Regarding less docs: okay I see now Shaz, we use the same key names as we
have in configurations, and so we defer to the documentation there.  Makes
sense.  Should we then have a separate section/attribute for run-time
adjustable config values?


On Mon, Feb 11, 2013 at 2:38 PM, Michal Mocny <[email protected]> wrote:

> I'm not sure why a single generic api would mean less docs?  We would
> still need to document each valid key, wouldn't we?  Anyway, I like the
> plugin approach suggested, especially since there will likely be very few
> configurations that need adjusting at runtime.
>
> (btw, another adjustable setting might be "set supported orientations")
>
> -Michal
>
>
> On Mon, Feb 11, 2013 at 2:31 PM, Shazron <[email protected]> wrote:
>
>> Not sure about explicit methods since we will have more docs (and separate
>> ones for the diff platforms). I personally would go with a generic API,
>> and
>> the setting to use would already be documented in the Project Settings
>> doc.
>>
>> e.g.
>>
>> navigator.app.setConfigSetting('KeyboardShrinksView', false, win, fail)
>>
>> or something to that effect. I reckon if the setting is "read-only", it
>> would call the fail callback.
>>
>>
>> On Mon, Feb 11, 2013 at 11:07 AM, Andrew Grieve <[email protected]
>> >wrote:
>>
>> > Hmm, yeah, definitely doesn't make sense for the whitelist. Not really
>> for
>> > BackupWebStorage either... If we're not saving across restarts, then
>> > BackupWebStorage won't use the updated settings anyways.
>> >
>> > Looking through the documentation, I think that most settings wouldn't
>> have
>> > a reason to change... Only ones are maybe:
>> > UIWebViewBounce
>> > EnableViewportScale
>> >
>> > And also the KeyboardShrinksView one I wanted to add.
>> >
>> > Perhaps another avenue here is to put this in an "App" plugin. Android
>> > currently has an navigator.app with methods like "clearCache",
>> > "overrideBackButton".
>> >
>> > We could explicitly have methods such as:
>> > navigator.app.setKeyboardShrinksView(true),
>> > navigator.app.setUIWebViewBounce(false)
>> >
>> > I'm not really sure if this would make more sense, or if it even matter
>> > which approach we go with :P.
>> >
>> >
>> >
>> >
>> > On Mon, Feb 11, 2013 at 11:59 AM, Shazron <[email protected]> wrote:
>> >
>> > > The only case (as Brian mentions) is not allowing the whitelist to be
>> set
>> > > at runtime.
>> > > Also - does the app have to "reset" if a setting that was read at
>> runtime
>> > > was changed? ie we probably need a "configchanged" event/notification
>> or
>> > > something. For example, BackupWebStorage in iOS, etc.
>> > >
>> > >
>> > > On Sun, Feb 10, 2013 at 11:01 AM, Andrew Grieve <[email protected]
>> > > >wrote:
>> > >
>> > > > First - I'd like to add yet another setting for iOS:
>> > > >
>> > > > KeyboardShrinksView (boolean).
>> > > >
>> > > > It applies to apps that position their elements relative to the
>> bottom
>> > of
>> > > > the webview. When the keyboard comes up, I'd like to shrink the
>> webview
>> > > > rather than shrink the viewport and have the page scrollable. This
>> is
>> > the
>> > > > default behaviour on Android, and makes a lot of sense when building
>> > apps
>> > > > as opposed to webpages.
>> > > >
>> > > >
>> > > > But... Reason for this email, is I think it'd be useful to have a
>> > > > x-platform API for changing the settings within config.xml at
>> runtime.
>> > > > Something like:
>> > > >
>> > > > cordova.setConfigValue(key, value)
>> > > >
>> > > > An example of the keyboard case:
>> > > > app.onPageChange = function(pageName) {
>> > > >   cordova.setConfigValue('KeyboardShrinksView', pageName != 'help');
>> > > > };
>> > > >
>> > > > And I suppose adding a getter makes sense:
>> > > > cordova.getConfigValues(successCallback)
>> > > >
>> > > > I'm thinking that setting these at runtime will *not* persist across
>> > app
>> > > > restarts to avoid the issue of what should happen when devs push
>> > updates
>> > > to
>> > > > their apps that change settings around.
>> > > >
>> > > > If everyone's okay with this, I'll file a bug & sub-bugs
>> per-platform.
>> > > >
>> > >
>> >
>>
>
>

Reply via email to