TSPSG is intended to generate and solve "travelling salesman problem" (TSP) tasks. It uses the Branch and Bound method for solving. Its input is a number of cities and a matrix of city-to-city travel costs. The matrix can be populated with random values in a given range (which is useful for generating tasks). The result is an optimal route, its price, step-by-step matrices of solving, and a solving graph. The task can be saved in an internal binary format and opened later. The result can be printed or saved as PDF, HTML, or ODF. TSPSG may be useful for teachers to generate test tasks or just for regular users to solve TSPs. Also, it may be used as an example of using the Branch and Bound method to solve a particular task.
W-Meter is a tool for testing the conformance and evaluating the performance of 802.11 products. The tool is meant for injecting arbitrary frames to test the conformance of any wifi product. The tool has templates for generating various kinds of frames which can be customized for generation of specific frames. The tool can be used to elicit responses from other wifi cards to check the conformance of the card to the protocol. At present, 802.11g frames are completed.
crpcut is the Compartmented Robust Posix C++ Unit Test system. crpcut (pronounced "Crap Cut") runs all test cases in their own process and their own working directory, which makes it perfectly normal to test that asserts do trap, and the test suite continues even in the event of an unexpected SIGSEGV. By using the C++11 features long available in GCC, the tests are extremely easy to write.
xCover is a code coverage library for C and C++. It uses non-standard features available with GCC 4.3+ and Visual C++ 7.0+. Users place line-marks in each branch of each function or method in a component's source file, and the library is able to produce, upon request, a report of which of these has not been executed. Reporting can be done on a per-file, per-alias, or per-file-group basis.
mutest is a micro unit testing framework for C (with some C++ support). It is easy to use and has no dependencies. The idea is simple: a source file is a test suite, and a function is a test case (special functions can be used for test suite initialization and termination) that can can have several checks. Checks comes in 2 flavors: one that only prints an error, and one that terminates the current test case too. A (normally) automated test program run all the test suites and prints some stats. It fails (returns non-zero) if any test suite fails.