[EMAIL PROTECTED] writes: > HN> The regression test texts should explain what the > HN> correct result should look like, so you could tell something is wrong > HN> by comparing the examples and texts. This doesn't really tell anything. > > Hm, I think the most important thing is that we have examples covering all the > special cases - so that one can actually see a bug if he knows how the result > sohuld look. There are SO MANY special cases that the examples has to be so > big if they should all be covered...
The trick is to find a small set of orthogonal cases. For examples to demonstrate how things should be used, we the separate directory test/. The directory regression/ is for test-files: i.e. small files that each demonstrates one aspect that could easily break (and yes, there are one or two absurdly big tests in regression/, I believe for lyric alignment and broken repeats. They should be fixed some time.) -- Han-Wen Nienhuys | [EMAIL PROTECTED] | http://www.cs.uu.nl/~hanwen/ _______________________________________________ Lilypond-devel mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/lilypond-devel
