From: Tom B. (Tehom) <te...@pa...> - 2013-12-04 23:06:27
|
Some odds and ends and bugfixes in the latest commit * Tested it with notation times way off from performance times, as Chris wisely suggested. Verdict: No apparent new problems and only old problems that we can live with. * Tested it with slivers and tiny notes. A few bugfixes in the new commit. * Collapsing notes and rests now removes hard spaces (I'll post something about hard spaces) On notation times way off from performance times: We can get crazy note-lengths. That's because it goes according to the metric events that the notation-times told it. If they lie, we can get (say) no metric events in a bar that actually has a note in it, resulting in a crazy-looking note durations. They do gradually get more honest as more notes are inserted, since we won't insert a note into a meterpiece that it doesn't really play in. For the same reason, we can get spurious split-and-ties, but they vanish the next time the bar is edited. This situation could be kinda fixed by using performance times when notation times are too far off. I won't do that unless it annoys me a lot more than it does now. It doesn't do real harm and we've been living with it for a long time. Tom Breton (Tehom) |