[webkit-dev] WebGL :: Win32 => build successful and rendering assumed so => but white page

[Brent Fulgham]

Hi Hugo,

I think most of us in the WebKit project are aware of the missing pieces on the Apple Windows port, and have plans to rectify this.

However, I don’t understand what your needs are regarding WebGL and the Apple Windows port. This port is not licensed for use outside of Apple’s own products, and only exists as a mechanism for outside developers to debug problems and run early-warning builds of our current sources.

If you would like to embed a Windows version of WebKit for use in your own software, I would suggest using the WinCairo port, which seeks to be feature-compatible with the Apple Windows port.  An added bonus is that WebGL *already exists and works* in this port.

Alex Christensen has been working to further expand the WinCairo port, and often hangs out on our IRC channel.

Thanks,

-Brent

On Oct 3, 2013, at 1:49 PM, Hugo Machefer <hugo.machefer at gmail.com> wrote:

> Thanks a lot Timothy. For some: part of my job might consist in doing this work  -for Windows-  required to actually display the result. Then, assuming <WebKit-r156532> compiles and runs WebGL content succesfully  for Mac  (probably provided some settings :  'defaults write com.apple.Safari WebKitWebGLEnabled -bool YES'  specified from the terminal) would that be a good approach, according to you, to watch step-by-step what happens across Mac OS, by using efficient debugger, and report what is missing   -for Windows-   ? To fill the gap...
> 
> Do you or anyone assess this analogy to mimic behaviours around OpenGLES2 from Mac OS X to WIN32 (involving Angle in-the-middle ;-) ?
> 
> PS: again, sorry for any piece of naivety or groundless indolence in my analysis: these technical areas turn out to be new to me.
> 
> On Thu, Oct 3, 2013 at 6:15 PM, Tim Horton <timothy_horton at apple.com> wrote:
> On 2013.10.03, at 09:06, Hugo Machefer <hugo.machefer at gmail.com> wrote:
> 
>> Dear all. Might not the best place  -here-  to describe my issue [Win32], but I don't know where exactly this shall be submitted, as a novice by WebKit. To recap, I managed to compile <WebKit-r156532> sources by Microsoft Visual Studio 2010 using CoreGraphics (ie without Cairo) in Debug mode. To activate WebGL, additionally, I have set: 
> 
> The AppleWin port (the one you describe) does not currently support WebGL. Much of the underlying mechanism is implemented, but there’s still some work required to actually display the result.
> 
>> #define ENABLE_WEBGL 1 
>> #define WTF_USE_3D_GRAPHICS 1 
>> #define WTF_USE_OPENGL 1 
>> #define WTF_USE_OPENGL_ES_2 1 
>> #define WTF_USE_EGL 1 
>> #define WTF_USE_GRAPHICS_SURFACE 1
>> 
>> I hope/guess this is the right approach => could anyone confirm ? Then LINK turns out to be "fortunate", by adding these libraries : translator_common.lib; libEGL.lib; libGLESv2.lib; preprocessor.lib; translator_glsl.lib; translator_hlsl.lib to WebKit sub-project.
>> 
>> However, execution fails by http://get.webgl.org. Indeed, HTML content pretends : 
>> Your browser supports WebGL
>> However: a blank area is displayed in place of expected movic 3D based cube... 
>> 
>> Apologize in case this concern forsooth comes to break any rule by lists.webkit.org , due to Windows matters...
>> 
>> _______________________________________________
>> webkit-dev mailing list
>> webkit-dev at lists.webkit.org
>> https://lists.webkit.org/mailman/listinfo/webkit-dev
> 
> 
> _______________________________________________
> webkit-dev mailing list
> webkit-dev at lists.webkit.org
> https://lists.webkit.org/mailman/listinfo/webkit-dev

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-dev/attachments/20131003/cfe56031/attachment.html>