aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* TiledPrintingAWTBase: Fix scaling - Fit frame to page, add MSAA ↵Sven Gothel2013-09-079-238/+348
| | | | | | | | | | | | | | RenderingHints test; setupPrint(Graphics2D): Query RenderingHints to use MSAA rendering - AWTPrintLifecycle.setupPrint(Graphics2D): Query RenderingHints to use MSAA rendering - Impl. in GLCanvas - TODO GLJPanel (would need a new offscreen buffer) - TiledPrintingAWTBase: - Fix scaling - Fit frame to page - add MSAA RenderingHints test - GLCanvas: Remove dumpStack() DEBUG output
* TileRenderer*: Fix FBO MSAA use-case, i.e. call swapBuffers() before ↵Sven Gothel2013-09-079-55/+234
| | | | | | | | | | | endTile(); Enhance unit tests for MSAA, also add TileRendererBase.TileRendererNotify to GearsES2 GL[Auto]Drawable.swapBuffers() must be called before endTile(). This is especially important if using multisampling offscreen FBO drawables, where swapBuffers() triggers the <i>downsampling</i> to the readable sampling sink. Otherwise, we will be 'one tile behind' !
* Fix TileRenderer*: Allow general usage w/ any GL profile, only image-buffer ↵Sven Gothel2013-09-073-25/+77
| | | | | | | | | | | | | requires >= GL2ES3; Always set pack-alignment, Set glReadBuffer(..) >= GL2ES3 - Allow general usage w/ any GL profile, only image-buffer requires >= GL2ES3 Due to GL2ES3.GL_PACK_ROW_LENGTH and image-width != tile-width - Always set pack-alignment Forgot for tile-buffer - Set glReadBuffer(..) >= GL2ES3 Required if using FBO offscreen, i.e. MSAA mode.
* Complete commit 54dcdf1f53c9fc1f7124cf77bbf5aa952d42053a, exposing GL2ES3 in ↵Sven Gothel2013-09-079-29/+31
| | | | GLProfile (just a query, no new profile)
* AWT/GL Printing WIP: Abstract AWT tile painting code out to AWTTilePainter, ↵Sven Gothel2013-09-073-301/+300
| | | | reused w/ GLCanvas and GLJPanel
* TestTiledPrintingGears*AWT: Add 4s delay before closing frame and ending ↵Sven Gothel2013-09-063-9/+5
| | | | | | unit test to wait for print-job (?) .. otherwise results are cut-off .. better solution ? callback ?
* GLCanvas Printing WIP: Add GLJPanel; GLCanvas TILE_SIZE 512; ↵Sven Gothel2013-09-067-88/+589
| | | | print(Graphics): Don't force on AWT-EDT -> Deadlock; releasePrint(): perform reshape/display!
* Add AWTPrintLifecycle interface describing impl. w/ ↵Sven Gothel2013-09-064-258/+367
| | | | setupPrint()/releasePrint(), modularize unit test TestTiledPrintingGearsAWT
* Added AWTPrinting document ..Sven Gothel2013-09-061-0/+53
|
* TestTiledPrintingGearsAWT: Perform AWT mods on AWT-EDT, always perform ↵Sven Gothel2013-09-061-13/+23
| | | | g.scale(..), pass 'offscreen' w/ manual printing as well
* GLCanvas Printing WIP: Perform all print* operations on AWT-EDT, clip before ↵Sven Gothel2013-09-062-173/+260
| | | | drawing image and show same; Test: Add offscreen print and add borders.
* TileRenderer*: Add more detailed 'toString()' - override for TileRendererSven Gothel2013-09-062-4/+24
|
* GLCanvas Printing WIP: Swap to offscreen at setupPrint(); Use TileRenderer @ ↵Sven Gothel2013-09-063-167/+278
| | | | | | | | | | | | | | | | | | | | | | | | print(..); Test: Add auto-print w/o UI 72dpi and 300dpi NOTE: Tile raster is added at print plus lots of debug output - WIP! - Swap to offscreen at setupPrint() - Swap GLEventListener (GLEL) and GLContext to fixed size offscreen GLAutoDrawable (GLAD) - Destroy onscreen drawable/context - Setup TileRenderer w/ offscreen GLAD - @ releasePrint() restore onscreen .. This shall ensure we don't hog the GPU resources when resizing the AWT component (upscaling). - Use TileRenderer @ print(..) The clip-rect will be used as the TileRenderer's image-size and we perform a whole tile-rendering loop. This helps to limit resources, see above. - Test: Add auto-print w/o UI 72dpi and 300dpi
* TileRendering / Gears: Clarify PMV math, if tiling use white background (-> ↵Sven Gothel2013-09-061-6/+14
| | | | print)
* TileRenderer*: TileRenderer adds setTileOffset(..), getParam(pname) shall ↵Sven Gothel2013-09-064-37/+25
| | | | | | | | | | | | | | | | | | | always return last result, no overloading of attachToAutoDrawable(..); Fix comments TileRenderer: - adds setTileOffset(..) A tile offset might be required, i.e. via a given rectangular clip bounds - getParam(pname) shall always return last result Even when finished tiling, the last value shall be returned, otherwise a post endTile() is unable to retrieve the value. - No overloading of attachToAutoDrawable(..) No reason to complicate usage by mutating semantics, call setTileSize(..) manually TileRendererBase: - Fix API doc: TR_CURRENT_TILE_X_POS, TR_CURRENT_TILE_Y_POS
* GLDrawableUtil.swapGLContextAndAllGLEventListener: Fix comment - Add remark ↵Sven Gothel2013-09-061-0/+4
| | | | about GLAnimatorControl's which association gets swapped as well.
* Add TestGLContextDrawableSwitch02AWT: Test AWT GLCanvas <-> ↵Sven Gothel2013-09-052-1/+195
| | | | GLOffscreenAutoDrawable switch, forth & back
* TestTiledPrintingGearsAWT: Fix margin translate for high dpi, i.e. other ↵Sven Gothel2013-09-051-6/+11
| | | | than 72dpi
* Test scripts ..Sven Gothel2013-09-051-6/+6
|
* GLVBOArrayHandler: Remove unused importsSven Gothel2013-09-051-3/+0
|
* GLCanvas Printing WIP - Add TestTiledPrintingGearsAWTSven Gothel2013-09-053-31/+340
|
* TileRendererBase: API-doc cleanup; Fix GLCanvas, WIP printing ..Sven Gothel2013-09-052-68/+124
|
* TileRendererBase: Add interface TileRendererNotify, to notify implementing ↵Sven Gothel2013-09-057-29/+57
| | | | | | | | | | | | | GLEventListener about attached/detached TileRenderer .. since GLEventListener's reshape(..) method must query certain tile renderer attributes (tile pos and image size), they have to be aware of the TileRendererBase. To simplify such awareness and hence automate this attachement and passing over the tile renderer reference, they should implement this new interface. Gears example implements the new interface, which caches the TR reference and pauses rotation.
* TileRendererBase: Add toString(), Proper tile unit test names, i.e. add NEWT ↵Sven Gothel2013-09-057-19/+25
| | | | for UI agnostic (no-awt tests).
* TileRenderer*: Enhance API Doc; Cleanup OO; Remove PMVMatrixCallback, use ↵Sven Gothel2013-09-0510-408/+844
| | | | | | | | | | | | | GLEventListener reshape(..) or manual reshape after beginTile(..) method. GLEventListener reshape(..) method should be aware of TileRenderer usage and get the missing tile position and image size from it (-> see Gears example). TestRandomTiledRendering3GL2AWT demos AWT GLCanvas onscreen being used for random tile rendering to produce a PNG file. TestTiledRendering1GL2 is now GLAutoDrawable/GLEventListener agnostic, hence demos plain GLDrawable tile rendering usage.
* TileRenderer*: Align impl. to common abstract superclass TileRendererBase ↵Sven Gothel2013-09-046-393/+389
| | | | (cleanup / API doc)
* TileRenderer: Unify TileRenderer, add RandomTileRenderer (both utilizing ↵Sven Gothel2013-09-049-523/+1126
| | | | GLPixelBuffer, and pre/post GLEventListener)
* PNGImage: Bounds check in getPixelRGBA8(..) and setPixelRGBA8(..)Sven Gothel2013-09-041-0/+8
|
* GLPixelBuffer*: Add componentCount==1 (ALPHA/RED), 'allowRowStride' ↵Sven Gothel2013-09-047-80/+152
| | | | | | | | | | | | | | | | | utilization AWTGLPixelBuffer* -> GLPixelBuffer* GLPixelBufferProvider: - Default*.getAttributes(): Add componentCount==1 (ALPHA/RED), validate values, throw exception if n/a or not supported - Add 'allowRowStride' (as for AWTGLPixelBufferProvider) - Add default for true and false GLPixelBuffer: - Add 'allowRowStride' (as for AWTGLPixelBuffer) - Fix requiresNewBuffer(..): - aquire minByteSize if passed one is <= 0 - validate minByteSize w/ currentByteSize according to allowRowStride. AWTGLPixelBuffer: 'allowRowStride' impl. moved to GLPixelBuffer.
* Test scripts ..Sven Gothel2013-09-031-4/+5
|
* Cleaned up TiledRenderer capable to be used w/ GL2ES3 and TiledRenderer2 to ↵Sven Gothel2013-09-035-601/+883
| | | | | | | | | | | | | | be used w/ GLAutoDrawable. - Remove GL2 dependencies - Only requires PixelStorage ROW_LENGTH -> GL2ES3 - Position target buffer position according to skip [pixels, rows] - Use an interface PMVMatrixCallback, allowing user to reshape the custom 'PMV Matrix' according to the current rendered tile - Properly adjust tile/image buffer to written position and flip for read operation
* Gears: Make rotation optionalSven Gothel2013-09-031-2/+7
|
* GLReadBufferUtil.readPixels(..): Use plain int values for inWidth/inHeight, ↵Sven Gothel2013-09-032-12/+9
| | | | instead int[] - no return value desired.
* GLPixelStorageModes: Fix API doc references, Add ↵Sven Gothel2013-09-031-5/+39
| | | | set[Pack|Unpack]RowLength(GL2ES3 gl, ..)
* GLBuffers: Align commentSven Gothel2013-09-031-1/+1
|
* GL*, GLContext: Expose isGL2ES3() and getGL2ES3() for convenience.Sven Gothel2013-09-036-0/+59
|
* Fix Bug 810: Adding Julien Gouesse's fix while moved the new ↵Sven Gothel2013-09-023-42/+28
| | | | 'addNativeJarLibsJoglCfg(..)' to GlueGen, commit c0ead6fa10280f8076704726d59f482b183fd77e
* Revert commit 4beef4fe856690b070ba06a6caf4515aebd7171b manually for testing ↵Sven Gothel2013-09-021-2/+2
| | | | purposes .. (ATI fglrx driver issues)
* X11GLXDrawableFactory.Shutdown: Disable shared context destruction since it ↵Sven Gothel2013-09-023-8/+11
| | | | | | | | | may lead to a JVM freeze .. .. on ATI fglrx driver 32bit on 64bit w/ a frozen shared GL context involved. Hence we have to rely on the driver cleanup when JVM hits 'exit', equal to the Windows implementation.
* Animator/GLWindow: Catch 'ThreadDeath/Throwable' and dump info in DEBUG mode ↵Sven Gothel2013-09-024-7/+21
| | | | (cosmetic change only); Typo in comment; TestSharedContextListNEWT2: Stop animator.
* Add '@FixMethodOrder(MethodSorters.NAME_ASCENDING)' to all *Test* classes; ↵Sven Gothel2013-09-02219-7/+659
| | | | | | | | Relying on junit 4.11 + hamcrest 1.2 (deployed in Gluegen junit.jar) See gluegen commits: - 10a28c860bd3f4ccecef156c9bbbc88087bf568d - 70c9ba258c5ba3d78f0f52faf37d88b4c62d0c96
* FFMPEGMediaPlayer: Handle use-case of having [av|sw]resample lib, but not ↵Sven Gothel2013-09-013-6/+7
| | | | | | | compiled for it -> pass Scenario ffmpeg-0.10, where we are not prepared (compiled-in) for sw-resample support. Don't use if compiled in version (CC) is < 0 (n/a), and allow to pass at load time.
* ffmpeg_impl_template: Remove DEBUG line ..Sven Gothel2013-09-011-1/+1
|
* GLMediaPlayer: pause() -> pause(boolean flush): Allowing to flush buffers, ↵Sven Gothel2013-08-314-20/+38
| | | | next frame after play() will provide new frame. Added API doc.
* WWW: Remove 'Meet us @ SIGGRAPH 2013' ; Siggraph 2013 link to local doc page!Sven Gothel2013-08-311-2/+3
|
* GLContext: Add missing 'isGLESProfile()' and fix 'isGLcore()', i.e. GL2ES2 ↵Sven Gothel2013-08-311-1/+2
| | | | -> GLES2 (in profile set)
* FFMPEG/GLMediaPlayer: Fix compiler errors w/ new MingW 4.8.1: 'strsafe.h' -> ↵Sven Gothel2013-08-312-4/+10
| | | | don't use tchar.h; Fix compiler warning: Add missing (intptr_t) cast.
* Scripts/Windows: Test x32 and x64 builds w/ ffmpeg 2.x !Sven Gothel2013-08-314-10/+16
|
* Doc/HowToBuild: Windows x32 and x64 require new MingW64 Versions, tested w/ ↵Sven Gothel2013-08-311-11/+30
| | | | | | | | | | >= 4.8.1 - The previous MingW64 would still work, however we should align 32bit and 64bit versions - Mingw 32bit does not support dshow/com properly! - Use MingW64 32bit builds solve the problem
* GLWindow: Shorten API-doc referencesSven Gothel2013-08-301-4/+6
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-11-30 03:30:53 +0000