Skip to content


Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time


Build Status

This is simple Bash command-line parsing. No loops, no shifting, no worrying about quoting. Just declare some variables and call ghettopt:


ghettopt "$@" || exit

After calling ghettopt, the results can be found in the variables. For example, if the user specified --config=/dev/null then $opt_config now contains /dev/null.

Any non-option parameters on the command-line can be found in the array $params. If you'd like these back in $@, do it like this:

ghettopt "$@" || exit
set -- "${params[@]}"


There are four kinds of declarations that ghettopt understands:

  • strings, including blank: opt_string=default or opt_string=. These require an argument on the command-line.

  • booleans: opt_bool=true or opt_bool=false. These automatically get a corresponding --no- prefixed option. Specifying --option sets the variable to true; specifying the --no-option sets the variable to false, regardless whether the default was true or false.

  • arrays: opt_array=(). This accumulates when specified multiple times on the command-line, for example --array=one --array=two results in opt_array=( one two ).

  • functions, which are called: opt_func() { echo hi; }. To accept (require) an argument, use a colon: opt_func:() { echo "$1"; }. (It looks strange, but yes, colon is a valid character in a function name.)

Short options

To support short options, ghettopt looks at a variable called shortopts. This is a list of associations between short and long options. It's impossible to have a short option without a long option equivalent.

For example, to add short options to the example program:

shortopts=( c:config p:proxy f:force )

Results of parsing

The results can be found in three places: parsed option values are in the variables, non-options are in the array $params, and for reference you can find the normalized parsed options in the array $parsed_opts.

If ghettopt encounters a parsing error, such as an unrecognized option, it will emit an error message and return non-zero status.


  1. Does ghettopt provide automatic help?

    No, ghettopt does not provide automatic help. You can declare a function for it, though. Don't forget to exit at the bottom. See the full example below which does this.

  2. Is ghettopt pure Bash?

    Yes, except it depends on external GNU getopt. For maximum portability, for example to OS X, you can avoid this dependency by using pure-getopt. See the full example below which does this.

  3. What Bash versions does ghettopt support?

    ghettopt is compatible with Bash versions >= 2.05b

  4. Does ghettopt have a test suite?

    Yes, just clone the repo and run test.bash.

  5. How do you pronounce "ghettopt?"

    The "pt" is silent, so it's pronounced like "ghetto" or "get-o". That way it's not confused with "getopt" in conversation.

Full example

See example.bash for a full example with some best practices. Or use this function whenever you want to start a new script:

ghettopt-new() {
    [[ -n $1 ]] || { echo "need script name" >&2; return 1; }
    [[ ! -s $1 ]] || { echo "won't clobber $1, aborting" >&2; return 2; }
    curl -s > "$1"

Now you can initialize a new script with:

ghettopt-new myscript.bash