The usage of Harry is covered in a classic manual page (man page), including command line options, configuration files and different operation modes.
Harry is developed in plain C. Harry's functionality for comparing strings is organized in different modules that are documented using Doxygen annotation. Although Harry should not be directly used as a library, this reference might help integrating Harry with other software frameworks.
Harry programming reference
Reference for modules and data structures
(Automatically generated from source code)
The following articles provide an overview of Harry and common similarity measures for strings. Harry supports edit distances and several string kernels; however, bag-of-words models are not supported and should be computed using Sally.
Harry: A Tool for Measuring String Similarity
Konrad Rieck and Christian Wressnegger.
Journal of Machine Learning Research, 17(9):1−5, March 2016.
Similarity Measures for Sequential Data
WIREs: Data Mining and Knowledge Discovery, 1 (4) 296-304, July 2011.