> >>> >> > > Joe
>>> >> >> >>> >> > >
>>> >> >> >>> >> > >
>>> >> >> >>> >> > > On Mon, May 14, 2012 at 9:43
AM, Filip Maj <
>>> fil@adobe.com>
>>> >> >> >>>wrote:
>>> >> >> >>> >> > >
>>> >> >> >>> >> > > > Sounds good to me. Joe
will setup call details.
>>> >> >> >>> >> > > >
>>> >> >> >>> >> > > > On 5/14/12 9:02 AM, "Bryce
Curtis" <
>>> curtis.bryce@gmail.com
>>> >> >
>>> >> >> >>> wrote:
>>> >> >> >>> >> > > >
>>> >> >> >>> >> > > > >How about tomorrow
at 9pt/11ct?  Simon is traveling
>>> today.
>>> >> >> >>> >> > > > >
>>> >> >> >>> >> > > > >On Mon, May 14, 2012
at 10:56 AM, Filip Maj
>>> >> >><fil@adobe.com>
>>> >> >> >>> wrote:
>>> >> >> >>> >> > > > >> Let's have a
call ASAP to discuss the issues.
>>> >> >> >>> >> > > > >>
>>> >> >> >>> >> > > > >> On 5/13/12 9:09
PM, "Bryce Curtis"
>>> >> >><curtis.bryce@gmail.com
>>> >> >> >
>>> >> >> >>> >> wrote:
>>> >> >> >>> >> > > > >>
>>> >> >> >>> >> > > > >>>Joe,
>>> >> >> >>> >> > > > >>>
>>> >> >> >>> >> > > > >>>I've gone
through the CordovaWebView branch with
>>> Simon
>>> >> >>and
>>> >> >> >>>have
>>> >> >> >>> >> some
>>> >> >> >>> >> > > > >>>comments and
updates based upon your branch for
>>> your
>>> >> >> >>>review.
>>> >> >> >>> >>  When I
>>> >> >> >>> >> > > > >>>upload, I
will let you know the link to look at.
>>>  If you
>>> >> >> >>>want,
>>> >> >> >>> we
>>> >> >> >>> >> > can
>>> >> >> >>> >> > > > >>>have a call
on Tue to go over.  There's still some
>>> >> >>things
>>> >> >> >>>to be
>>> >> >> >>> >> > worked
>>> >> >> >>> >> > > > >>>out, so I
don't want to rush merging into master
>>> until
>>> >> >>the
>>> >> >> >>>same
>>> >> >> >>> >> > level
>>> >> >> >>> >> > > > >>>of function/robustness
has been demonstrated.  We
>>> can
>>> >> >>make
>>> >> >> >>>an
>>> >> >> >>> >> > > > >>>assessment
when you get back.
>>> >> >> >>> >> > > > >>>
>>> >> >> >>> >> > > > >>>Bryce
>>> >> >> >>> >> > > > >>>
>>> >> >> >>> >> > > > >>>On Fri, May
11, 2012 at 12:55 PM, Joe Bowser <
>>> >> >> >>> bowserj@gmail.com>
>>> >> >> >>> >> > > wrote:
>>> >> >> >>> >> > > > >>>> Hey
>>> >> >> >>> >> > > > >>>>
>>> >> >> >>> >> > > > >>>> This
will probably make things trickier as far as
>>> >> >> >>> CordovaWebView
>>> >> >> >>> >> > is
>>> >> >> >>> >> > > > >>>> concerned,
but I will be away from my keyboard
>>> from
>>> >> >>May
>>> >> >> >>>17th
>>> >> >> >>> to
>>> >> >> >>> >> > May
>>> >> >> >>> >> > > > >>>>23rd.
>>> >> >> >>> >> > > > >>>>  I'll
still be reachable via e-mail, but I won't
>>> be
>>> >> >>able
>>> >> >> >>>to
>>> >> >> >>> do
>>> >> >> >>> >> > much
>>> >> >> >>> >> > > > >>>>more
>>> >> >> >>> >> > > > >>>> than
that for that time.
>>> >> >> >>> >> > > > >>>>
>>> >> >> >>> >> > > > >>>> Joe
>>> >> >> >>> >> > > > >>
>>> >> >> >>> >> > > >
>>> >> >> >>> >> > > >
>>> >> >> >>> >> > >
>>> >> >> >>> >> >
>>> >> >> >>> >>
>>> >> >> >>>
>>> >> >> >
>>> >> >>
>>> >> >>
>>> >>
>>> >>
>>>
>>
>>
>

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