summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* Test: More 'layouting' in NewtCanvasAWT test caseSven Gothel2011-10-291-0/+8
|
* Nativewindow/OSX: GetLocationOnScreen() must use [win frame] for window height.Sven Gothel2011-10-291-3/+7
|
* NEWT/Applet: Add 'newt.debug.Applet' ; Be more verbose in DEBUG Mode;Sven Gothel2011-10-292-2/+51
|
* Win32: Reuse Platform's OS VersionNumberSven Gothel2011-10-291-1/+1
|
* NEWT/OSX: Using NewtView softLock ; Adding debug info to solve assertion: ↵Sven Gothel2011-10-271-43/+86
| | | | | | | | | | | | | 'NSThemeFrame lockFocus' NSView.m:6053 - surfaceLock/Unlock: Use NewtView softLock - createWindow: - enable lock/unlockFocus to force realization of view - add retainCount in dbg-msg - clarify changeContentView
* NEWT/OSX NewtView: Add 'soft' pthread locking impacts: ↵Sven Gothel2011-10-272-14/+78
| | | | | | | | needsDisplay/displayIfNeeded, .. - Add 'soft' pthread locking (blockin) - impacts: needsDisplay/displayIfNeeded, .. - needsDisplay/displayIfNeeded also respects destroyNotifySend (ignore action if set)
* NEWT/OSX: Refine comment for native requestFocus(); Remove warningSven Gothel2011-10-261-2/+5
|
* NEWT/OSX: Can't use sending key/mouse directly to WindowImpl, deadlock .. ↵Sven Gothel2011-10-261-1/+7
| | | | see comment
* NEWT/OSX: Add missing sendKeyEvent(..) specialization to produce the keyCode ↵Sven Gothel2011-10-261-1/+10
| | | | (like enqueueKeyEvent())
* GLProfile: Add GLProfile getImpl(); Use getImpl() for getGL2ES[12](..), ↵Sven Gothel2011-10-262-72/+71
| | | | dropping GL2ES[12] lists
* OSX: Enable OpenGL profile support, ie OpenGL 3.2 -> GL3 for OSX >= 10.7Sven Gothel2011-10-265-152/+365
|
* Minor edits / cleanup of GLContext*Sven Gothel2011-10-266-26/+13
|
* MacOSX: Pull down (and fix releaseContext) NSOPENGL/CGL mode/impl, fixes SWT ↵Sven Gothel2011-10-2512-594/+499
| | | | usage and FBOMRT
* Tests: AWT EDT: invokeLater(..) -> invokeAndWait(..) for a more predictable ↵Sven Gothel2011-10-2310-72/+120
| | | | workflow
* Test: AWTRobotUtil TO 1s -> 2sSven Gothel2011-10-231-1/+1
|
* Partially revert cba9a070f9649bec42627631d393963d548e320c: Skip ↵Sven Gothel2011-10-232-23/+23
| | | | focusAction() and native focus request on X11/Windows. On both platforms it's not required (proper focus traversion) and AWT would deadlock on Windows
* NEWT/ScreenMode: Make getCurrenctScreenMode..() more fail proof in case the ↵Sven Gothel2011-10-232-15/+31
| | | | (native) impl. doesn't work.
* Newt/OSX: Disable VERBOSESven Gothel2011-10-221-1/+1
|
* Test/Focus: Add sleep for manual focus testsSven Gothel2011-10-222-1/+11
|
* Test/AWTRobotUtil: Proper titlebar location (using insets) ; Real UI/Robot ↵Sven Gothel2011-10-221-31/+37
| | | | requestFocus impl.
* NEWT/Native RequestFocus: Even if owning focus, run the focusAction() call ↵Sven Gothel2011-10-223-34/+35
| | | | incl. native focus request (X11, Windows, OSX)
* NewtCanvasAWT: Cleanup / Remove FIXME tag (impl. is ok)Sven Gothel2011-10-221-6/+2
|
* NEWT/Test FocusEventCountAdapter: Detailed test whether focus gained/lostSven Gothel2011-10-226-60/+79
|
* NEWT/MacOSX: Fix resignKeyWindow super call. Summarize ↵Sven Gothel2011-10-222-44/+15
| | | | sendFocusLost/sendFocusGained -> focusChanged
* Test/Ruler/ScreenMode: Validate whether current ScreenMode existsSven Gothel2011-10-221-1/+8
|
* NEWT/ScreenMode: If unimplemented have at least current mode in list. ↵Sven Gothel2011-10-227-119/+121
| | | | Current mode uses ScreenSize w/ dummy values if unimplemented.
* Test NEWT Remote GLWindow: Show VBO usageSven Gothel2011-10-211-19/+26
|
* Test: NEWT Remote GLWindow w/ GL2ES1 (GLSL n/a on remote GLX, eg. NV)Sven Gothel2011-10-212-15/+19
| | | | | | | - X11GLXContext: Better error message (requested / has) - Test: - Need to fetch remote GLProfile for GLCapabilities, since local machine differs - Can't use GLSL (GL2ES2) on remote .. n/a w/ NV
* Remove warningsSven Gothel2011-10-201-1/+5
|
* Using math functions in initializers fails on MacOSX - fixedSven Gothel2011-10-201-1/+3
|
* NEWT/OSX: Attempt to stabilize native parenting (ie w/ AWT) ; Add ↵Sven Gothel2011-10-204-100/+174
| | | | | | | | | | | | | | | | | windowRepaint() callback - act on positionChanged(..) only for realized windows, otherwise we could end in an AWT deadlock (AWT parent window). - add view parenting calls (addSubView/removeFromSuperview) - attachToParent after view configuration - allow concurrently view draw - add windowRepaint() callback (native -> java) - add more debug tracing
* Only revert ScreenMode (destroy/shutdown) if owner (Screen) changed itSven Gothel2011-10-193-11/+36
|
* Fix Solaris gcc args ; Cleanup dependencies and linker args; Adding JVM arch ↵Sven Gothel2011-10-182-4/+4
| | | | | | | | flags (-d32/-d64) ; glxext.h uses 'gl-64bit-types.h' Cleanup dependencies and linker args - JOGL/NEWT: remove Xxf86vm dependency - NativeWindow Solaris: Use std 'Xinerama' and 'xf86vmode'
* NEWT/OSX: Adding 'fake' lockSurface based on [view canDraw]. Can't use ↵Sven Gothel2011-10-162-6/+53
| | | | lockFocus(), deadlocks since we render separate to focus mgmt.
* OSX: Fix CGL deleteContext()'s release call on MainThread (typo, shall have ↵Sven Gothel2011-10-161-2/+2
| | | | no ':' for void func call)
* NEWT/OSX: Send events direct w/o EDT queue (like X11 and Win32)Sven Gothel2011-10-161-10/+28
|
* NEWT/OSX: Proper impl. of NEWT's focus management (fixes NEWT/AWT focus ↵Sven Gothel2011-10-165-34/+96
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | behavior/tests) - Old code was just requesting the focus and made the window upfront and notifying a gained focus to WindowImpl. (hack) - Using proper requestFocus impl. issuing focusAction() and utilizing native focus gained/lost messages. This distinguish between 'makeKey' and 'orderFront'. Also requesting and accepting (view) first responder role, which is a precursor to proper gained/lost focus handling on OSX. - NEWTCanvasAWT: Adding 'steal AWT focus': +++ void requestFocus() { super.requestFocus(); // AWT < steal AWT focus > NEWTChild.requestFocus() } +++ Helps make the focus traversal between NEWT/AWT more reliable. Happend on OSX that AWT (NewtCanvasAWT instance) didn't release the focus after NEWT child gained the same. We are not able to use the 'focusAction()' here (disabled in this code path) due to AWT-EDT blocking and recursive focus changes. The latter is also intendend to request the AWT focus first .. - AWT/NEWT focus test 01 passes on OSX
* NEWT/OSX: exec. makeKeyAndOrderFront/makeKeyWindow on MainThreadSven Gothel2011-10-161-5/+6
|
* Test GearsES2: Split confine mode to centered 'J' and non centered 'j', to ↵Sven Gothel2011-10-162-5/+25
| | | | test actual confinement
* NEWT/OSX: Add mouse enter/exit and pointer features (visible, confined, ↵Sven Gothel2011-10-165-98/+299
| | | | warp) - cleanup pos/coord translations
* NEWT/Mouse: Skip 'move' event w/ same position. Add Enter/Exit eventsSven Gothel2011-10-154-7/+107
| | | | | | | | | | Skip 'mouse move' event w/ same position - On Windows, the OS sends us multiple event w/o change in position, suppress them Add Enter/Exit events incl. synthesize 'enter' event for windows/osx - X11: using native Enter/Leave events - Windows: using native Leave event (tracking) and synthesized enter event - OSX: TODO (required for the confined feature, etc)
* NEWT/Pointer Confined: Dispatch mouse move events before enabled; Only ↵Sven Gothel2011-10-152-29/+40
| | | | | | | | | | | | | | | | | | | request-focus/warp-ptr if enabled Dispatch mouse move events before enabled - Allows user app listener to track to the new centered mouse position before using the confined position. This is important for position change usage. See simplified demo GearsES2 Only request-focus/warp-ptr if enabled - No need to request focus and center mouse if leaving confinement Demo GearsES2: - No need to assume some position changes are erroneous (jumps) due to confinement. - Track unconfined mouse position, allowing confined navigation to have the proper position change value
* Fix NEWT/Pointer Invisible: INVISIBLE_MASK was equal w/ CONFINED_MASK (duh)Sven Gothel2011-10-151-1/+1
|
* NEWT/EDTUtil: Allow get/set poll period, defaults to defaultEDTPollPeriodSven Gothel2011-10-153-5/+32
|
* OSX: Avoid invoking JNI or performSelectorOnMainThread in JNI if already ↵Sven Gothel2011-10-132-2/+6
| | | | mainThread
* OSX/SWT: Adding OSXUtil: RunOnMainThread(), IsMainThread() / Utilizing those ↵Sven Gothel2011-10-136-34/+224
| | | | | | | | | | | | | | for SWT access/calls Adding OSXUtil: RunOnMainThread(), IsMainThread() - Issuing a native call where the user Runnable is to be performed on the main thread - Enable query if we are on the main thread. Utilizing those for SWT access/calls - Using the above to call all SWT functions on the main thread if required (incomplete) TODO/Issues: - JOGL OSX CGL Context fails, ie expecting NS, but having CGL
* MacOSX: Always release ctx on main thread (offscreen was frozen as well); ↵Sven Gothel2011-10-1310-69/+33
| | | | | | Cleanup Warnings and OSX ContextUpdater -
* MacOSX: Fix shared ctx release [onMainThread]; Make GLContextShareSet ↵Sven Gothel2011-10-136-212/+202
| | | | | | | | | | | | | | | | | lifecycle deterministic; Remove warnings Fix shared ctx release [onMainThread] - Releasing the shared contexts caused a freeze of about 10s from one of the shared release operations. [NSOpenGLContext release] - Thorough triage concluded the workaround to release the shared ctx on the main thread. - Using enhanced GLContextShareSet, see below Make GLContextShareSet lifecycle deterministic - Programmatically control the lifecycle of tracked shared ctx allows us using 'hard' references. - Features queries for isShared() and ofc unregister a share set if all are destroyed. Remove warnings - MacOSXWindowSystemInterface.m used 'long', where 'GLint' was requested.
* TestParenting03AWT: Add manual code to disable 2nd GLWindow for internal ↵Sven Gothel2011-10-131-21/+38
| | | | testing (def.: enabled)
* NEWT/OSX: Fix resize behavior / Cleanup coordinate transormation ↵Sven Gothel2011-10-133-51/+86
| | | | | | | | | | | | (client-space/top-level, child/parent) At resizing a perent window w/ a NEWT OSX child, the window position needs to be updated since it's absolute. Re-adding sending *Changed notifications via the appropriate WindowImpl methods. Turns out they are missing in some parent/child situations (fullscreen for example). Native getLocationOnScreen0(..) queries totalHeight by it's own to have correct values.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-23 20:25:48 +0000