GSOC project updates

Tomaz Canabrava tcanabrava at kde.org
Mon Jul 28 10:55:46 PDT 2014


On Mon, Jul 28, 2014 at 2:48 PM, Lubomir I. Ivanov <neolit123 at gmail.com> wrote:
> On 28 July 2014 20:33, Tomaz Canabrava <tcanabrava at kde.org> wrote:
>> On Mon, Jul 28, 2014 at 2:20 PM, Lubomir I. Ivanov <neolit123 at gmail.com> wrote:
>>>
>>> dunno, my tests were based on Qt5.2 for Win32.
>>> RotationAnimation on a declarative image, 15% CPU on a dual core with OpenGL.
>>> rotating with QGraphicsScene / C++ ~0% CPU with or without OpenGL.
>>
>> Then I hope they fix this bug, as they are selling QML as the big thing in Qt.
>> We can do a touch ui based on QGraphicsView, but it will be so much
>> easier in QML ( mostly because now there's a widget's QML package )
>>
>>
>
> me and Dirk are asking the same question in a way, do we even have to
> create a new UI? can we just style the existing UI to be usable on
> android with #ifdefs and things like setStyleSheet(...). also only
> show one Subsurface view at a time?
>
> from what i've gathered both you and Anton mention that the main issue
> is the tiny UI elements.
> that aside, aren't all parts like the profile and dive list already
> working on Android - do they receive touch events, does input text
> fields open the Android keyboard etc..?

well, There are some parts of the interface that will not ever be good
even if we style it:

The Dive Info part should be vertical, two fields side by side is too
big for phones ( but works for tablets )

the dive list has too many fields.

"do we even have to create a new UI" -> to get the most of the platform, yes.
at least three UI's:

- Tablet
- Phone
- Desktop

how hard it's going to be?
Right now, on Subsurface 4.2, a tiny bit hard since we have some
'desktop' assumptions ( like the existence of a mainwindow, that other
classes calls ).

I plan to clean all that for 4.3 and further separate the code from
interface ( thing that I'm doing for quite a long time already ) in a
way that will be easy to integrate new interfaces to the existing
core.

that's going to be too much of work? I sincerely hope not. :)

> lubomir
> --


More information about the subsurface mailing list