Nice! I am using some of the funny characters! Some remarks: I doubt if it is needed to launch a sub-process for each markup directive. This is fine for small

Nice! I am using some of the funny characters! Some remarks: I doubt if it is needed to launch a sub-process for each markup directive. This is fine for small output, but will slow down processing of larger chunks of text. I would be interested in a bash solution based on associative arrays. As an example, one might want to look at the zsh implementation of the colors script in the main distribution[0], which makes colors available as an associative array: # Set once in shell init script: autoload -U colors && colors # And use it anywhere: print $fg[blue]hello colorful world!$fg[reset] [0]: http://sourceforge.net/p/zsh/code/ci/master/tree/Functions/Misc/colors http://sourceforge.net/p/zsh/code/ci/master/tree/Functions/M...