summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* NEWT/Android: 565, 5551 or RGBA_8888 (no RGBX/888) ; surfaceRealized() ASAPSven Gothel2011-10-014-39/+70
|
* AndroidWindow: get required PixelFormat ID by req. CapsSven Gothel2011-09-302-4/+48
|
* Android Launcher: Change properties .. debug, etc - disable traceSven Gothel2011-09-301-2/+3
|
* RedSquareES1: remove GLU dependencySven Gothel2011-09-301-11/+10
|
* Android/Tests: Adding missing activitiesSven Gothel2011-09-306-0/+267
|
* Android Launcher: Sort properties / enable DebugGL/TraceGLSven Gothel2011-09-301-8/+13
|
* GLContext/GLDebugMessageHandler: Fix/clarify init and usage of dbg handlerSven Gothel2011-09-303-4/+23
|
* GLPipelineFactory: Generics && Using enhanced GlueGen's ↵Sven Gothel2011-09-301-16/+13
| | | | ReflectionUtil.getConstructor(..)
* Android/JogAmp ClassLoader: Split CL JogAmp + User-APK, reuse JogAmp CL if ↵Sven Gothel2011-09-301-33/+54
| | | | in same ClassLoader
* Android: Use standard launch mode; Add default actions to NewtBaseActivity; ↵Sven Gothel2011-09-307-176/+35
| | | | Add RedSquare ES1 + ES2
* NEWT/Android: More ANativeWindow coop;Sven Gothel2011-09-302-48/+129
| | | | | | | | | | | | | | | | Note: surface's format _is_ the nativeVisualID of EGL. More ANativeWindow (ANW) coop: - add acquire/release for ANW - validate ANW format - Add fixCaps: 'format -> GLCaps' validation Test: - Constructor: Start w/ surface format RGB_565 (default) - createNative: Filter EGLConfig w/ surface format - Move surfaceHandle fetching from surfaceCreated -> surfaceChanged, where the format is available. - surfaceChanged: use fixedCaps validation
* EGL: eglSwapBuffers() show failure in DEBUG modeSven Gothel2011-09-301-1/+6
|
* ES Library lookup order - Favor spec lib nameSven Gothel2011-09-302-11/+17
|
* EGL: chooseGraphicsConfigurationStatic() public and w/ optional visualIDSven Gothel2011-09-302-14/+35
|
* Graph GLSL: Use global precicision settings enhancing readabilitySven Gothel2011-09-307-60/+54
|
* Android Launcher Demos: Adept to df6021f9ef50623e6ac8f0822e827fd31a0a0cf2Sven Gothel2011-09-304-11/+43
|
* Android Launcher: Generalize ClassLoaderUtil. Use local TempFileCacheSven Gothel2011-09-302-8/+506
|
* demo: es2 shaders: use default precisionSven Gothel2011-09-282-113/+106
|
* Android/Demos: Add ElektronenMultiplizierer (fix shader and keylistener ↵Sven Gothel2011-09-284-21/+24
| | | | add/remove)
* GearsES1/ES2 demos: remove junit assertionsSven Gothel2011-09-282-15/+1
|
* NEWT DisplayImpl: Check for null event in taskSven Gothel2011-09-281-0/+6
|
* NewtBaseActivity: Enable 'slave' mode, ie as a downstream for external ↵Sven Gothel2011-09-282-94/+88
| | | | launched Activity, see NewtLauncherActivity
* NEWT/Android: Send MouseClicked event, MouseEvent enhancement, Demo: real 2 ↵Sven Gothel2011-09-285-32/+155
| | | | | | | | finger zoom MouseEvent: Check array sizes at cstr. Enhance 'toString()', pointer arrays added. GPUUISceneGLListener0A now uses 2 finger distance for zoom
* TypecastFontConstructor: Use IOUtil's createTempFile() to comfort AndroidSven Gothel2011-09-281-4/+2
|
* Fix ES2 shadersSven Gothel2011-09-285-7/+9
| | | | | | | | | | | Add missing floating point precision qualifiers. '#extension OES_standard_derivatives : require' somehow doesn't pass the GLSL compiler, however '#extension GL_OES_standard_derivatives : enable' does. Currently works on ARM's Mali-400 MP, NV Tegra still don't show a picture. TODO: More in depth evaluation.
* Fix GLProfile ES2: Component numbers can't be validated (??)Sven Gothel2011-09-281-5/+3
|
* Android Activity Launching (jogl.test)Sven Gothel2011-09-288-69/+421
| | | | | | | | | | | | | | | | Launching activity is in: jogl.android-launcher.apk and directly derives from NewtLauncherActivity. It daisy chains apk's via ClassLoaderUtil: - gluegen-rt.apk - jogl.all-android.apk - jogl.test.apk (*) (*) This has to made configurable so the generic NewtLauncherActivity can be reused by any user application. After preparing the ClassLoader (see above), NewtLauncherActivity instanciates the configurable user Activity and passes all it's activity calls down to it.
* NEWT/OSX: 'Better' child window positioning, still, after reparenting into ↵Sven Gothel2011-09-275-9/+29
| | | | | | | | | parent, a white window rectangle remains. .. we also need to understand the absolute screen position better, ie. when required and when not (at window creation currently).
* GLContextImpl: Fix bug: Surface was not unlocked in case new context ↵Sven Gothel2011-09-271-7/+9
| | | | creation failed.
* tests: added some commented out 'wait for press enter' startups, reduce ↵Sven Gothel2011-09-273-6/+13
| | | | getCurrentScreenMode() loop to 100
* NEWT/Threading: MainThread / DefaultEDTUtilSven Gothel2011-09-275-224/+94
| | | | | | | | | | | | | | | | | | | | | | | - MainThread: This class no more implements EDTUtil! This class just provides a main-thread utility, forking of a main java class on another thread while being able to continue doing platform specific things on the main-thread. The latter is essential for eg. MacOSX, where we continue to run NSApp.run(). - DefaultEDTUtil: - if Lock.DEBUG validate that no recursive locks are being hold, where it shall not (EDT: startup and return from task execution) - If task execution's result wasn't waited for (checked), at least dump exeception's stack trace if i happened. - MacDisplay: Just use DefaultEDTUtil - MacWindow: - No more need of special locking -> removed nsViewLock, since: - using proper EDT - capability to run from multiple threads (native Java thread attachment)
* NEWT: Adapt to GlueGen's Lock ChangeSet, all java callbacks for native have ↵Sven Gothel2011-09-2715-132/+193
| | | | | | | | | | | | | | 'defer' 1st argument - Adapt to GlueGen's Lock ChangeSet: e4baba27507ce78e64a150ec6f69fb96f5721a34 - All java callbacks for native have 'defer' 1st argument. This allows enqueuing resulting events to the EDT if required, ie. the native thread may not be 'compatible' (MacOSX). - MacOSX-Native: enqueue key/mouse events and defer:=true for all java callbacks Since we are comming from a 3rd-party thread (AWT/NSApp-MainThread) we shall not abuse it.
* Adapt to GlueGen's Lock ChangeSet: e4baba27507ce78e64a150ec6f69fb96f5721a34 ↵Sven Gothel2011-09-277-40/+40
| | | | ; Use generics
* NEWT/JOGL: MacOSX UpdateSven Gothel2011-09-2516-514/+617
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Feature related: - Added always-on-top - Added translucency - Child Window Position - AWT parent: manual traverse up the tree and calc position on screen (Problem: the parent view rect is not at the proper position, but covers the whole frame) EDTUtil related: - Works now w/ AWT ot headless (again) - OSX native JNI callbacks gathering JNIEnv properly and attaches/detaches thread. - AWT case: using AWT-Event which properly dispatches our cocoa events - MainThread (headless) case: Fork off thread w/ main class and kick off NSApp run(). This leads to same behavior as w/ AWT case. - Using DefaultEDTUtil - Cleanup MainThread (implements EDTUtil) - Currently not used as EDTUtil (osx), just as launcher - Removed EDTUtil impl code, reuse DefaultEDTUtil - Cleanup AWTEDTUtil (implements EDTUtil) - Currently not used as EDTUtil (osx)
* NativeWindow/OSX: Add nativewindow_macosx lib and nativewindow.os.macosx.jar ↵Sven Gothel2011-09-254-2/+159
| | | | for 'GetLocationOnScreen()'
* CreateWindow set visible; lockSurface: lock AWT if availableSven Gothel2011-09-241-11/+15
|
* OSX: Sync MainThread w/ DefaultEDTUtil and proper deledation AWT EDT, ↵Sven Gothel2011-09-242-77/+176
| | | | MacWindow: create/visible at native creation
* Cleanup DefaultEDTUtil: Generics and DEBUGSven Gothel2011-09-241-32/+22
|
* Use new JNILIbLoaderBase addNativeJarLibs(all, atomic) for JOGL, ↵Sven Gothel2011-09-243-26/+19
| | | | NativeWindow and NEWT; no more LD_LIB_.. in setenv.sh for test scripts
* Use Platform's initSingleton() instead of JVMUtil's (private package) ; Use ↵Sven Gothel2011-09-234-21/+53
| | | | TempJarCache if used.
* Proper usage of loadLibrarySven Gothel2011-09-212-9/+4
|
* NEWT/WindowImpl: Remove wait for position (keep waitForSize for ↵Sven Gothel2011-09-164-31/+9
| | | | | | reparent/fullscreen) Window position is not deterministic enough and slows down processing while sync on it
* NEWT Cleanup: AWT driver fix, WindowImpl createNative .., minor editsSven Gothel2011-09-165-35/+22
| | | | | | | | | AWT driver fix - setVisible at creation, using reconfigureWindow impl (fixed WindowImpl createNative .., - no double visibility check on creation - visible:=true anyway for later creation, if failed or not available yet
* minor editsSven Gothel2011-09-162-5/+10
|
* waitForPosSize/createWindow: exclude 0/0, at creation wait for size as wellSven Gothel2011-09-161-37/+9
| | | | | | | - 0/0 may result in -1/-1, which is impl. specific (X11), might get deleted if causes more harm than ham - waiting for size after creation is actually a good thing todo
* Test NEWT Parenting: Incr. test duration to 600ms (new state poll of ↵Sven Gothel2011-09-161-1/+1
| | | | reparenting may take ~400ms)
* NEWT Window: Allow positionChanged(..) notification for child windowsSven Gothel2011-09-161-1/+1
|
* Fix AWT Test Cases: Use AWT-EDT for modifying visible frame and call ↵Sven Gothel2011-09-163-58/+84
| | | | validate() afterwards
* NEWT/Reparent: Make new local position/size persitent if not promoted ↵Sven Gothel2011-09-161-0/+14
| | | | natively (invisible or recreation)
* NEWT/X11: Fix ABOVE handling, notify Java on reparentNotify, cleanup of ↵Sven Gothel2011-09-162-52/+118
| | | | setFullscreenEWMH