cordova-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Lisa Seacat DeLuca <ldel...@us.ibm.com>
Subject Re: cordova-android LinearLayout vs. RelativeLayout
Date Thu, 15 May 2014 18:05:05 GMT
Joe might be responding to my comment within the JIRA issue about 
potentially leaving the layout up to the user as a config setting.  And I 
agree that would make it tougher as far as implementing and testing all of 
the possible scenarios.  I say we start with the easy solution of 
switching to the RelativeLayout which, as Andrew pointed out, would have a 
similar experience for developers as iOS has as far as overlaying 
components. 

Yes, let's put it on the list for 4.0 even if it's only to discuss at an 
upcoming hangout for pros/cons of the switch.


Lisa


Lisa Seacat DeLuca
Mobile Engineer | t: +415.787.4589 | ldeluca@apache.org | | 
ldeluca@us.ibm.com | lisaseacat.com | | 



From:   Andrew Grieve <agrieve@chromium.org>
To:     dev <dev@cordova.apache.org>
Date:   05/15/2014 12:12 PM
Subject:        Re: cordova-android LinearLayout vs. RelativeLayout
Sent by:        agrieve@google.com



Is it that you think LinearLayout is the best default Layout? Or is it 
that
you don't want to make a non-backwards-compatible change?

Be happy to stick this on the "should do for 4.0" list (it's not urgent by
any means). But if you're arguing for LinearLayout being a better option,
then I'd like to know why.

On Wed, May 14, 2014 at 11:53 AM, Joe Bowser <bowserj@gmail.com> wrote:

> I disagree.  iOS is not Android and doesn't have to deal with the 
Android
> life cycle or intents.   I don't think supporting extending 
CordovaActivity
> is a good idea, especially considering other libraries that the view may 
be
> used with will also be extending the Activity.


> If people want to do Android development and not Cordova development,
> which sounds like the case here, they should have to handle their own
> intents.
>
> Also, if we're counting this as an ATI change, we can't do it until a 
major
> version.
>  On May 14, 2014 7:36 AM, "Andrew Grieve" <agrieve@chromium.org> wrote:
>
> > I think Lisa's reasons make sense.
> >
> > True, you can make your own Activity, but there's a lot of value in 
not
> > requiring this. iOS has one level of indirection for the UIWebView (it
> > lives in a generic UIView), and that allows the exact use-case that is
> > being discussed here.
> >
> >
> > On Wed, May 14, 2014 at 12:43 AM, Joe Bowser <bowserj@gmail.com> 
wrote:
> >
> > > Ok, here's my opinion on LinearLayout vs RelativeLayout and why I
> closed
> > > the bug as "Won't Fix"
> > >
> > > 1. We went with LinearLayout because it was the easiest at the time.
>  We
> > > experimented with FrameLayout as well, and decided to stick with
> > > LinearLayout because it was the easiest to setup programmatically.
> > > 2. CordovaActivity is not intended to be extended or modified in any
> way.
> > >  If people need a custom activity, they should create their own and
> > > implement CordovaInterface and use the CordovaWebView. That's why we
> > > created the CordovaWebView.
> > >
> > > The fact is that we actually use a class called
> > > LinearLayoutSoftKeyboardDetect, which at one point detected when the
> > > keyboard was visible by measuring the screen to see if the WebView 
and
> > the
> > > Layout containing it resized, but since the keyboard's behaviour
> changed
> > in
> > > 3.x, this has been broken for quite a long time and simply doesn't
> work.
> > > That being said, changing this would be non-trivial, and I don't see
> any
> > > benefit in doing another refactor.
> > >
> > >
> > >
> > > On Tue, May 13, 2014 at 1:19 PM, Lisa Seacat DeLuca <
> ldeluca@us.ibm.com
> > > >wrote:
> > >
> > > > I opened a JIRA issue on the android default webview layout use of 
a
> > > > LinearLayout.  Chatting with our product team that uses Cordova
> they've
> > > had
> > > > a number of customer requests to use a RelativeLayout.  I wanted 
to
> get
> > > > some input from people to see if there is a reason we decided to 
go
> > with
> > > > the LinearLayout instead of a RelativeLayout in case there is a 
good
> > > reason
> > > > and someone goes and fixes this issue.  :)
> > > >
> > > > https://issues.apache.org/jira/browse/CB-6681
> > > >
> > > >
> > > > Lisa
> > > >
> > > > Lisa Seacat DeLuca
> > > > Mobile Engineer | t: +415.787.4589 | *ldeluca@apache.org*<
> > > ldeluca@apache.org>| |
> > > > *ldeluca@us.ibm.com* <ldeluca@us.ibm.com> | *lisaseacat.com*<
> > > http://www.lisaseacat.com/>| [image:
> > > > follow @LisaSeacat on twitter] <http://www.twitter.com/LisaSeacat
>|
> > > [image:
> > > > follow Lisa Seacat DeLuca on linkedin]<
> > > http://www.linkedin.com/in/lisaseacat>
> > > >
> > >
> >
>


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