While quite nice as it is, there is always room for improvement. I've been thinking on features that I should/could add.
- Ability to use images from an RGBA Source for fonts. Including a .png library of some sort would help with that.
- Auto-scanning routine to at least make the base .kerning file. It may still need to be hand tweaked before deployment.
- Have more scanning areas besides our current three, should a font warrant it, or to have LESS scanning areas for those cases where you are using a monospace or non-overlapping font.
- Include a development version of the library that supports loading the two separate .pcx and .kerning files. Possibly make the 'converter' part of the object.
- Support for a larger character set, hopefully the full byte range. I'd not try for anything fancier than that, though UTF-8 support would be kind-of cool.
- LZW or other encoding of the proprietary format's pixel buffer.
- Add a toggle to decide if mipmaps or a single image map is to be used - Possibly even include the ability to specify your own custom mipmaps.
If you have any ideas that might help make this better, or comments/corrections/bugfixes/questions/flames, please feel free to fill out my Contact Form and let me know.