[cairo] Cairo SkyOS backend

[Robert Szeleney]

Hi!

Does anyone have an answer to these questions?

skyos at liwest.at wrote:
> Hi!
>
> I'm in the middle of making a SkyOS backend for cairo-1.2.0 which works quite nice already. (at least according to the cairo test suite).
>
> Anyway, I hope you can answer me a few technical questions regarding the recommended communication between cairo and the native window system.
>
> 1.
> Using a "cairo_skyos_surface_create" function the SkyOS backend establishes a "connection" between a cairo surface and a native window. (Similar to the other backends).
>
> The cairo_skyos_surface_create allocates a memory buffer with the same size as the current window content and copies the current window content into this buffer. 
> When now cairo draws into this buffer, how is the content supposed to get "on screen" again? (SkyOS internally double buffers all drawing functions, so as long as no GI_SwapWindow is called nothing will be displayed). Is this call to the swap function supposed to be somewhere in the cairo/SkyOS backend or is the application using cairo responsible for handling this?
>
> For instance, in order to make the cairo-test.c display content on the screen I had to insert a GI_SwapWindow right before terminating the test/releasing the surface. 
>
> 2. 
> What happens if the native window system draws into the same window which is "in use" by cairo currently? Is this allowed at all, by definition?
>
>
> 3. Color space
> Does the native windowing system has to handle all color spaces (RGB16,RGB24,RGB32, ARGB32, ....) or does cairo implicit "conversions" if the backend doesn't support a specific color depth?
>
>
> Anyway, with the GI_SwapWindow workaround the testsuite works fine, no test failed. and indeed, its rather fast, and very cool to see all the graphic effects while the test is running.
>
> Hope someone can help me with my questions,
>    Thanks,
>     Robert!
> _______________________________________________
> cairo mailing list
> cairo at cairographics.org
> http://cairographics.org/cgi-bin/mailman/listinfo/cairo
>
>
>