Older

• Aug 12, 2009

  About (A brief description of the Nebula3 project.) Wiki page commented on by lazalong   -   The last commit was on 19 oct 08 but flohofwoe blog mention a lot of changes. So what is the relation between both? Is this google repository the official one from Radon coders? Or is it simply a copy-paste?

  The last commit was on 19 oct 08 but flohofwoe blog mention a lot of changes. So what is the relation between both? Is this google repository the official one from Radon coders? Or is it simply a copy-paste?
• Aug 11, 2009

  issue 14 (debug version's result will be wrong if there is a gamepad c...) reported by xoyojank   -   What steps will reproduce the problem? 1. compile a debug version of testviewer 2. connect a gamepad to the computer 3. run testviewer.debug.exe What is the expected output? What do you see instead? what rendering is not the same to the release version, and cannot operate by mouse What version of the product are you using? On what operating system? Nebula3 SDK April 2009 Please provide any additional information below. RenderApplication::RenderApplication() : //time(0.0), //frameTime(0.0), quitRequested(false) { // empty } those two member variables were not inited!!!

  What steps will reproduce the problem? 1. compile a debug version of testviewer 2. connect a gamepad to the computer 3. run testviewer.debug.exe What is the expected output? What do you see instead? what rendering is not the same to the release version, and cannot operate by mouse What version of the product are you using? On what operating system? Nebula3 SDK April 2009 Please provide any additional information below. RenderApplication::RenderApplication() : //time(0.0), //frameTime(0.0), quitRequested(false) { // empty } those two member variables were not inited!!!
• Aug 03, 2009

  issue 13 (Window style is wrong for window mode) reported by xoyojank   -   DisplayMode D3D9DisplayDevice::ComputeAdjustedWindowRect() { ... AdjustWindowRect(&r, this->windowedStyle, 0); ... } and the position of the DisplayMode may be negative

  DisplayMode D3D9DisplayDevice::ComputeAdjustedWindowRect() { ... AdjustWindowRect(&r, this->windowedStyle, 0); ... } and the position of the DisplayMode may be negative
• Dec 14, 2008

  issue 12 (the DisplayDevice's DisplayMode is not correct) commented on by xoyojank   -   this bug will also cause nVidia's PerfHUD could not be operated by mouse, because the coordinate is wrong

  this bug will also cause nVidia's PerfHUD could not be operated by mouse, because the coordinate is wrong
• Dec 14, 2008

  issue 12 (the DisplayDevice's DisplayMode is not correct) reported by xoyojank   -   the DisplayMode's width & height is ought to the Window's client width & height, instead of Window's width & height. Therefore, the render target size is wrong, so the text/image is not so clear. What version of the product are you using? On what operating system? Nebula3 SDK Sep 2008 Please provide any additional information below. //------------------------------------------------------------------------- ----- /** Open the application window. */ bool Win32DisplayDevice::OpenWindow() { ....... // the rendering area is not the whole window RECT rect; GetClientRect(this->hWnd, &rect); this->displayMode.SetWidth(rect.right); this->displayMode.SetHeight(rect.bottom); return true; }

  the DisplayMode's width & height is ought to the Window's client width & height, instead of Window's width & height. Therefore, the render target size is wrong, so the text/image is not so clear. What version of the product are you using? On what operating system? Nebula3 SDK Sep 2008 Please provide any additional information below. //------------------------------------------------------------------------- ----- /** Open the application window. */ bool Win32DisplayDevice::OpenWindow() { ....... // the rendering area is not the whole window RECT rect; GetClientRect(this->hWnd, &rect); this->displayMode.SetWidth(rect.right); this->displayMode.SetHeight(rect.bottom); return true; }
• Dec 14, 2008

  issue 11 (D3D9Texture::Unmap will assert) reported by xoyojank   -   What steps will reproduce the problem? 1. D3D9Texture::Map() 2. D3D9Texture::Unmap() What is the expected output? What do you see instead? What version of the product are you using? On what operating system? Nebula3 SDK Sep 2008 Please provide any additional information below. this->GetD3D9Texture()->UnlockRect(mipLevel); now the mapCount is not 0, so this->GetD3D9Texture() will assert

  What steps will reproduce the problem? 1. D3D9Texture::Map() 2. D3D9Texture::Unmap() What is the expected output? What do you see instead? What version of the product are you using? On what operating system? Nebula3 SDK Sep 2008 Please provide any additional information below. this->GetD3D9Texture()->UnlockRect(mipLevel); now the mapCount is not 0, so this->GetD3D9Texture() will assert
• Dec 03, 2008

  issue 7 (DisplayDeivce should have a method for tool application's wi...) commented on by xoyojank   -   that is similar to what i do. hope those code would be merged into the trunk. thank u ^_^

  that is similar to what i do. hope those code would be merged into the trunk. thank u ^_^
• Nov 30, 2008

  issue 7 (DisplayDeivce should have a method for tool application's wi...) commented on by hwpeterkim   -   See the "BrowserPlugin" under "branches". And find SetParentWindow function.(I hope this is correct name of that) You might find proper implementation of what you want to do. Cheers, -Kim

  See the "BrowserPlugin" under "branches". And find SetParentWindow function.(I hope this is correct name of that) You might find proper implementation of what you want to do. Cheers, -Kim
• Nov 29, 2008

  issue 8 (XACTAudioDevice may have a momery leak) commented on by xoyojank   -   pGlobalSettingsBuffer Buffer containing the global file data. This file is generated when you build an XACT project file. It contains category, variable, and RPC information. Xbox 360: Memory for this member must be allocated using XMemAlloc when specifying the XACT_FLAG_API_CREATE_MANAGEDATA flag. Windows: Memory for this member must be allocated using CoTaskMemAlloc when specifying the XACT_FLAG_API_CREATE_MANAGEDATA flag.

  pGlobalSettingsBuffer Buffer containing the global file data. This file is generated when you build an XACT project file. It contains category, variable, and RPC information. Xbox 360: Memory for this member must be allocated using XMemAlloc when specifying the XACT_FLAG_API_CREATE_MANAGEDATA flag. Windows: Memory for this member must be allocated using CoTaskMemAlloc when specifying the XACT_FLAG_API_CREATE_MANAGEDATA flag.
• Nov 28, 2008

  issue 10 (There may be a delay on 3D Cue's transform information to ta...) reported by xoyojank   -   What steps will reproduce the problem? 1. build a cue which has a distance attenuation with XACT(DXSDK's Tool) 2. Setup the N3's Audio System 3. Play that cue at a distance(the sound volume should be lower) What is the expected output? What do you see instead? at the start of playing, the sound effect has no distance attenuation, but after about 0.1 second, this attenuation takes place. What version of the product are you using? On what operating system? Nebula3 SDK (Sep 2008), Microsoft DirectX SDK (March 2008) Please provide any additional information below. the 3d effect won't take place before xactEngine->DoWork(), so you should update the engine when the cue starts playing: Ptr<Cue> AudioDeviceBase::PlayCue(const CueId& cueId, bool play, bool enable3D, const Math::matrix44& transform, const Math::vector& velocity) { n_assert(this->IsOpen()); Ptr<SoundBank> soundBank; IndexT cueIndex = this->FindCue(cueId, soundBank); if (InvalidIndex != cueIndex) { // create and setup a new cue instance Ptr<Cue> newCue = Cue::Create(); newCue->Setup(soundBank, cueIndex, play, enable3D); this->activeCues.Append(newCue); if (enable3D) { newCue->SetTransform(transform); newCue->SetVelocity(velocity); this->OnFrame(); } return newCue; } else { n_error("PlayCue(): cue '%s' not found in any loaded sound bank!", cueId.Value().AsCharPtr()); return Ptr<Cue>(); } }

  What steps will reproduce the problem? 1. build a cue which has a distance attenuation with XACT(DXSDK's Tool) 2. Setup the N3's Audio System 3. Play that cue at a distance(the sound volume should be lower) What is the expected output? What do you see instead? at the start of playing, the sound effect has no distance attenuation, but after about 0.1 second, this attenuation takes place. What version of the product are you using? On what operating system? Nebula3 SDK (Sep 2008), Microsoft DirectX SDK (March 2008) Please provide any additional information below. the 3d effect won't take place before xactEngine->DoWork(), so you should update the engine when the cue starts playing: Ptr<Cue> AudioDeviceBase::PlayCue(const CueId& cueId, bool play, bool enable3D, const Math::matrix44& transform, const Math::vector& velocity) { n_assert(this->IsOpen()); Ptr<SoundBank> soundBank; IndexT cueIndex = this->FindCue(cueId, soundBank); if (InvalidIndex != cueIndex) { // create and setup a new cue instance Ptr<Cue> newCue = Cue::Create(); newCue->Setup(soundBank, cueIndex, play, enable3D); this->activeCues.Append(newCue); if (enable3D) { newCue->SetTransform(transform); newCue->SetVelocity(velocity); this->OnFrame(); } return newCue; } else { n_error("PlayCue(): cue '%s' not found in any loaded sound bank!", cueId.Value().AsCharPtr()); return Ptr<Cue>(); } }
• Nov 26, 2008

  issue 9 (The Audio System's "Wave" has no 3D effect) reported by xoyojank   -   What steps will reproduce the problem? 1. AudioPlayer::PlayWave3D() What is the expected output? What do you see instead? because the XACT does not support 3D wave, it only supports 3d cue... What version of the product are you using? On what operating system? Nebula3 SDK (Sep 2008)

  What steps will reproduce the problem? 1. AudioPlayer::PlayWave3D() What is the expected output? What do you see instead? because the XACT does not support 3D wave, it only supports 3d cue... What version of the product are you using? On what operating system? Nebula3 SDK (Sep 2008)
• Nov 25, 2008

  issue 8 (XACTAudioDevice may have a momery leak) reported by xoyojank   -   What steps will reproduce the problem? this method is not called by Close() XACTAudioDevice::UnloadGlobalSettings() What version of the product are you using? On what operating system? Nebula3 SDK (Sep 2008) Please provide any additional information below.

  What steps will reproduce the problem? this method is not called by Close() XACTAudioDevice::UnloadGlobalSettings() What version of the product are you using? On what operating system? Nebula3 SDK (Sep 2008) Please provide any additional information below.
• Nov 22, 2008

  issue 7 (DisplayDeivce should have a method for tool application's wi...) reported by xoyojank   -   I want to use n3 as the base library of a tool application, which is build with MFC/CLR. But I have to change the code of Win32DisplayDevice to render things in another window in MFC/CLR. The InputServer also do not provide friendly interface to use child window's handle(HWND). although i can edit the code by myself , i think n3 should have those code, then someone can write many tools for n3. ^_^

  I want to use n3 as the base library of a tool application, which is build with MFC/CLR. But I have to change the code of Win32DisplayDevice to render things in another window in MFC/CLR. The InputServer also do not provide friendly interface to use child window's handle(HWND). although i can edit the code by myself , i think n3 should have those code, then someone can write many tools for n3. ^_^
• Oct 26, 2008

  issue 6 (Bug] PSSMutil's CameraFrustum FOV is doubled.) reported by amugana   -   in PSSMUtil::CalculateFrustumPoints cameraNormWidth and cameraNormHeight are doubled, thus it makes so humble tiny shadow. thanks.

  in PSSMUtil::CalculateFrustumPoints cameraNormWidth and cameraNormHeight are doubled, thus it makes so humble tiny shadow. thanks.
• Oct 19, 2008

  r53 (Added the CMake build file for the texture batcher tool, now...) committed by vadim.macagon   -   Added the CMake build file for the texture batcher tool, now that I have DevIL building properly.

  Added the CMake build file for the texture batcher tool, now that I have DevIL building properly.
• Oct 19, 2008

  r52 (Added the IL and ILU source that gets built into a DevIL lib...) committed by vadim.macagon   -   Added the IL and ILU source that gets built into a DevIL library used by the texture batcher tool. Note that I've disabled support for any formats (like png and jpeg) that require additional libraries, if you wish to add support for those format modify config.h in the DevIL source, tweak the CMake build file and rebuild. I may add the source for the png and jpeg libraries in the future to eliminate the extra step.

  Added the IL and ILU source that gets built into a DevIL library used by the texture batcher tool. Note that I've disabled support for any formats (like png and jpeg) that require additional libraries, if you wish to add support for those format modify config.h in the DevIL source, tweak the CMake build file and rebuild. I may add the source for the png and jpeg libraries in the future to eliminate the extra step.
• Oct 19, 2008

  r51 (Added CMake build files for most of the toolkit code with th...) committed by vadim.macagon   -   Added CMake build files for most of the toolkit code with the exception of the texture batcher which relies on DevIL, and DevIL is proving to be a real pain to build statically so I'm still working on it.

  Added CMake build files for most of the toolkit code with the exception of the texture batcher which relies on DevIL, and DevIL is proving to be a real pain to build statically so I'm still working on it.
• Oct 18, 2008

  r50 (Added CMake build files for foundation, render, application,...) committed by vadim.macagon   -   Added CMake build files for foundation, render, application, addons and extlibs, I've only tested on VC 2005. CMake supports more platforms than Radon Labs custom TCL-based build system, or my Python based N2 build system.

  Added CMake build files for foundation, render, application, addons and extlibs, I've only tested on VC 2005. CMake supports more platforms than Radon Labs custom TCL-based build system, or my Python based N2 build system.
• Oct 18, 2008

  r49 (Fixed typo in filename.) committed by vadim.macagon   -   Fixed typo in filename.

  Fixed typo in filename.
• Oct 18, 2008

  r48 (Some fixes to RakNet to make it build with _HAS_EXCEPTIONS=0...) committed by vadim.macagon   -   Some fixes to RakNet to make it build with _HAS_EXCEPTIONS=0 defined (an undocumented/unsupported MSVC feature that's used by the other Nebula3 libs). Also properly disabled the deprecation warnings for unlink and mkdir (there was a typo, the warning number is 4996, not 4966).

  Some fixes to RakNet to make it build with _HAS_EXCEPTIONS=0 defined (an undocumented/unsupported MSVC feature that's used by the other Nebula3 libs). Also properly disabled the deprecation warnings for unlink and mkdir (there was a typo, the warning number is 4996, not 4966).
• Oct 18, 2008

  r47 (Added the source for the RakNet library so it can be built w...) committed by vadim.macagon   -   Added the source for the RakNet library so it can be built with the same compiler/linker options as the rest of Nebula3.

  Added the source for the RakNet library so it can be built with the same compiler/linker options as the rest of Nebula3.
• Oct 18, 2008

  r46 (Merged the September2008Merge branch into Trunk.) committed by vadim.macagon   -   Merged the September2008Merge branch into Trunk.

  Merged the September2008Merge branch into Trunk.
• Oct 16, 2008

  r45 (Change how we implement singletons on OS X since it doesn't ...) committed by bruce.mitchener   -   Change how we implement singletons on OS X since it doesn't support a thread-local qualifier, so we need to use pthread_set/getspecific and friends. This appears to work to make thread-specific singleton instances work. It also required changing the __ConstructSingleton and __DestructSingleton macros to have corresponding versions __ConstructInterfaceSingleton and __DestructInterfaceSingleton.

  Change how we implement singletons on OS X since it doesn't support a thread-local qualifier, so we need to use pthread_set/getspecific and friends. This appears to work to make thread-specific singleton instances work. It also required changing the __ConstructSingleton and __DestructSingleton macros to have corresponding versions __ConstructInterfaceSingleton and __DestructInterfaceSingleton.
• Oct 16, 2008

  r44 (More work in progress on the porting branch after the merge....) committed by bruce.mitchener   -   More work in progress on the porting branch after the merge. Everything that used to compile and link before does again. There are crashes at runtime still and I've been stubbing out the memory management rather than actually fixing it.

  More work in progress on the porting branch after the merge. Everything that used to compile and link before does again. There are crashes at runtime still and I've been stubbing out the memory management rather than actually fixing it.
• Oct 16, 2008

  r43 (More merging of the September 2008 release from Radon Labs. ) committed by bruce.mitchener   -   More merging of the September 2008 release from Radon Labs.

  More merging of the September 2008 release from Radon Labs.
• Oct 16, 2008

  r42 (Work in progress on merging the September 2008 release into ...) committed by bruce.mitchener   -   Work in progress on merging the September 2008 release into the porting branch. Foundation doesn't entirely compile yet.

  Work in progress on merging the September 2008 release into the porting branch. Foundation doesn't entirely compile yet.
• Oct 15, 2008

  r41 (Fixes to portability work from Haron. ) committed by bruce.mitchener   -   Fixes to portability work from Haron.

  Fixes to portability work from Haron.
• Oct 11, 2008

  r40 (Second part of the merge of the Nebula3 September 2008 SDK, ...) committed by vadim.macagon   -   Second part of the merge of the Nebula3 September 2008 SDK, these should be the last of the code changes. I haven't committed most of the resources (in the work, export, export_win32 directories/archives) due to their size, you'll have to grab them from the SDK for now, likewise with the RakNet static libs.

  Second part of the merge of the Nebula3 September 2008 SDK, these should be the last of the code changes. I haven't committed most of the resources (in the work, export, export_win32 directories/archives) due to their size, you'll have to grab them from the SDK for now, likewise with the RakNet static libs.
• Oct 08, 2008

  r39 (First part of the merge of the Nebula3 September 2008 SDK, o...) committed by vadim.macagon   -   First part of the merge of the Nebula3 September 2008 SDK, only 'foundation' will build for now, the rest will follow in the next couple of days.

  First part of the merge of the Nebula3 September 2008 SDK, only 'foundation' will build for now, the rest will follow in the next couple of days.
• Sep 30, 2008

  r38 (Branch for merging the September 2008 drop of source from Ra...) committed by bruce.mitchener   -   Branch for merging the September 2008 drop of source from Radon Labs.

  Branch for merging the September 2008 drop of source from Radon Labs.
• Sep 20, 2008

  issue 5 (volume texture code bug) reported by xoyojank   -   d3d9texture.cc, line 249 ============================================================== void D3D9Texture::SetupFromD3D9VolumeTexture(IDirect3DVolumeTexture9* texVolume) { n_assert(0 != texVolume); this->d3d9VolumeTexture = texVolume; HRESULT hr = texVolume->QueryInterface(IID_IDirect3DBaseTexture9, (void**) &this->d3d9VolumeTexture); =============================================================== should changed to: HRESULT hr = texVolume->QueryInterface(IID_IDirect3DBaseTexture9, (void**) &this->d3d9BaseTexture);

  d3d9texture.cc, line 249 ============================================================== void D3D9Texture::SetupFromD3D9VolumeTexture(IDirect3DVolumeTexture9* texVolume) { n_assert(0 != texVolume); this->d3d9VolumeTexture = texVolume; HRESULT hr = texVolume->QueryInterface(IID_IDirect3DBaseTexture9, (void**) &this->d3d9VolumeTexture); =============================================================== should changed to: HRESULT hr = texVolume->QueryInterface(IID_IDirect3DBaseTexture9, (void**) &this->d3d9BaseTexture);
• Sep 20, 2008

  issue 4 (the device lost is not handled cerrectly) reported by xoyojank   -   What steps will reproduce the problem? 1. run testgame.exe 2. Win+L to lock Windows 3. the D3D9RenderDevice::Present() will fail the n_assert will exit the program, so i change the code to: ================================================================= if (0 != D3D9DisplayDevice::Instance()->GetHwnd()) { HRESULT hr = this->d3d9Device->Present(NULL, NULL, 0, NULL); n_assert(SUCCEEDED(hr) || D3DERR_DEVICELOST == hr); } ================================================================= but the num of device lost handlers is zero, so the Reset() would always return FAIL! so it renders nothing. ================================================================= else if (D3DERR_DEVICENOTRESET == hr) { // notify event handlers that the device was lost this->NotifyEventHandlers(RenderEvent(RenderEvent::DeviceLost)); ================================================================= I switch to DEBUG version of DirectX, and get some information. I think the the render target should recreate when the device is lost

  What steps will reproduce the problem? 1. run testgame.exe 2. Win+L to lock Windows 3. the D3D9RenderDevice::Present() will fail the n_assert will exit the program, so i change the code to: ================================================================= if (0 != D3D9DisplayDevice::Instance()->GetHwnd()) { HRESULT hr = this->d3d9Device->Present(NULL, NULL, 0, NULL); n_assert(SUCCEEDED(hr) || D3DERR_DEVICELOST == hr); } ================================================================= but the num of device lost handlers is zero, so the Reset() would always return FAIL! so it renders nothing. ================================================================= else if (D3DERR_DEVICENOTRESET == hr) { // notify event handlers that the device was lost this->NotifyEventHandlers(RenderEvent(RenderEvent::DeviceLost)); ================================================================= I switch to DEBUG version of DirectX, and get some information. I think the the render target should recreate when the device is lost
• Sep 14, 2008

  About (A brief description of the Nebula3 project.) Wiki page edited by vadim.macagon
• Sep 14, 2008

  r36 (Reverted changes to Win32ConsoleHandler because they broke n...) committed by vadim.macagon   -   Reverted changes to Win32ConsoleHandler because they broke nsh3 by making Console::GetInput() non- blocking, this caused nsh3 to print the prompt character '>' to the console window over and over again. Also removed a duplicate include from renderapplication.h.

  Reverted changes to Win32ConsoleHandler because they broke nsh3 by making Console::GetInput() non- blocking, this caused nsh3 to print the prompt character '>' to the console window over and over again. Also removed a duplicate include from renderapplication.h.
• Sep 12, 2008

  About (A brief description of the Nebula3 project.) Wiki page edited by vadim.macagon
• Sep 12, 2008

  About (A brief description of the Nebula3 project.) Wiki page added by vadim.macagon
• Sep 12, 2008

  r33 (Merged in floh's code from last month. Notable changes inclu...) committed by vadim.macagon   -   Merged in floh's code from last month. Notable changes include: - Dynamic memory allocation is now done using various specialized heaps. - Rendering is now done in a separate dedicated thread. - N2 particle system has been removed (but there is no replacement as yet). - Character animation subsystem has undergone some work, but doesn't appear to be complete yet. - Vertex component packing. A nice overview of the major changes can be found on floh's blog. Unfortunately the vertex component packing changes are not backwards compatible with existing mesh resources, so meshes need to be re-exported before they can be loaded again. Since we don't have the updated resources as yet, all the Nebula apps that display pretty things on screen currently crash on startup. We hope to rectify the situation in the near future.

  Merged in floh's code from last month. Notable changes include: - Dynamic memory allocation is now done using various specialized heaps. - Rendering is now done in a separate dedicated thread. - N2 particle system has been removed (but there is no replacement as yet). - Character animation subsystem has undergone some work, but doesn't appear to be complete yet. - Vertex component packing. A nice overview of the major changes can be found on floh's blog. Unfortunately the vertex component packing changes are not backwards compatible with existing mesh resources, so meshes need to be re-exported before they can be loaded again. Since we don't have the updated resources as yet, all the Nebula apps that display pretty things on screen currently crash on startup. We hope to rectify the situation in the near future.
• Aug 20, 2008

  issue 2 (warning C4819 (code page problems of non-unicode source file...) commented on by xoyojank   -   some characters are displaying as "'?", you can change them to "'?'"

  some characters are displaying as "'?", you can change them to "'?'"
• Aug 16, 2008

  issue 3 (=> VS8.0: idlc.sln ERROR: target 'idlc' not defined!) reported by shkim5   -   What steps will reproduce the problem? 1. cd code 2. tclsh update.tcl 3. output: :: GENERATING... ... ... -> VS8.0: benchmarks_win32.benchmarkbase_win32.vcproj -> VS8.0: benchmarks_win32.benchmarkfoundation_win32.vcproj -> VS8.0: benchmarks_win32.benchmarkaddon_win32.vcproj => VS8.0: idlc.sln ERROR: target 'idlc' not defined! What is the expected output? What do you see instead? no error expected. What version of the product are you using? On what operating system? Vista x64, VS2005 SP1, ActiveState ActiveTcl 8.5.3.0 Please provide any additional information below. I don't know well about Tcl, I can't edit the generation script. Please check this error message. Thanks.

  What steps will reproduce the problem? 1. cd code 2. tclsh update.tcl 3. output: :: GENERATING... ... ... -> VS8.0: benchmarks_win32.benchmarkbase_win32.vcproj -> VS8.0: benchmarks_win32.benchmarkfoundation_win32.vcproj -> VS8.0: benchmarks_win32.benchmarkaddon_win32.vcproj => VS8.0: idlc.sln ERROR: target 'idlc' not defined! What is the expected output? What do you see instead? no error expected. What version of the product are you using? On what operating system? Vista x64, VS2005 SP1, ActiveState ActiveTcl 8.5.3.0 Please provide any additional information below. I don't know well about Tcl, I can't edit the generation script. Please check this error message. Thanks.
• Aug 16, 2008

  issue 2 (warning C4819 (code page problems of non-unicode source file...) commented on by shkim5   -   I still got an error from /code/tests/testfoundation/binaryreaderwritertest.cc.

  I still got an error from /code/tests/testfoundation/binaryreaderwritertest.cc.
• Aug 15, 2008

  issue 2 (warning C4819 (code page problems of non-unicode source file...) commented on by amugana   -   in /bin/makesInWin32.tcl turn "WarnAsError" off then, update.tcl again..

  in /bin/makesInWin32.tcl turn "WarnAsError" off then, update.tcl again..
• Aug 14, 2008

  r32 (Branch for merging new code from Radon Labs. ) committed by bruce.mitchener   -   Branch for merging new code from Radon Labs.

  Branch for merging new code from Radon Labs.
• Aug 11, 2008

  issue 2 (warning C4819 (code page problems of non-unicode source file...) reported by shkim5   -   What steps will reproduce the problem? 1. In my case, I use Korean Windows whose code page for MBCS is 949. 2. just type on cmd shell: 'msbuild compile.msbuild' 3. I got many C4819 warning and other errors. What is the expected output? What do you see instead? no compile error expected :) What version of the product are you using? On what operating system? Windows Vista Ultimate x64 Korean Ed, Visual Studio 2005 SP1 Please provide any additional information below. Please do not use any special characters which is not compatible with ANSI or 7bit charset.

  What steps will reproduce the problem? 1. In my case, I use Korean Windows whose code page for MBCS is 949. 2. just type on cmd shell: 'msbuild compile.msbuild' 3. I got many C4819 warning and other errors. What is the expected output? What do you see instead? no compile error expected :) What version of the product are you using? On what operating system? Windows Vista Ultimate x64 Korean Ed, Visual Studio 2005 SP1 Please provide any additional information below. Please do not use any special characters which is not compatible with ANSI or 7bit charset.
• Aug 09, 2008

  r31 (Changed the mime type used by the Gecko plugin from applicat...) committed by vadim.macagon   -   Changed the mime type used by the Gecko plugin from application/nebula3 to application/x-nebula3 since unregistered mime types are supposed to be prefixed with "-x". The IE plugin already uses application/x-nebula3.

  Changed the mime type used by the Gecko plugin from application/nebula3 to application/x-nebula3 since unregistered mime types are supposed to be prefixed with "-x". The IE plugin already uses application/x-nebula3.
• Aug 08, 2008

  r30 (Initial commit of the Nebula3 ActiveX Control aka Internet E...) committed by vadim.macagon   -   Initial commit of the Nebula3 ActiveX Control aka Internet Explorer 6+ plugin that allows running Nebula3 applications within the IE window. To test, just build the neb3ac project and open up test.htm in IE, no additional setup required.

  Initial commit of the Nebula3 ActiveX Control aka Internet Explorer 6+ plugin that allows running Nebula3 applications within the IE window. To test, just build the neb3ac project and open up test.htm in IE, no additional setup required.
• Aug 05, 2008

  r29 (Delay launching Nebula until the size of the parent window i...) committed by vadim.macagon   -   Delay launching Nebula until the size of the parent window is known.

  Delay launching Nebula until the size of the parent window is known.
• Aug 04, 2008

  r28 (Fixed flickering of embedded Nebula3 window during page scro...) committed by vadim.macagon   -   Fixed flickering of embedded Nebula3 window during page scrolling. Rapid scrolling (hold the scrollbar and move it up and down very quickly and repeatedly) can still result in weird shifting of the embedded window... can't make Google's Lively break in the same way though so there's got to be a way to fix it!

  Fixed flickering of embedded Nebula3 window during page scrolling. Rapid scrolling (hold the scrollbar and move it up and down very quickly and repeatedly) can still result in weird shifting of the embedded window... can't make Google's Lively break in the same way though so there's got to be a way to fix it!
• Aug 04, 2008

  r27 (Fixed issues with sizing and positioning of the Nebula windo...) committed by vadim.macagon   -   Fixed issues with sizing and positioning of the Nebula window when switching between embedded and windowed modes.

  Fixed issues with sizing and positioning of the Nebula window when switching between embedded and windowed modes.
• Aug 03, 2008

  r26 (Implemented ability to switch between child window and popup...) committed by vadim.macagon   -   Implemented ability to switch between child window and popup window display modes at runtime. Use the 'P' key to switch between the two modes. This feature is still a work in progress, the window positioning and sizing when switching modes is one of the more obvious things that still need to be fixed up.

  Implemented ability to switch between child window and popup window display modes at runtime. Use the 'P' key to switch between the two modes. This feature is still a work in progress, the window positioning and sizing when switching modes is one of the more obvious things that still need to be fixed up.
• Aug 03, 2008

  r25 (The win32 input server will now unacquire the DirectInput mo...) committed by vadim.macagon   -   The win32 input server will now unacquire the DirectInput mouse device when input focus is lost, if Nebula is running in a child window.

  The win32 input server will now unacquire the DirectInput mouse device when input focus is lost, if Nebula is running in a child window.