Graham Percival <gra...@percival-music.ca> writes: > On Fri, Jan 14, 2011 at 09:58:15PM +0000, Neil Puttock wrote: >> On 14 January 2011 21:04, David Kastrup <d...@gnu.org> wrote: >> >> > Sometimes I am inclined to just commit and be done. After all, this >> > would just omit a warning in working cases, and would fix the totally >> > broken warning message when breves and longas get into play. Of course, >> > this latter part would require translations of the fixed message to be >> > placed in the message catalogs. >> >> It looks fine to me, apart from a whitespace nitpick. The only thing >> missing is a regtest. > > Why a regtest? I mean, would what it check...? would it just be > verifying that lilypond doesn't print a message in this particular > case? At the moment we don't have a (good) infrastructure for > such regtests, but I hope to get that into play after 2.15.
And that the message gets more useful once \breve and \longa come into play. Anyway, I am currently confused about what engraver/settings are responsible for what kid of operation. This particular case would seemingly just cover a) putting several noteheads on a common compatible stem which works for chording quarter notes with half notes. It would apparently not help with b) putting several noteheads on a common incompatible stem which would be required for chording eighth notes with half notes or any kind of c) merging incompatible noteheads as long as the stem is incompatible which is required for identical-sounding concurrent notes with different durations, indicated for example by having just a half-note head on an eighth-note stem (I think that the head-collision logic caters for that subcase already, with several options). I think I want at least b) in place before I call it a feature rather than something working by accident. So the calls for regtests at least had the advantage of me trying a few other things and getting surprised. -- David Kastrup _______________________________________________ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
