I find it hilarious that there's a ruby version of a tool that can be implemented in 10 lines of bash. It's like atwood's law all over again. Use the right tool for the job, please.
I'd be happy to see your 10 lines bash version of this ruby tool, with the same options/properties (debug messages, customizable prompt, history logging, same safeguards, rlwrap support, man page).
You're missing the point. I could do all of that in a similar (or likely smaller) number of lines with bash compared to the ruby tool, only mine would be much faster and wouldn't depend on the enormous ruby runtime.
the bash version might be technically faster, but i doubt it would be faster in any practical, meaningful way: most of the time spent in the script is waiting for input, or waiting for a command to return. that it need not depend on ruby is a perfectly valid argument though, and is in fact why i bothered redoing it in POSIX shell.
to do so however didn't just require a similar number of lines, it actually required 4 more (though to be fair, both scripts have comments and a fair bit of formatting, so it may not be a fair comparison. also it probably doesn't need stating, but i'm a bit of a hack, so my implementation maybe shouldn't be the reference point. but i digress)
in any event, i don't think atwood's law comes into play, given that a) this isn't in javascript and b) defunkt is a well known member of the ruby community who spends a lot of time working in ruby and probably wrote this to actually use it, as quickly as he could think of it. his choice of language here is perfectly reasonable in that context.
if you remain convinced that the only correct language for such a tool is bash/shell, i imagine OP certainly welcomes PRs; i know i do.
It's likely that with the feature set growing, a version that uses a slightly more advanced language would be easier to maintain and have less "traps" than the bash version...
