On Sat, 29 Oct 2016 at 20:32 Urs Liska <u...@openlilylib.org> wrote: > > That's what I thought, and that's of course a good thing. But would it be > conceivable to actually start doing unit tests? One should probably not be > frightened by the issue that we won't be able to apply that backwardly, to > the existing code. > > Urs > > It's an excellent idea Urs; and I for one feel very uneasy writing code without a test framework.
I think the problem is, how would you define an assertion, and what are you intending to test (i.e. what's the Subject Under Test?). I think something like Lilypond might require some quite elaborate test fixtures / fakes. There are probably C++ functions you could test with a C++ assert based test framework - Catch for example - or something simpler. Obviously anything doing a bit of maths is easy to test; the interesting functions tend to manipulate "Grob" objects; would the tests be examining their properties after the function call? But what about Scheme code?... Chris _______________________________________________ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
