Skip to content

Latest commit

 

History

History
73 lines (46 loc) · 5.44 KB

README.md

File metadata and controls

73 lines (46 loc) · 5.44 KB
Raw

drawing_hands

Example output of drawing_hands

Installation

  1. Download the Ubuntu Font Family.
  2. Unzip the Ubuntu Font Family within the assets/ directory so that assets/ubuntu-font-family-0.83/UbuntuMono-R.ttf is a valid path.
  3. Install FIGlet. My command line prompted me with a variety of commands to use to install this package/executable.
  4. Install vim. I doubt you don't already have it installed.

Execution

The easiest command to execute the full program is cargo run -- -r. I recommend reading the Explanation section for better context on the usage parameters.

main.rs

Usage: drawing_hands [OPTIONS]

Options:

  • -f, --font-path <FONT_PATH> (default: ./assets/ubuntu-font-family-0.83/UbuntuMono-R.ttf). This works best as a mono-spaced font, where each character is as close to twice as tall as it is wide as possible. When encoding the font into the literal, each printable ASCII character of the font is converted into a 16x32 greyscale image, with 6 bits of luminance/alpha.
  • -i, --image-path <IMAGE_PATH> (default: ./assets/Drawing_Hands_3x2.png). The image to convert into ASCII art for the big literal.
  • -e, --example-name <EXAMPLE_NAME> (default: drawing_hands). The output program will be created as ./examples/<EXAMPLE_NAME>.rs.
  • -t, --tab-size <TAB_SIZE> (default: 4). This will be the tab size that drawing_hands.rs has in its TAB_SIZE constant, used to keep track of positions of non-whitespace text following tab characters.
  • -r, --run. This automatically runs drawing_hands.rs after generating it.
  • -h, --help. Print help.

drawing_hands.rs

Usage: drawing_hands [OPTIONS]

Options:

  • -i, --input-text-path <INPUT_TEXT_PATH> (default: ./examples/drawing_hands.rs). The path to a text-based file to open up in vim.
  • -o, --output-image-path <OUTPUT_IMAGE_PATH> (default: ./output/drawing_hands.jpg). The output image file to produce.
  • -w, --min-pixel-width <MIN_PIXEL_WIDTH> (default: 0). The minimum pixel width of the output image file.
  • -h, --min-pixel-height <MIN_PIXEL_HEIGHT> (default: 0). The minimum pixel height of the output image file.
  • -t, --tab-size <TAB_SIZE> (default: 4). The tab size to use to keep track of positions of non-whitespace text following tab characters.
  • -d, --debug. This will produce some debug files containing the (text, position), (text, color), and (text, position, color) ordered tuple lists.
  • -s, --skip-print. Used in conjunction with -d, this will skip the image printing process. This is useful for iterating on code that's used to produce the list of (text, position, color) ordered trios.
  • --help. Print help.

Background

I wanted to create a piece of art that could, to some extent, encapsulate the beauty that I see in code.

I drew inspiration from a few works, including

Explanation

This project consists of three programs, main.rs, content.rs, and drawing_hands.rs. Together, they can produce a file that can print an image of itself as rendered in vim at a resolution high enough for a 24"x32" poster.

main.rs

main takes an input font .ttf file, an input image, and an output example name. Using the font data and the input image, main constructs a large &[u8] literal that encodes alpha bitmaps for printable ASCII characters, with a large ASCII-art representation of the input image in the center. It then combines this literal with the existing contents of content.rs to make the third program, which will carry the example name (drawing_hands by default). Inserted at the top of this third program is a signature of the project, produced by FIGlet.

content.rs

content takes an input text path and an output image path. content opens up the input text file in vim to construct a list of (text, color) ordered pairs. By opening up the file itself and keeping track of whitespace characters, content also constructs a list of (text, position) ordered pairs. Due to some vim shenanigans, "zippering" these two lists is more difficult than one might think, but eventually content has a list of (text, color, position) ordered trios. With these, and the font alpha bitmaps decoded from the literal, content then outputs an image of the text file, colored as it would be in the user's vim settings. The key piece missing here is that if content tries to decode its own literal, it fails, as it's literal definition is just const BYTES: &[u8] = br"";. That's where the third program, drawing_hands.rs, comes in.

drawing_hands.rs

With all the functional code of content.rs, albeit it condensed into a much less readable format, plus a valid BYTES literal, drawing_hands is actually capable of producing the image output that content wishes it could.

Shortcomings

  • My testing of this project leaves a lot to be desired. I'm sure there are some vim settings that make this fail in mysterious ways. I have tested removing FIGlet and verifying that the fallback text prints correctly.
  • This is limited to printing files composed entirely of ASCII text.