Changes

*External*Context Impl: Remove unused 'lastContext' field, incl. their makeCurrent() and release() override.

Semantics of 'lastContext' are plain wrong, since release() override does claim the previous 'lastContext'
is current at the end - however, it wasn't technically made current.

MemoryObject: Use GlueGen's new HashUtil

GlueGen commit 1a4514accc8f61ab7ff5fe8c82d22a5ef356c865

Bug 665: Allow re-association of GLContext/GLEventListener to a GLDrawable (Part 4)

Note:

  - GLEventListenerState preservs the GLAutoDrawable state,
    i.e. GLContext, all GLEventListener and the GLAnimatorControl association.

  - GLEventListenerState may be utilized to move the state from a dying GLAutoDrawable,
    to be moved to a new created GLAutoDrawable at a later time.

  - GLEventListenerState will be made public soon.

+++

Exessive unit tests cover the new feature, tested manually on GNU/Linux/X11 and OSX(Java6/Java7).

+++

- GLAutoDrawable
  - Change 'setContext(..)' to allow the destruction of the unbound old context:
      'setContext(GLContext newCtx)' -> 'setContext(GLContext newCtx, boolean destroyPrevCtx)'

  - Implementations: Properly implement 'setRealized(..)' incl. obeying threading constraints if exists.
    Method is being utilized at least for GLEventListenerState.moveTo(..)
    to unrealize and realize the drawable resources.

+++

Fix propagation of GLContext/GLDrawable association change (Bottom -> Top):

  GLDrawableImpl.associateContext
      GLContextImpl.associateDrawable
          GLContextImpl.makeCurrent
          GLContextImpl.destroy
          GLContext.setGLDrawable
              ...
              GLDrawableHelper.switchContext
                  GLAutoDrawble.setContext

associateDrawable(..)/associateContext(..) unifies and hence:
    - GLContextImpl.contextRealized() (removed)
    - GLDrawableImpl.contextRealized() (removed)
    - GLDrawableImpl.associateContext(..) (merged)
    - MacOSXCGLContext.drawableChangedNotify(..) (removed)

+++

- EGLUpstreamSurfaceHook.evalUpstreamSurface() validates the surface's device for reusage,
  which is valid in case of GLEventListenerState.moveTo(..)

- MacOSXCGLContext.NSOpenGLImpl: pixelFormat replaces NSOpenGLLayerPfmt and has simplified lifecycle [create..destroy],
  while native NSOpenGLLayer code only holds the reference until released.

Bug 665: Allow re-association of GLContext/GLEventListener to a GLDrawable (Part 5)

- GLAutoDrawableBase:

  - Add 'setPreserveGLStateAtDestroy(..)' to preserve the GLEventListenerState at destroy() operation,
    and impl. details pullGLEventListenerState()/pushGLEventListenerState().
    pullGLEventListenerState() is called automatic at destroyImplInLock(),
    where pushGLEventListenerState() has to be called after drawable realization
    instead of context creation.

  - Note/TODO: Method will become public in GLAutoDrawable in general!

- NEWT/GLWindow:
  - Use GLEventListenerState preservation for reparenting case w/ destruction,
    i.e. keep GLContext/GLEventListener alive while reparenting in recreation mode.

    Scenario: NewtCanvasAWT Child <-> Top on OSX w/ CALayer

Minor Changes: EGLDisplayUtil: Add stack trace in DEBUG mode for opened DPYs; JAWTWindow: Add JAWT info in toString()

GLEventListenerState: Moved to public package 'com.jogamp.opengl.util'

Fix TestParenting01NEWT: Since recreational reparenting resets the fps-counter, wait for a few frames!

OSX/CALayer: Revise CALayer 'RunOnMainThread' utilization, avoiding deadlocks

RunOnMainThread(waitUntilDone:=true,..) can deadlock the main-thread if called from AWT-EDT,
since the main-thread may call back to AWT-EDT while injecting a new main-thread task.

This patch revises all RunOnMainThread CALayer usage, resulting in only one required left:
  - OSXUtil.AddCASublayer() w/ waitUntilDone:=false

Hence the CALayer code has no more potential to deadlock main-thread/AWT-EDT.

OSXUtil.AddCASublayer() must be performed on main-thread, otherwise the
CALayer attachment will fail - no visible rendering result.

+++

Note: A good trigger to test this deadlock is to magnify/zoom
      the OSX desktop (click background + ctrl-mouse_wheel)
      before running some unit tests.

      TestGLCanvasAWTActionDeadlock01AWT and TestAddRemove02GLWindowNewtCanvasAWT
      also have the potential to trigger the mentioned deadlock.

OSX/CALayer: Simplify FixCALayerLayout()/layoutSurfaceLayer() call, no more need for explicit call

- OffscreenLayerSurface.layoutSurfaceLayer() removed, no more required

- JAWTWindow adds a ComponentListener, which issues FixCALayerLayout() at resized, moved and shown.

- MyNSOpenGLLayer no more requires fix*Size() methods

- MyNSOpenGLLayer::setDedicatedSize() need no explicit CATransaction, performed by caller.

Fix NEWT WindowImpl reparent-recreate w/ GLEventListenerState: Bug introduced w/ commit e2506d7663b752f00f0a98f793ebad52e65bd1e3

In case a reparent action takes place w/ recreate,
only preserve the GLEventListenerState if the window is valid and will become visible again (wasVisible).

Also add proper DEBUG log prefix to GLEventListenerState.

Remodel OSX/CALayer Threading (commit 896e8b021b39e9415040a57a1d540d7d24b02db1): Run on main-thread w/o blocking ; Misc Changes

Commit 896e8b021b39e9415040a57a1d540d7d24b02db1 moved all native CALayer calls to the current thread
to avoid deadlocks.

Even though this seemed to be fine at least resource GC (release/dealloc calls) were issued
very late in time, probably due to multithreading synchronization of JAWT and/or OSX API.

Example: Our 'TestAddRemove01GLCanvasSwingAWT' test didn't freed CALayer resources incl. GL ctx
         when destroying the objects (AWT Frame, GLCanvas, ..), leading to resource starvation .. eventually.

Remedy is a compromise of behavior before commit 896e8b021b39e9415040a57a1d540d7d24b02db1
and that commit, i.e. to run CALayer lifecycle methods on main-thread, but do not block!

The careful part within MacOSXCGLContext.associateDrawable(..) performs the following block on main-thread:
  - lock the context
  - create NSOpenGLLayer (incl. it's own shared GL context and the DisplayLink)
  - attach NSOpenGLLayer to root CALayer
  - unlock the context

Due to the GL ctx locking, this async offthread operation is safe within our course of operations.

Details:
  - NSOpenGLContext
    - Context and CVDisplayLink creation at init
    - Call [ctx update] if texture/frame size changed
    - 'waitUntilRenderSignal' uses default TO value if given TO is 0 to avoid deadlocks

+++

Misc Changes:
  - Fix object type detection: isMemberOfClass -> isKindOfClass
    - OSXUtil_isNSView0
      OSXUtil_isNSWindow0,
      CGL_isNSOpenGLPixelBuffer

  - MacOSXCGLDrawable/MacOSXPbufferCGLDrawable: remove getNSViewHandle() method.
    MacOSXCGLContext uses common code to detect nature of the drawable handle.

  - MacOSXCGLContext/CALayer: Use safe screenVSyncTimeout values, never 0 to avoid deadlock!

  - JAWTWindow.invalidate: Call detachSurfaceLayer() if not done yet

Adapt to GlueGen commit b1eb7ca6b9d7dec7ff62c1f1e8ef0a0545724d2f: Function- RunnableTask adds PrintStream 'exceptionOut' argument in ctor.

NativeWindow OSXUtil RunOnMainThread: Use daemon attachment and do not detach; Add RunLater0(..)

OSX CGL: Don't issue [NSOpenGLContext clearDrawable] for [NSOpenGLContext setView: view] which breaks pbuffer; Add [NSOpenGLContext clearDrawable].

MyNSOpenGLContext::dealloc: Avoid 'invalid context'

OSX/CALayer: OSX/CALayer Threading Part3 - Run CALayer ops in a streaming design on main-thread w/o [infinitive] blocking

History:
  Part1   commit 896e8b021b39e9415040a57a1d540d7d24b02db1 (Run CALayer Ops on current thread to avoid blocking)
  Part2   commit 28c6472335b924080d638b33a28f8f4eedb459b1 (Run CALayer Ops on main-thread w/o blocking)

Dependency:
  GlueGen commit 4becdfa125b07ff969d6540e1112735b53cd15eb (Fix RecursiveLockImpl* Timeout corner case)

Part2 misses essential locking of the OpenGL context (and it's surface upfront) while creating the
NSOpenGLLayer instance. The latter instantiates a OpenGL context shared w/ JOGL's, hence it cannot be locked.

Encapsulating NSOpenGLLayer creation/attachment and it's detachment/release in sub-classes
AttachNSOpenGLLayer and DetachNSOpenGLLayer, where instances will be streamed on main-thread.

Both tasks are triggered at associateDrawable(boolean bound).

The mentioned GL context locking requires disturbs the 'streaming' design considerably in AttachNSOpenGLLayer.
It is solved by attempt to acquire the recursive lock of the surface and the context via 'tryLock(maxwait)'
w/ screen-vSync-period/2. If the locks could not be acquired completly, the AttachNSOpenGLLayer instance
will be re-queued to the main-thread for later execution.

Before DetachNSOpenGLLayer is being streamed, it is validated whether AttachNSOpenGLLayer did run.
A recursive situation does happen w/ resizing an offscreen pbuffer drawable! Hence extra care is being taken.

OSX/NEWT: Following CALayer streaming design, i.e. issue NSWindow/NSView Ops on main-thread w/o blocking; NEWT/WindowImpl: Volatile multithreaded mutable values

Similar to commits:
  28c6472335b924080d638b33a28f8f4eedb459b1
  f354fb204d8973453c538dda78a2c82c87be61dc

main-thread operations cannot block main-thread.

Luckily we are able to create the NSWindow and NSView instance uninitialized (deferred) on the current thread,
while issuing their initialization on the main-thread w/o blocking.

Further more a size glitch is fixed, which didn't take the title bar into account.

+++

NEWT/WindowImpl: Volatile multithreaded mutable values

Since position, size and other attributes might get changes off-thread, these fields needs to be volatile.

GLJPanel: No need to utilize AWT Component's default ImageObserver code - pixel data is ready when invoked.

MacOSXCGLContext: Remove Debug output

OSX/NEWT: NSWindow/NSView Ops on main-thread w/o blocking - Part2

Continues commit 81cbcdc8469143587b2044661dd613c798ae02ba

Perform on main-thread invocation from Java, allowing to issue
visibleChanged(..) after creation/visible calls.

This fixes the 'Visibility not reached ..' regressions.

Handles indexed PNGs

Rename PNG test files and add RGBA variations (To be used by TestPNGTextureFromFileNEWT, etc)

Newt PNG test file locations

TestPNGTextureFromFileNEWT: Use new PNG test file locations, add RGBA tests and use alpha in drawable

PNGImage: Cleanup code and reuse scanlineRGBA buffer in indexed mode. Add DEBUG property 'jogl.debug.PNGImage'

TestGearsES2NEWT/TestGearsES2NewtCanvasAWT: Add -noanim option to manually validate Bug 649

OSX/NEWT: Fix native window parenting freeze, invisible/orderOut0, Position bugs w/ parenting

Fix native window parenting freeze:

  - Pull out setJavaWindowObject(..) of changeContentView(..) to be called seperately,
    add param for changeContentView(..) to enable/disable setJavaWindowObject()

  - initWindow0(..):
    - Call changeContentView(..) w/o setJavaWindowObject()
    - setJavaWindowObject(..) at end of initialization

Fix native window parenting orderOut0:
  If parent window is invisible or no parent used call orderOut(..),
  otherwise call orderBack().

Fix updatePosition(..): positionChanged(..) - Position bug w/ parenting
  - AWT parent passed 0/0
  - call positionChanged(..) w/ client-pos instead of screen-pos

Fix getLocationOnScreenImpl(..) - Position bug w/ parenting
  - Position < 0/0 is valid!

Misc:
  - setWindowClientTopLeftPointAndSize0(..), setWindowClientTopLeftPoint0(..):
    Add 'display' param, deciding whether area should be display (invalidated)

MacOSXCGLContext: NPE in DEBUG mode while releaseNSOpenGLLayer on main-thread, drawable maybe null already

OSX/NEWT: Fix visible -> false when closing and reparenting window. This lead to non functional recreational reparenting.

OSX recreational reparenting moves the saved GLEventListenerState at destroy
to the new dawable/surface, which must be valid.
The flaky visible state caused seemingly random reparenting failures.

- WindowImpl.ReparentActionRecreate.run() set 'visible:=true', which circumvented OSX to wait for actual realization.
- OSX WindowDriver.closeNative(): Issue visibleChanged(true, false); ASAP

NEWTEvent: Add isConsumed() and setConsumed(boolean) methods to simply usage, using the existing consumedTag attachment for compatibility and efficiency.

NEWT/Android: Fix BACK button implementation, use different KeyCode mappings and allowing native action to be suppressed.

- Don't trust soft-kbd visibility state, but perform invisible action. If the latter
  was successful - soft-kbd was visible before.

- Map BACK to VK_KEYBOARD_INVISIBLE and propagate it,
  if soft-kbd was visible before.
  No native default action is performed.

- Map BACK to VK_ESCAPE event and propagate it,
  if soft-kbd was invisible _and_ an activity was registered via registerActivity(Activity),
  i.e. by NewtBaseActivity.
  Otherwise proceed w/ default action (-> activity.finish()).

- If the KeyListener consumed the [EVENT_KEY_RELEASED, VK_ESCAPE] event,
  it will be suppressed and no default action performed.
  This allows applications to have a custom 'ESCAPE' or 'BACK' handling.
  Otherwise (not consumed) the default action is performed.

NEWT/Android: Remove KeyEvent.KEYCODE_HOME handling, since it doesn't work - i.e. cannot be intercepted this way.

Add GLStateKeeper handling GLEventListenerState preservation and restauration for GLAutoDrawable

- New GLStateKeeper interface, package com.jogamp.opengl

  Implemented by:
    - GLAutoDrawableBase

  Currently supported by:
    - NEWT GLWindow

- GLEventListenerState package move:
   com.jogamp.opengl.util ->  com.jogamp.opengl

Animator: Cleanup DEBUG output

NEWT/Android WindowDriver.closeNative0(..): Catch Exception at surface destroy allowing to continue destruction.

NEWT/Android (Bug 665): Add Support for GLStateKeeper ; onPause() always destroys ; Recognizing all GLAutoDrawable's GLAnimatorControl for pause/resume ; Use GLAnimatorControl instead of Animator

- Add Support for GLStateKeeper
  If !isFinishing() (HOME button), preserve the GLEventListener if an GLStateKeeper instance

- onPause() always destroys
  onDestroy() is too late, i.e. surfaceDestroyed() already called

- Recognizing all GLAutoDrawable's GLAnimatorControl for pause/resume
  pause/resume the GLAnimatorControl of all GLAutoDrawable instances

- Use GLAnimatorControl instead of Animator
  We used an Animator reference .. duh!

Note: The EGL native WindowDriver (Android and BCM.IV) must retain their own copy of EGLGraphicsDevice,
      which preserves the EGLDisplay handle due to EGLDisplayUtil reference counting per nativeHandleID.

NEWT/Android: Fix suspense regression of commit d514ecbf052d013ea8c0982c490757678075a9ea

Explicit win.destroy() shall only be called at activity.onDestroy(),
 - GLStateKeeper preservation is marked at pause,
 - and the Window's surfaceDestroyed() will also issue destroy() - so it's safe

Reason: On Suspense (or power button), application is paused _without_ surfaceDestruction !

NEWT/MouseEvent: Refine MultiTouch Event Spec regarding associated 'action' pointer Index; Add PointerType[PointerClass]; Add 'wheelScale' Attribute;

- Refine MultiTouch Event Spec regarding associated 'action' pointer Index
  In case an instance represents multi-touch events, i.e. {@link #getPointerCount()} is > 1,
  the first data element represents the pointer which triggered the action if individual to one pointer.
  For example {@link #getX(int) e.getX(0)} at {@link #EVENT_MOUSE_PRESSED} returns the data of the pressed pointer, etc.

- Add PointerType[PointerClass]
  This allows applications to identify the type and it's class [On-/Offscreen]
  helping to interpret semantics, e.g. wheel-rotate and new wheel-scale

- Add 'wheelScale' Attribute
  Returns the scale used to determine the {@link #getWheelRotation() wheel rotation},
  which semantics depends on the {@link #getPointerType() pointer type's} {@link PointerClass}.
  See API doc for details ..

TODO: NEWT/Android changes adopting these changes.

Android/NEWT MouseEvent: Fix Delivery of MultiTouch PRESSED/RELEASE; Fix Detection/Processing of 2-Finger-Scroll Gesture ; GearsES2: Add NEWT based 'zoom' gesture detection.

- Fix Delivery of MultiTouch PRESSED/RELEASE
  Adopting MouseEvent changes of commit 18cb57246372eda72c25a5cd9a69a873bdf09489

- Fix Detection/Processing of 2-Finger-Scroll Gesture

  Dropping utilization of Android's GestureDetector and implementing our own,
  which turns out to simplify keeping track of states.

  Our gesture detection works well w/ user NEWT based gesture detection (-> See GearsES2 zoom and rotate),
  using following criteria related to Android parameter:
     - ScaledDoubleTapSlop:
        - Max 2 finger distance

     - ScaledTouchSlop:
        - Min. movement w/ 2 pointer withing ScaledDoubleTapSlop starting 'scroll' mode
        - Max. change of finger distance in respect to initiating 2-finger distance (2x ScaledTouchSlop)
        - Max. distance growth in respect to initiating 2-finger distance.

- GearsES2: Add NEWT based 'zoom' gesture detection.

Newt/MouseEvent: Add 'float[3] getRotation()', getWheelScale() -> getRotationScale(), refinement of commit 18cb57246372eda72c25a5cd9a69a873bdf09489

Turns out the 'wheel' semantics are not generic enough and confining rotation values
to one axis only satisfies the traditional mouse wheel.

Widen the definition of 'rotation' and delivering 3-axis if supported.

On NEWT/Android, we deliver the 2-axis for example, allowing to rotate around both
or scrolling using both directions (-> GearsES2).

Android/NEWT MouseEvent: Fix Detection/Processing of 2-Finger-Scroll Gesture - Part 2

Multiple tests on different devices disclosed that:
  - 1 of 2 pointers get disconnected every now and then ..
    -> Shall tolerate this case

  - dist-delta within gesture may shrink below doubleTouchSlope
    -> Remove constraint after gesture detection

- Always validate pointer-id

GearsES2
  - Works quite stable on several devices now
  - Moved soft-keyboad show to 4-pointer pressed >= 0.7f pressure

GLJPanel: Refine API doc

NEWTEvent: Fix toString message - add comma

Unit Tests: Unify 'AWT Robot Key Press/Release', waitForIdle, action, delay, waitForIdle ; TestFocus0*: Wait until closed after win.destroy().

PNGImage: Add license header ; TestPNGImage01NEWT package move ..

SharedRessourceRunner: Add isDeviceSupported(..) query before spawn off thread, allowing a more gracefull detection of n/a GLX on X11

Complete 9a4fcc7ea4ec61e4ceed791acced734ac04ea270: SharedResourceRunner.Impl.createSharedResource(..): isGLXAvailableOnServer(..) call redundant since checked upfront at isDeviceSupported(..)

GLStateKeeper: Add clearPreservedGLState()

NewtBaseActivity: Perform cleanup() even on onCreate() due to possible call w/ initialized states (App process is killed for memory exhaustion etc)

.. even though I was not able to reproduce this case, it will not harm.

NEWTGearsES2Activity*: Adding property to enable 'process kill' behavior (memory and killSelf)

Adapt to new version scheme, see GlueGen a3f2ef50ad33c58a240a17fcf03e415d772207c3, etc; Fix NewtVersion, NativeWindowVersion and NewtVersionActivityLauncher

NewtVersion, NativeWindowVersion: Also search for extension javax.media.opengl (all packaging)
NewtVersionActivityLauncher: Use new launcher URI

NewtBaseActivity: Move setVisible(false) from onPause() -> onStop(), since only onStop() declares invisible status. onPause() could imply other non fullscreen activity on top.

Fix Bug 671: Add JPEG Decoder w/o AWT Dependencies

Original JavaScript code from <https://github.com/notmasteryet/jpgjs/blob/master/jpg.js>,
author 'notmasteryet' <async.processingjs at yahoo.com>.

Ported to Java.

Enhancements:
 * InputStream instead of memory buffer
 * User provided memory handler
 * Fixed JPEG Component ID/Index mapping
 * Color space conversion (YCCK, CMYK -> RGB)
 * More error tolerant

+++

Features:
              JOGL   AWT
  RGB          ok     ok
  YCCK         ok     Exception
  CMYK         ok     Exception
  YUV Store    ok     n/a
  Need Y-Flip  no     yes

+++

Benchmark: TestJPEGJoglAWTBenchmarkNewtAWT

  JOGL.RGB Loops 100, dt 1199 ms, 11.99 ms/l
  JOGL.YUV Loops 100, dt  351 ms,  3.51 ms/l
  AWT..... Loops 100, dt 2144 ms, 21.44 ms/l

File:    jogl/src/test/com/jogamp/opengl/test/junit/jogl/util/texture/j1-baseline.jpg
Machine: GNU/Linux PC (AMD 8 core), JavaSE 6 (1.6.0_38)

.++++ UITestCase.setUp: com.jogamp.opengl.test.junit.jogl.util.texture.TestJPEGJoglAWTBenchmarkNewtAWT - benchmark
libEGL warning: DRI2: failed to authenticate
0: JPEGImage[261x202, bytesPerPixel 3, reversedChannels false, JPEGPixels[261x202, sourceComp 3, sourceCS YCbCr, storageCS RGB, storageComp 3], java.nio.DirectByteBuffer[pos=0 lim=158166 cap=158166]]
0: TextureData[261x202, y-flip false, internFormat 0x1907, pixelFormat 0x1907, pixelType 0x1401, border 0, estSize 158166, alignment 1, rowlen 0, buffer java.nio.DirectByteBuffer[pos=0 lim=158166 cap=158166]
JOGL.RGB Loops 100, dt 1199 ms, 11.99 ms/l
0: JPEGImage[261x202, bytesPerPixel 3, reversedChannels false, JPEGPixels[261x202, sourceComp 3, sourceCS YCbCr, storageCS YCbCr, storageComp 3], java.nio.DirectByteBuffer[pos=0 lim=158166 cap=158166]]
0: TextureData[261x202, y-flip false, internFormat 0x1907, pixelFormat 0x1907, pixelType 0x1401, border 0, estSize 158166, alignment 1, rowlen 0, buffer java.nio.DirectByteBuffer[pos=0 lim=158166 cap=158166]
JOGL.YUV Loops 100, dt 351 ms, 3.51 ms/l
0: TextureData[261x202, y-flip true, internFormat 0x1907, pixelFormat 0x80e0, pixelType 0x1401, border 0, estSize 158166, alignment 1, rowlen 261, buffer java.nio.HeapByteBuffer[pos=0 lim=158166 cap=158166]
AWT..... Loops 100, dt 2144 ms, 21.44 ms/l
++++ UITestCase.tearDown: com.jogamp.opengl.test.junit.jogl.util.texture.TestJPEGJoglAWTBenchmarkNewtAWT - benchmark

NewtCanvasAWT: Cleanup up private method calls; Attach/Detach JAWTWindow's SurfaceUpdateListener to NEWT Window when it gets attached/detached.

OSX CALayer minor edit: Call [super setFrame..] directly

JAWTWindow: Issue component.repaint() after attachSurfaceLayer(..), otherwise no rendering result might be visible w/o Animator

OSX/CALayer Threading Part4: Stream all JAWT Root CALayer Operations on OSX Main-Thread

Previous code created, set and unset the root CALayer on the current thread,
which lead to a very delayed destruction of the root CALayer w/.
With Java7 this lead to a possible resource starvation in certain situations,
since Java7 uses an CAOpenGLLayer.

Similar w/  f354fb204d8973453c538dda78a2c82c87be61dc,
creation, set and unset is operated on main-thread.

OSX CALayer/NewtCanvasAWT: Force relayout of component (size +1, size -1) when attaching NEWT Child to avoid false CALayer position.

OSX CALayer Unit Tests: NEWT child is destroyed if NewtCanvasAWT is disposed/removed, due to offscreen/onscreen transition

SWT unit tests: Sort w/ AWT and Headless - Fix threading w/ AWT.

This especially impacts (fixes) Mac OSX.

Bug 713: Handle broken EGL setup - Catch all Exception types, not only JogampRuntimeException at *DrawableFactory* instantiation trial

In case EGL is not completly installed, EGLDisplayUtil.eglGetDisplayAndInitialize(..)
will throw a GLExeception which was not catched in GLDrawableFactory.

The latter only catched JogampRuntimeException caused by ReflectionUtil due to n/a classes,
but the actual initialization code is capable to throw others.

Bug 707: Fix NEWT EVENT_MOUSE_EXITED not sent on Windows - Regression of commit 85338858f5c58694fa88e77df1386d0556887944

Commit replaced enqueueMouseEventID w/ sendMouseEventID, while not removing the 'jboolean wait' argument.
This also lead to staying in DRAGGED mode when mouse left the window.

Adds a unit test and a fix for the bug 697

TestGLAutoDrawableGLCanvasOnOffscrnCapsAWT: Wrong test comment, no NEWT involved

Bug 678: Differentiate ALT (left) and ALT_GRAPH (right) on X11, EventDev, Windows and OSX

- X11: Memorize pressed Alt_R to decide which 'alt' has to be used for non key modifier fetching
- Windows: Only use GetKeyState(..) and compare the US vkey, since int. kbd layout use reduced scancode

Bug 714: Add GL_COPY_READ_BUFFER_BINDING and GL_COPY_WRITE_BUFFER_BINDING, which were added to spec post release.

Bug 709: Testing texured demo w/ FBO, comparing w/ onscreen - No corruption.

Adds the missing file

Bug 641 NEWT: X11 Deliver keyCode layout independent, keySym layout dependent and UTF-16 keyChar value

On X11, the layout dependent keySym was not delivered [1],
as well as the UTF-8 to UTF-16 translation was missing [2].

[1] is solved using XLookupString w/o ShiftMask
[2] is solved using JNI's NewStringUTF, which takes UTF-8.

NEWT/Native: NewtCommon_GetJNIEnv(..) adding 'asDaemon' flag, used by all OSX JNI attachments to save time since detachment is skipped.

Bug 641 NEWT/X11: Deliver 'J_VK_WINDOWS'; keySym is layout independent for dead-keys (zero keyChar); For 'unshifted' keySym's also unmask Ctrl and Mod* states

NEWT/OSX: Remove USE_SENDIO_DIRECT conditional compilation, by now it's clear we need to enqeue the events

NEWT KeyEvent: More tight definition of isPrintable(), i.e. must have defined keyCode [and keyChar]; VK_KEYBOARD_INVISIBLE -> isActionKey

Bug 641 NEWT/OSX Deliver keySym, derived from the current layout's lower-case UTF-16 character.

Bug 641 NEWT: Refine definition of keySym, i.e. for printable keys, keySym is the 'unshifted' UTF-16 char value ; Add isPrintable() to toString().

Fix Bug 703: Quaternion slerp function. Disclaimer: Not tested / No unit test.

LinuxEventDeviceTracker: shift-modifier

Signed-off-by: Xerxes Rånby <xerxes@zafena.se>

LinuxEventDeviceTracker: 4 space indentation

Signed-off-by: Xerxes Rånby <xerxes@zafena.se>

LinuxEventDeviceTracker: ALT, ALT_GRAPH & CTRL modifiers.

Signed-off-by: Xerxes Rånby <xerxes@zafena.se>

Debug Messages: Unify getThreadName() usage and 'Catched Exception' messages

GLDrawable: Refine API doc; GLDrawableImpl: Balance createHandle()/destroyHandle(); Handle LOCK_SURFACE_CHANGED in lockSurface() ; GLContextImpl.makeCurrent(): Fix drawable handle validation.

GLDrawable: Refine API doc (realized/handle)
  - Lifecycle of the drawable handle was not clear
  - Ephasizing handle's dependency on NativeSurface's lock state and drawable's realization

GLDrawableImpl: Balance createHandle()/destroyHandle()
  - updateHandle() -> createHandle()
  - ensure both are balance, see below

GLDrawableImpl: Handle LOCK_SURFACE_CHANGED in GLDrawableImpl's lockSurface()
  - call destroyHandle() and createHandle()

GLContextImpl.makeCurrent(): Validate drawable handle if realized only.
  - it is valid to have an invalid drawable handle if not realized (see above)

GLContextImpl.destroy(): Fail fast in case of invalid dawable [null, !realized] and makeCurrent() failure.

jogl: remove executable bit from some java files and a png

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>

NEWT KeyEvent: Use UTF-16 UniCode for key-code and key-symbol exposing well defined key encoding and simplify usage.

Note, we use one collision to reduce key-code range:
  [0x61 .. 0x78] keyCodes [F1..F24] collide w/ ['a'..'x']

  Since keyCode/Sym won't use lower capital a-z, this is a no isssue.

KeyEvent:
  - 'printable' type is being determined by a nonPrintableKeys table,
    while 'action' type is set in case !printable and !modifier.

  - public ctor hidden, use create(..) method instead.
    This allows us to ensure modifier bit are properly set (incl. the keySym one) w/o performance loss.
  - ctor validates that only one of the type flags is set, printable, modifyable or action.

WindowImpl:
  - Using IntBitfield of 255 bits to track pressed state,
    while removing the repeat state tracking since it is redundant.
    The Windows impl. uses a single field to validate whether a key
    was already repeated or not.
  - Properly cast keyCode short values to int for tracking!

AWTNewtEventFactory, SWTNewtEventFactory:
  - Add translation of keyCode/Sym from and to NEWT
  - All tested via:
    - Newt -> Awt for AWTRobot
    - OSX CALayer: AWT -> NEWT
    - SWT tests

X11:
  - Add VK_CONTEXT_MENU mapping (XK_Menu)

LinuxEventDeviceTracker:
  - Fix apostrophe and grave mapping, i.e. to VK_QUOTE and VK_BACK_QUOTE.

Adapted all unit tests, especially:
  - TestNewtKeyCodesAWT: More fine grained keyCode ranges to test
    using proper keyCode symbols.

LinuxEventDeviceTracker: Reapply changes from commit b13868b612689307ebf4e46ee4ede2fd0560e504, which got lost via my merge conflict resolution (duh).

NEWT/WindowsWindow.c: Simplify pointerVisible branch for calling SafeShowCursor(..)

Bug 697: Fix commit 4db745e84fac610f85ab085e5c147e571e82e008 - Not compile clean! Also brought back my safe showCursor logic.

The code was not compile clean:
  - declaration after use w/o prototype!
  - missing var declaration 'success'!
  - 'pointerVisible=0', instead of 'wud->pointerVisible=0' var. not found!

Used my old safe show cursor logic, while removing the max count,
which was redundant, since we can check whether the counter moves
in the right direction.

GLContextImpl: destroy(): Fix null drawable check; makeCurrentWithinLock(): Add drawable size validation before attempt to create context; MacOSXCGLContext: Explicit exception for Null AttachGLLayerCmd.

GLContextImpl:
  - destroy():
    - Fix null drawable check
      Only if the GLContext isCreated(), we require a drawable to be set.

    - Proper name of associateDrawable Exception

  - makeCurrentWithinLock():
    - Add drawable size validation before attempt to create context.
      - 'makeCurrent()' shall never be called w/o proper sized drawable, i.e. > 0x0.
      - returns CONTEXT_NOT_CURRENT, if drawable size is <= 0x0

MacOSXCGLContext:

  - Explicit exception for Null AttachGLLayerCmd.
    In case context creation has failed via makeCurrent(),
    AttachGLLayerCmd may never been issued and hence maybe NULL.
    Catch this case and send a meaningful exception,
    which is catched and fwd in GLContext.destroy().

FBObject.init() uses safe 1x1 size like reset(..) to behave equal; GLFBODrawableImpl.contextMadeCurrent(ctx, true): Throw exception if !initialized

NEWT/AWT: Fix 'AWT driver' to work w/ OSX CALayer ; WindowImpl: 'Object getWrappedWindow()' -> 'NativeSurface getWrappedSurface()'

WindowImpl: 'Object getWrappedWindow()' -> 'NativeSurface getWrappedSurface()'
  - AWT driver itself instantiates the JAWTWindow for eager initialization at createNative().

Fix 'AWT driver' to work w/ OSX CALayer
  - See above
  - size reconfig changed to ease OSX CALayer,
    i.e. set frame's size if already visible reducing CALayer artefacts.

NEWT/AWT-Wrapper-Driver: Fix MS-Windows Case (Use NativeWindow window handle); Use set size by frame only for OSX/CALayer

GLES[12] Impl: Cleanup 'checkBufferObject(..)', remove unused and 'avail' argument and 'initBufferObjectExtensionChecks()'

Bug 692: Add modified version of Edgar Velazquez-Armendariz's test case for testing alternating VAO and VBO usage.

Alternating VBO/VAO usage triggers bug 692, where our VBO enable check throws an exception:
  javax.media.opengl.GLException: element vertex_buffer_object must be enabled to call this method
        at jogamp.opengl.gl4.GL4bcImpl.checkBufferObject(GL4bcImpl.java:34318)
        at jogamp.opengl.gl4.GL4bcImpl.checkElementVBOEnabled(GL4bcImpl.java:34361)
        at jogamp.opengl.gl4.GL4bcImpl.glDrawElements(GL4bcImpl.java:4395)
        at javax.media.opengl.DebugGL3.glDrawElements(DebugGL3.java:1006)
        at com.jogamp.opengl.test.junit.jogl.acore.TestBug692GL3VAO$GL3VAODemo.displayVAONormal(TestBug692GL3VAO.java:254)

Fix Bug 692: Add tracking of VERTEX_ARRAY_BINDING, and enable allowing a bound non default VAO to pass VBO enabled test, even if VBO is disabled.

VAO is available if: GL >= 3.0 or is having GL_ARB_vertex_array_object extension.

checkBufferObject(..) checks whether VERTEX_ARRAY_BINDING has a non default VAO bound in case
no VBO is being bound and VAO is allowed.

glBindVertexArray(int) is being tracked, i.e. on state VERTEX_ARRAY_BINDING

Bug 705: This is the version of displayMultiple02 which uses glXCreateNewContext instead of glXCreateContext. Segmentation fault w/ Mesa 7.0

Bug 705: This is the version of displayMultiple02 which uses glXCreateNewContext instead of glXCreateContext. X11 Error w/ Mesa 7.0

FFMPEGMediaPlayer: Fix mpeg seek using time_base.num 1 , time_base.den 90000

Prevent division and multiplication by zero errors in native code
after mpeg video seek caused by type truncation.

Signed-off-by: Xerxes Rånby <xerxes@zafena.se>

Fix Bug 705 - Make Mesa X11 driver < 8.0 use new Quirk DontCloseX11DisplayConnection

Fix Bug 705 - Cleanup & Generalize Commit 5b47372590ec715647ebbd75d70c41ec7a64485a ; Close X11 Display in isDeviceSupported()

- Moved GL vendor version parsing to GLVersionNumber

- Moved X11Util.markAllDisplaysUnclosable() trigger into SharedResource creation of
  - X11GLXDrawableFactory
  - EGLDrawableFactory

- GLProfile is back to pre 5b47372590ec715647ebbd75d70c41ec7a64485a,
  i.e. contains no quirk artifact (clean)

- Close X11 Display in X11GLXDrawableFactory.isDeviceSupported()
  Regression of 9a4fcc7ea4ec61e4ceed791acced734ac04ea270

- TODO: Remove X11Util markAllDisplaysUnclosable detection code ?

Notes to Martin:
  - Use TAB == 4 SPACES
  - No author names into source code, git commit log is enough.
  - No need to tag your edits, the diff is enough.

Fix Bug 705 - Validated that Quirk DontCloseX11Display is not required w/ Mesa >= 8.0

Tested w/ kubuntu 12.04, Mesa 8.0.4 (glx/dri/gallium and X11 SW rasterizer)

Fix Bug 705 - Set Quirk DontCloseX11Display also if using X11 and X11Util has set 'markAllDisplaysUnclosable' (Down -> Upstream)

Bug 718: Windows BITMAP Offscreen Fails w/ GLCaps other than simple RGB888 - Refine Unit Tests ; UITestCase.Snapshot: Add RGBA bits to filename.

Split TestGLAutoDrawableFactoryOffscrnCapsNEWT to
      TestGLAutoDrawableFactoryGL2OffscrnCapsNEWT and TestGLAutoDrawableFactoryES2OffscrnCapsNEWT

TestGLAutoDrawableFactoryGL2OffscrnCapsNEWT contains more 'Bitmap' tests of various caps, w/ snapshots.

*Capabilities: Cleanup string ctor; GLGraphicsConfigurationUtil: Clean getExclusiveWinAttributeBits(..); WGLGLCapabilities: Add PFD2String(..)

Bug 718: Windows BITMAP Offscreen Fails w/ GLCaps other than simple RGB888 - Filter invalid PFD configs

- Filter invalid PFD configs
  - WindowsBitmapWGLDrawable: Clip chosenCaps to RGBA888[0|8]
  - WindowsBitmapWGLDrawable: Only use BITMAPINFOHEADER.BiBitCount=24
  - WindowsWGLGraphicsConfiguration: Only allow GDI BITMAP PFD's w/ RGB888 w/ alpha <= red, others may fail

- WindowsWGLGraphicsConfigurationFactory.getAvailableGLCapabilities()
  - Fetch ARB caps w/o BITMAP
  - Concat GDI [BITMAP] caps

GLJPanel.toString(): Avoid NPE

Fix Bug 720: Unify all platform specific GLContextImpl specializations; Fix Bug 719 - Windows BITMAP Offscreen Orientation is not propagated through API

Fix Bug 719 - Windows BITMAP Offscreen Orientation is not propagated through API
  Depends on Bug 720, since cleaning up GLContextImpl* is required to move
  property 'GLContext.isGLOrientationFlippedVertical()'
  to 'GLDrawable.isGLOriented()' where it belongs!

  Windows BITMAP GLDrawable impl. isGLOriented() shall return false,
  while we keep the BITMAPINFOHEADER's height field negative
  to remove the need for vertical flip when used w/ AWT or Windows, ..

  Then property 'GLDrawable.isGLOriented()' has to be recognized throughout the
  utility functions, i.e. TextureData's mustFlipVertically and hence TextureIO writer.

Fix Bug 720: Unify all platform specific GLContextImpl specializations
  GLContextImpl shall have only _one_ unique platform derivative
  to allow proper swapping of GLDrawables of any type via:
    - 'GLAutoDrawable.setContext(GLContext newCtx, boolean destroyPrevCtx)', which calls
    - 'GLContext.setGLDrawable(GLDrawable readWrite, boolean setWriteOnly)'

  Exception: External context may be specialized.

  All drawable specific property handling shall be provided
  and implemented (if possible) via GLDrawable specializations.

- GLContext.isGLOrientationFlippedVertical() -> GLDrawable.isGLOriented()
  - PNGImage.createFromData() takes 'isGLOriented' to properly handle vertical flipping simply by line ordering
  - TextureIO's PNG writer passes TextureData's getMustFlipVertically() as isGLOriented to PNGImage.createFromData()
  - GLReadBufferUtil respects GLDrawable's isGLOriented() when creating TextureData instance.
  - Screenshot respects GLDrawable's isGLOriented()

- Screenshot is deprecated, use GLReadBufferUtil.

- Removed all PBuffer attributes, i.e. floatingPoint, RenderToTexture and RenderToTextureRectangle.
   - Allows removal of special pbuffer handling in GLContext* implementations.
   - Removed also from GLCapabilities*
   - Removed from deprecated GLPbuffer

  Impact:
    - Low, users who desire to render into a texture shall use our FBO GLOffscreenDrawable.
    - Only use case was the deprecated GLPbuffer
    - floating point framebuffer technology is still patented anyways :)

- Removed Java2DGLContext, which was only used for OSX's GLJPanel Java2D bridge,
  which is no more supported anyways.

GLJPanel: setFocusable(true) at ctor to allow keyboard input

Bug 721: NEWT/X11 RandR 1.3 Support - Part 1 - Delegate RandR functionality to pluggable impl. for RandR 1.1 and 1.3 (todo)

Bug 718: Further restriction of Windows BITMAP PFD's - RGB888 and !alpha only

On Windows Vista machines, using alpha, i.e. RGB8888 failed to SetPixelFormat(..),
hence only allow RGB888.

Bug 718: Exclude updateGraphicsConfigurationARB(..) usage for BITMAP (Windows Vista does report BITMAP w/ ARB PFD)

On Windows Vista machines, ARB PFD selection reports BITMAP which is attempted to be used.

Excluse updateGraphicsConfigurationARB(..) from BITMAP query, similar to getAvailableGLCapabilitiesARB()
from commit 61a47e07975eb2fd8b1f5f60552935c993a6eef6.

TODO: Re-evaluate commit db24615ebaebcda88ffb7275d3a60e6400226099

type: weather -> whether :) .. it's spring again :)

Bug 718: WindowsBitmapWGLDrawable: Requires GLProfile.GL2, fix BITMAP GLCaps ASAP at Ctor instead of setRealized(true); WindowsWGLContext: Exclude ARB creation for BITMAP

Unit Test TestGLAutoDrawableFactoryGLnBitmapCapsNEWT added using BITMAP on GLProfile.getDefault()

Also:
  X11GLXContext, WindowsWGLContext: Cleanup formatting in createImpl(..)

GLJPanel: Avoid NPE of 'backend' in getChosenGLCapabilities() and DisposeGLEventListenerAction.run() ; Make 'backend' volatile due to possible multi-threading access.

Bug 721: NEWT/X11 RandR 1.3 Support - Part 2 - Rename impl class/files, start RandR13 coding.