mpr finds malloc/realloc memory leaks and memory allocation statistics by using a simple, brute force strategy: log all malloc/free calls to a file and post-process this log file when the program terminates. mpr keeps track of the entire call chain leading up to a malloc/free call. This is superior to conventional methods that keep track of only the immediate caller using __FILE__ and __LINE__ pre-processor tricks.
HL7lib is a project that will provide a free, correct implementation of Health Level 7 functions. Health Level 7 is commonly used in large hospitals to send patient information among computer systems from different vendors. Since there is no reference implementation of HL7 many of these vendor systems vary widely in their interpretation of HL7.
LibKMid is a library that implements features to play MIDI events on MIDI devices. It also includes easy-to-use functions if all you want is play MIDI files in your application (for example, background music in a game), and a console MIDI player. The API is well documented, and the design is very simple for the most common uses.
Mondriaan produces a set of images that look a lot like the paintings of the artist Piet Mondriaan. By way of selection and genetic operators this set can be manipulated by you. You can try to find images that you think look even more like Mondriaan art, or search for images you find attractive.
libwayne is a library of reasonably abstract datatypes and algorithms written in C. They have a consistent user interface. Many of the algorithms are taken verbatim from textbooks, but are coded well. The library contains data structures and code for binary trees, stacks, queues, combinatorial analysis, bit strings, graphs, dynamic arrays, event-driven simulation, priority queues, about half a dozen different O(n log n) sorting algorithms, ODE integration, long long integer I/O, matrix and vector operations, random number generation, search algorithms, and elementary statistics.
LK is an implementation of the Lin-Kernighan heuristic for the Traveling Salesman Problem and for the minimum weight perfect matching problem. It is tuned for 2-d geometric instances, and has been applied to certain instances with up to a million cities. Also included are instance generators and Perl scripts for munging TSPLIB instances. <P>This implementation introduces ``efficient cluster compensation'', an experimental algorithmic technique intended to make the Lin-Kernighan heuristic more robust in the face of clustered data.