drop sdl_net support and extract sdl_image/sdl_ttf/sdl_rotozoom from sdl egg

[felix winkelmann]

Installing the SDL egg requires all sorts of third-party libraries. The TTF/image/rotozoom stuff should go into separate eggs and the support for sdl_net should be dropped completely. To avoid backwards compatibility issues, we probably need a new SDL egg.

[Christian Kellermann]

Work to migrate is in progress, the plan is to separate the bits into sdl-base, sdl-img, sdl-gfx, sdl-ttf and sdl-net. The original sdl egg will stay as it is or alternatively be converted into a "meta-egg" that includes all sub components.

I am pretty much indifferent about throwing out sdl-net, is there a specific reason to do so? If it does not harm, why not leave it in, it's optional then anyway.

[felix winkelmann]

Replying to ckeen:

Work to migrate is in progress, the plan is to separate the bits into sdl-base, sdl-img, sdl-gfx, sdl-ttf and sdl-net. The original sdl egg will stay as it is or alternatively be converted into a "meta-egg" that includes all sub components.

That sounds good.

I am pretty much indifferent about throwing out sdl-net, is there a specific reason to do so? If it does not harm, why not leave it in, it's optional then anyway.

The wrappers need to be exported, even if they contain no functionality. I think this is not the best approach (Personally, I don't see the point in sdl-net support anyway. What's wrong with using the tcp unit?)

[Christian Kellermann]

Thanks to megane, sdl-base, sdl-img, sdl-ttf have been tagged 0.1
sdl-gfx has also been released as version 0.1

The old sdl egg will remain as it is, we may include a warning to deprecate it.

After having a look at the sdl-net API I will not provide bindings, as they are not in anyway more convenient than core functionality.

I am closing this ticket now.