Skip to content
Guillaume Piolat edited this page Dec 18, 2018 · 17 revisions

Important: a lot of (non-binding) discussion is happening on the D Discord Server: https://discord.gg/fUHy4s

Current disagreements If you need ideas for proposals.

Image abstraction:

  • error codes vs exceptions for various components
  • stance on -betterC
  • TBD: what is a Color, the "pixel" type of an Image
    • and whether it's a compile-time or runtime thing
  • whether there should be several different Color format and type
    • and what they would be
  • whether the Image abstraction is 2D (N = 2), 3D-or-less (N <= 3) or N-dimensional (any N)
  • whether there can be an additional IAllocator in Image loaders for temporary buffers, used for decoding
  • what exactly owns IAllocator
  • who owns the Image abstraction itself (GC, RC, scoped ownership)
  • how animated Image would be represented

Image loaders:

  • how much Image loader are supposed to convert from one Color format to another
  • what happens when loading something with another colorspace as sRGB? should it auto-convert?
  • whether Image loaders should have short-paths for getting image dimensions
  • whether/how Image loaders should have short-paths for progressive downsampled images
  • a 2-layer architecture for image loading (format decoders + generic interface)
  • what meta-data is accessible in the outer layer

Solved disagreements:

Clone this wiki locally