bitedutils

bited-utils

Pipeline helpers and utilities for building fonts from bited BDFs. Built with Go and Nix flakes.

• bited-build: A command that generates vector (TTF, WOFF2) and bitmap (BDF, PCF, OTB, DFONT) fonts from bited BDFs. Supports integer scaling and Nerd Font patching.
• bited-img: A command that generates image specimens for bited BDFs. Features Base16 color support.
• bited-clr: A TUI tool for fine-tuned coloring of TXT/CLR pairs.
• bited-scale: A command for scaling bited BDFs.

Built with bited-utils

kirsch

ANAKRON

Nix flake template

In a blank directory where you want your font project to reside, run:

nix flake init --template github:molarmanful/bited-utils

This will create an annotated template project with Nix scaffolding around bited-utils to help get your font project up and running.

TXT/CLR files

To generate colorful images, bited-img accepts pairs of .txt and .clr files. For example, if you have a file test.txt that you wish to color, you would include test.clr containing your desired color codes in the same directory.

A .clr file is really just lines of spaces and color codes. The color codes are as follows:

| Code | Definition | | ------------ | --------------------------- | | 0-9, A-F | Set color to a Base16 color | | . | Reset to foreground color |

All other characters are treated as no-ops and simply pass on the current color.

Codes are interpreted left-to-right, top-to-bottom. The position of a code matches a position in the TXT file where you wish to color. Say you have the following TXT:

Hello, world!
Testing Testing 123

And the following CLR:

A    . D    .
5       6       BCE

Would produce:

Why TXT/CLR?

When designing the pipeline for creating color specimens for my fonts (what would eventually become bited-img), I initially considered several existing approaches to creating the specimen sources:

• Writing Pango markup directly, which would make passing them to ImageMagick super easy.
• Creating ANSI via an editor and then generating the Pango markup.

Both of these approaches suffer from a rather important flaw, however: because they embed color data alongside content in the same file, their source files are difficult to read and/or write in a text editor. So my caveman solution was to just separate the color data into its own file, which preserves a readable TXT source.

My workflow for creating these TXT/CLR files goes something like:

1. Create the TXT file with uncolored contents.
2. Copy the TXT file to CLR.
3. Block out the CLR file with color codes. Neovim's visual block mode is especially well-suited for this.
4. Remove extraneous text and use bited-clr for finer adjustments if necessary.