Dilettante Habits

Knowing nothing about everything

A library for parsing command line options

Abstract: A small library for parsing command line options.

Topics: software engineering/options, libraries, downloads, open source

© Copyright Daniel Krajzewicz, 07.08.2021 19:50, cc by


Command line applications are still a valid technology, especially for writing software that is meant to be processed in batch. Usually, each command line application needs some parameters or information about what has to be processed and how. There are some rules or common practices regarding the syntax for passing these parameters on the command line.

While working on the SUMO package I missed a proper object oriented solution in C++. Therefore, I have developed a library for handling of a command line application's options in a common way, including methods for things such as type checking or pretty printing a help screen. For simplicity, the libraries' name is “optionslib”. You may find it at github, the C++ version at optionslib_cpp, and the Java version at optionslib_java. It is meant to be a replacement for C's getopt function or Java's commons-cli. The library assumes that your options are split and given as the case on Linux/UNIX and works on Microsoft Windows command shell as well. The library is licensed under the Eclipse Public License — v 2.0.

In the following, you will find some information about standards, conventions and common practices for defining a command line application's parameters, first. Then, I will briefly outline why I designed this library. Then, the library itself is presented. Finally, you may find some further references as well as the list of changes.

Command Line Options

There are some conventions for command line options[1]:

The Issue

I have designed this library because of being annoyed by the getopt function that is usually used in applications written in C. When using getopt, your application has to deal with every single option one by one. You have to store them in a container or static (sic!) variables, do type checking for each option individually, maybe even multiple times, and stuff like that. It is not really nice.

The option is solving these issues by storing the given options in a single container. While being filled, it is checked whether the supported values can be converted into the type of the respective option. In addition, the library supports configuration files.

The library


The library should have the following features:

Design and Implementation Notes

The library uses a very basic object oriented approach. Every single option is represented by a single instance of a class that is derived from Option. The derived classes represent the realised types — Option_Bool, Option_Int, Option_Double, Option_String, Option_FileName. The options are stored in an OptionsCont — options container. Basically, that's all. Some further classes are given for parsing the application's parameters and storing them or for loading them from files.

C++ Options Library (optionslib_cpp)

The library is located at optionslib_cpp. The current release is v1.2:

You may as well find a verbose documentation of the optionslib_cpp library at github. A doxygen class documentation for optionslib_cpp exists as well.

C++ Options Library (optionslib_cpp)

The library is located at optionslib_java. The current release is v1.2:

You may as well find a verbose documentation of the optionslib_java library at github. A doxygen class documentation for optionslib_java exists as well.

Opportunities in Python

I personally use optparse. It comes with almost all the features that were put into requirements for this library.


I do not know any. If you have any issues, please let me know.

Possible Extensions

You may find current extension requests — and add own ones — at the respective issues page:


Both, the C++ as well as the Java versions are licensed under the Eclipse Public License v2.0.


In the following, you may find some further references to additional information.


Version 1.2 (06.08.2021)


Version 1.0 (08.10.2019)

A first version was released on the 8th of October 2019.

Creative Commons Attribution 4.0 International
Your message

Supporting an e-mail address allows us to answer.

Your message has been sent. Thank you.