Rewrite RGBGFX in C++ #981

Currently missing from the old version: - `-f` ("fixing" the input image to be indexed) - `-m` (the code for detecting mirrored tiles is missing, but all of the "plumbing" is otherwise there) - `-C` - `-d` - `-x` (though I need to check the exact functionality the old one has) - Also the man page is still a draft and needs to be fleshed out More planned features are not implemented yet either: - Explicit palette spec - Better error messages, also error "images" - Better 8x16 support, as well as other "dedup unit" sizes - Support for arbitrary number of palettes & colors per palette - Other output formats (for example, a "full" palette map for "streaming" use cases like gb-open-world) - Quantization? Some things may also be bugged: - Transparency support - Tile offsets (not exposed yet) - Tile counts per bank (not exposed yet) ...and performance remains to be checked. We need to set up some tests, honestly.

Be consistent with what we settled on in code

And also the inverted alpha channel condition causing the wrong colors to be treated as transparent

Force version.c to be compiled as C++ (bodge, will need a proper fix for `version.c`) Remove user-defined `ProtoPalette` assignment operator (same as default, anyway)

That's what it actually is, oops (Required for `minmax_element`)

The conversion is OK because of the tile amount cap

Conversion may be unspecified, but Windows better do the right thing (or else.)

We are about to set all of the pixels anyway

As explained by the comment

Also allow counting an `AssignedProtos`'s number of proto-palettes

Describe current options, and add some TODOs for functionality to be implemented

Should fix compilation with GCC before 9

This fixes many whitespace issues

Support all intended formats and allow partial parsing <charconv> is not available in GCC 7

GCC 7 fails to deduce the contained type

Duh

Not available in libstdc++ 7, apparently

Fixes build on some compilers for some reason I fail to understand

MSVC's (broken) ABI breaks otherwise. What the f@!$#ck, Microsoft? (Thank you based Clang for warning, by the way.)

Though none of the code involved has been tested in any capacity yet, lol

And also some ASCII art, perhaps?

Rename a poorly-named attribute, and add a bunch of debug logging

This notably caused decantation to delete non-empty palettes, which crashes

Though none of this has been tested so far...

Co-Authored-By: Rangi <[email protected]>

For reproducibility

Ensure that the color count is properly used, and that transparency is not counted as a color when packing palettes

More consistent with its "main" function's name

Useful for persisting flags outside of the build system

Apply review comments

Not hooked to all RGBGFX flags yet, but good enough for most use cases (and as a base for future development, should I need to `reset --hard`.) TODOs marked appropriately.

Alter the return status accordingly, and print the error counts on exit

PSP, ACT, and ACO are complete

As documented, oops

This is now described in its own section

Silences some static analysis warning

Including removal of a dead & useless check. By the way, fuck integer promotion.

Thanks @GreenAndEievui!

That check was when the image width was in tiles, and another check right below is the correct equivalent for the new pixel-unit argument

It's tool-specific, so categorize it where it belongs

Which of the two is erroring out can be inferred from the file name, anyway.

It gets overwritten right after, and uncovers an error in the interlaced read

Either we let libpng handle the transform, or we don't. But make up your mind!

Which otherwise trips a later assertion in debug mode (phew!) and crashes in release mode (oops)

Co-Authored-By: ISSOtm <[email protected]>

The sorting was performed without updating the mappings, which broke the world. We can instead sort the IDs as they are inserted into the packing queue, which should also be faster than moving the actual proto-pal objects around.

Teeny tiny

Some disjoint sets were mistakenly reported not as such For example, {0} was considered to include {1}.

…tilemap

Compare CGB colors only, including ignoring RGB components if alpha says the color is transparent

A black+white palette would turn into white+transparent, removing black pixels from the palette (the third slot would be empty, and the 4th slot be set directly)

These offsets should only be applied to a tile ID read as input... but this ain't one!

No such tests yet, but the infrastructure will be there.

This is not intended to test libpng as much as checking that we behave correctly if libpng gives us an error

Simplifies iterating over tiles and attributes at the same time

Should improve performance. This version is cooler, and also does not suffer from iteration limits

Less duplication = good

As it turns out, it is really difficult to implement, and can be dealt with later.

Also correct minor blunders in the man page

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Rewrite RGBGFX in C++ #981

Rewrite RGBGFX in C++ #981

Commits on Jun 12, 2022

Commits on Jul 2, 2022