David Kastrup <d...@gnu.org> writes: > Urs Liska <u...@openlilylib.org> writes: > >> I've only read the documentation, and I agree with Trevor that this is a >> real improvement. >> >> One question/suggestion: Would it be possible to have \voicify accept a >> single symbol with a "pattern", which should allow two entries like >> #'top-bottom and #'bottom-top? This could count the number of following >> expressions and calculate the list (like 1,3,5,6,4,2 in the docs example). > > No. A symbol already has meaning. > > You could try separate commands \voicifyUp and \voicifyDown . I am not > sure whether or not \voicify should not be \voices or \voicing or > \voicings instead, possibly making for nicer compounds like that. > > I mean, something like > > \voices 1,3,4 ... > > possibly stresses the connection with \voiceOne, \voiceThree, \voiceFour > better. And then \voicesUp and \voicesDown ? Somehow a bit too close > to \stemUp / \stemDown meaning something entirely different. > Upwards/Downwards ? Rising/Falling ?
And frankly, I don't think it a good idea. What should \voicesUp << a \\ b \\ c >> be equivalent to? \voicify 2,3,1 or \voicify 2,4,1 ? And \voicesDown ? \voicify 1,3,2 or \voicify 1,4,2 ? There are obvious use cases for either and people will be confused when the pattern does not fit their use after all, and people will always forget which is which. Those numbered lists are brief and painless to enter, clear in the semantics, and they never force you to warp your input to match the scheme when it would be nicer in another manner. That's a strong point of the command, and a \voicesUp/\voicesDown does not save enough typing to make up for this. -- David Kastrup _______________________________________________ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
