Convert images to beautiful ANSI escape codes for display in command line interfaces.
Available as both a CLI application and a Python library.
- Custom sized images
- ASCII or Unicode support
- Unicode enables 4x more detail
- 8/16/256/Truecolor support, for a wider gamut of colors
- Selectable system palettes to adjust for user terminal themes
- Fast color lookup with KDTrees & memoization
CLImage is available as both a standalone CLI program, or available for import as a Python3 library.
By default converting an image will output in 256 color, as 80 columns, and ASCII (for reasonable compatibility).
$ climage image.png
A nicer image can be obtained when enabling unicode and truecolor flags.
$ climage --unicode --truecolour image.png
For display in TTYs such as the Linux terminal before starting X11 (also accessible by Ctrl-Alt-F3, etc), you should restrict yourself to 8 color, and ASCII only. As 8 and 16 colors are solely constructed out of system colors (often set by the terminal theme), it is recommended to select a palette for these, depending on what you have selected.
$ climage --8color warhol.png --palette solarized
Further options may be found by running
It is recommended that you use the Python library if converting a large number of images. Usage of the library mirrors that of the CLI tool.
import climage output = climage.convert('image.png', is_unicode=True) print(output)
Two functions are exposed in the Python library
convert(filename, is_unicode=False, is_truecolor=False, is_256color=True, is_16color=False, is_8color=False, width=80, palette="default")
This will read in the file given by
filename, and return the output.
to_file(infile, outfile, is_unicode=False, is_truecolor=False, is_256color=True, is_16color=False, is_8color=False, width=80, palette="default"):
This is will in the file given by
filename, and write the output to
- write docstrings
- investigate different scaling modes? 256 color sometimes looks better for colors (than truecolor)?
- rename the
_toAnsifn, as it's not really right (ANSI is restricted to the 16 color stuff, right?)
- add a detect option to --palette, to automatically detect mapping of system colors? this might be hard.
- run the fabled pylint on the codebase
- Python library section link in setup.py, and in this README
- Add option to select