innix / github-avatar-generator

A GitHub-esque avatar image generator written in Python.

GitHub Avatar Generator

A GitHub-esque avatar image generator written in Python.

Installation

1. Install the Pillow package:

   pip install Pillow
   

2. Clone repository.

Usage

An example script is provided to demonstrate usage. The example creates a single image and saves it to a file.

python example.py <canvas_size> <block_count> <output_filename>

Parameter	Summary
canvas_size	The size (width and height) of the image in pixels.
block_count	The number of colored blocks to use in the image.
output_filename	The output image file path.

For example:

$ python example.py 420 12 avatar_01.png

License

You may use GitHub Avatar Generator under the terms of the MIT License (see LICENSE).

github-avatar-generator's Issues

The number of blocks in an avatar is sometimes wrong

Sometimes the generated avatars will have extra or missing blocks in them (e.g. 6 blocks when you asked for 5, or 5 blocks when you asked for 6).

Block placement is RNG so reproducing may take several runs. Run this a few times and eventually you will get an image with more or less than than 6 blocks:

$ python3 example.py 420 6 avatar_6_blocks.png

It is not isolated to even numbers, using an odd number 5 and the same issue will be seen:

$ python3 example.py 420 5 avatar_5_blocks.png

The problem is caused by how the generator creates vertically symmetrical images and dealing with "central" blocks that don't need an opposing block to ensure symmetry. The generator needs to be smarter about where to place blocks to ensure the avatar has the correct number of blocks but still maintains the vertical symmetry.

Infinite loop if too many blocks are requested

If you try to create an avatar with more blocks than can be fitted on the image, the avatar generator will loop forever.

Reproducible example:

$  python3 example.py 420 32 avatar_01.png

The generator should detect if too many blocks have been asked for, and raise an exception.

Use requirements.txt to manage dependencies

Instead of a list of pip commands in the README for documenting dependencies, we should use a requirements.txt file instead. The README should also be updated with new instructions on installing the library.