Hi,
I am against making Android-specific classes or methods public. This
would break the idea of an independent graphic layer between the
rendering application and the low level drawing implementation. The
current visibility modifiers have been chosen purposely.
Adding missing stuff is fine as long it fits into the concept. If a
certain method is needed, it should rather be added to the interface and
implemented in the AWT graphics adapter too. This is a small overhead
but it allows shared code between Android and desktop Java. Can you
describe what exactly is missing for your use case?
Regarding your second question:
I just read
http://stackoverflow.com/questions/7549182/android-paint-measuretext-vs-gettextbounds
to learn the difference between these two methods. The Android API is
pretty confusing here as the documentation does not mention this. Let's
try to avoid copying that design flaw into our graphics API and rather
have two clearly distinct methods or one method which takes a second
parameter. Which solution would you prefer?
Greetings,
Thilo
> To view this discussion on the web visit
>
https://groups.google.com/d/msgid/mapsforge-dev/333e970d-2e73-422a-9660-7122b5f513df%40googlegroups.com.