cute_aseprite: continued support

[mathewmariani]

This pull requests adds support for color modes grayscale and indexed color modes, as well as, support for tilesets and tilemaps

struct ase_frame_t now has a field void* pixels; instead of a ase_color_t* pixels to reflect how the specs defined a pixel. We can cast to our desired color types (struct ase_color_t, struct ase_grayscale_t, and struct ase_index) as needed.

A pixel as described in the specs:

* `PIXEL`: One pixel, depending on the image pixel format:
  - **RGBA**: `BYTE[4]`, each pixel have 4 bytes in this order Red, Green, Blue, Alpha.
  - **Grayscale**: `BYTE[2]`, each pixel have 2 bytes in the order Value, Alpha.
  - **Indexed**: `BYTE`, each pixel uses 1 byte (the index).

Because blending is not supported for indexed color the blending step is skipped for routes where ase->mode == ASE_MODE_INDEXED.

I also added support chunk 0x2023; tilesets and tilemaps through struct ase_tileset_t and struct ase_cel_tilemap_t respectively.

An example of loading and reading the pixels of a tileset:

ase_t *ase = cute_aseprite_load_from_file(path, NULL);
ase_tileset_t ase_tileset = ase->tileset;

// `struct ase_color_t`, `struct ase_grayscale_t`, or `struct ase_index`
const ase_color_t *pixels = (ase_color_t *) ase_tileset.pixels;

An example of loading and reading the tiles from a tilemap:

ase_t *ase = cute_aseprite_load_from_file(path, NULL);

for (int i = 0; i < ase->frame_count; ++i) {
  ase_frame_t *frame = ase->frames + i;
  for (int j = 0; j < frame->cel_count; ++j) {
    ase_cel_t *cel = frame->cels + j;
    if (cel->is_tilemap) {
      memcpy(tilemap, cel->tiles, size);
    }
  }
}

cute_aseprite_free(ase);

[RandyGaul]

Put your name in the docs credits somewhere!

[mathewmariani]

Put your name in the docs credits somewhere!

Many thanks ! I added credits along with the date.

[RobLoach]

I tried this out, and while I haven't dove into the root cause, I think there's a regression in how the pixels are loaded? By just replacing the old cute_aseprite with this one in pntr_aseprite, it aseprite character no longer displays.

[mathewmariani]

I tried this out, and while I haven't dove into the root cause, I think there's a regression in how the pixels are loaded? By just replacing the old cute_aseprite with this one in pntr_aseprite, it aseprite character no longer displays.

Thank you @RobLoach. Ill take a look at it later this week

[RobLoach]

Same with this one too:
https://github.com/RobLoach/raylib-aseprite

Again, I'm unsure what goes wrong with it, could likely dive deeper on it next week.

[RandyGaul]

I think to merge this in it would be much easier to separate into smaller and different PRs. Suggestion: One for tileset, one for tilemap (if it's different enough from tileset, I haven't looked and don't know), one for greyscale/indexed. It's risky to adjust code and takes a lot of time/effort to test, so smaller more atomic PR really help save me time. Otherwise I don't feel like we can merge it in as-is.

[mathewmariani]

I have broken up the pull request starting with indexed and grayscale color mode #423.

[RandyGaul]

Closing this for now since we're breaking it up into smaller PRs.