Sefira library is meant to provide a well tested and reasonably optimized implementations of some tree comparison algorithms, so that they can be tried out in practical applications. Finding the largest common subtree of two ordered labeled trees is a problem with increasing applications (e.g. in XML processing and computational biology), on which substantial progress has been made in recent years. However, papers on tree comparison may not be available without subscription to the appropriate journals and even when they are, they generally don't contain executable code, so using the new algorithms is non-trivial. Also, there isn't one "best" algorithm yet. Distance definitions differ, and even for the same edit distance, time complexity can be measured on different tree features.
