rgbds

rgbds copied to clipboard

Improve performance of emitting data

4

Primarily, avoids reserving the max size for ROM sections' data. Also performs bulk reads. Fixes #1483

ISSOtm

Traverse the charmap trie to reconstruct mappings instead of saving them

5

There's a fundamental issue with this approach that makes me not want to use it: it loses the order in which charmaps were defined, instead outputting them in the ASCII/UTF-8...

Rangi42

Fixes #1483

Rangi42

Thanks to ISSOtm for first using cygpath in the rgbasm tests (#1472).

Rangi42

- The `#` component is renamed from "prefix" to "exact" - The `#` component for type `s` now escapes the string characters - Example: `"%#s"` given `"\"\t\\"` formats as `"\\"\\t\\\\"`...

Rangi42

Fixes #1485 In particular, a lot of the less clear interactions with other flags (e.g. the lack of `-u`, how should deduplication be performed, if at all?) have been made...

ISSOtm

This could be a `-i/--input-tileset` flag for RGBGFX. Previously discussed in https://github.com/gbdev/rgbds/issues/575#issuecomment-1991456862.

Rangi42

This could be a `-B/--background` flag for RGBGFX. If given `-B #fff`, it should match the `--remove-whitespace` behavior of pret's gfx.c postprocessor.

Rangi42

Implement `INCLUDE_ONCE` directive

1

Implements `INCLUDE_ONCE`, as suggested in #1478. This behaves like regular `include` the first time a file is included, except the name of the file is added to a list. The...

sukus21

Feature request: `#pragma once` or `include_once` equivalent

9

Include guards are a common problem in languages with include files. In some C/C++ compilers, the problem is remedied by supporting the `#pragma once` precompiler directive. I would like to...

sukus21