John Chambers wrote: > > Phil Taylor writes: > | > | >2. L=1/4 and [FG/]G , what beat is the second 'G' on? 2 or and-of-one? > | > | Undefined, I'm afraid. > | > | MusicXML has an interesting construct to deal with this kind of situation. > | The <backup> and <forward> tags have the effect of moving the time point, > | so you can use <backup> to go back to the start of a measure in order to > | add an extra layer of notes. This means that you can deal with temporary > | voices which appear and disappear in the course of a piece. > | > | Maybe we need something similar in abc? > > I sorta recall reading about just such a feature in > abc2mtex, with a comment that it probably wouldn't work > with other abc programs. I've never read about anyone else > ever implementing it. > > Now what was that syntax? ... >
ABC2MTEX v. 1.6.1 User Guide section 4.2 allows multistave input with syntax similar to MusicTeX's as illustrated below (sample from the U.G.): DEFG & ABcd && A4 & e2 c2 | equivalent to a two-stave system as this: [V:1] ABcd e2 c2 | [V:2] DEFG A4 | (i.e., groups joined by '&' make a set where the groups are rendered in vertical sequence upwards in diferent staves aligned on the first note while '&&' signals the start of the next &-joined group set). Nevertheless, it says nothing about temporary voices (section 2.2.14, "Chords & unisons", on the regular abc construct for chords, also shows same-length voicings only). What, I think, comes closer to > [...] a comment that it probably wouldn't work > with other abc programs [...] is UG section 2.3.2 "New notation", on the use of letters H-Z to generate whatever TeX input the user ultimately assigns to the letter in the "header.tex" file or in the abc file, but still it doesn't address the problem of different-length temporary voices directly. Perhaps what Phil Taylors suggests might be achieved by combining the &-&& ABC2MTEX extension with the x-rest extension included in some packages for an invisible rest, with the additional rule that the group sets belong to the same staff, e.g. x4 & CDEF & x>c && x2D2 & GABc & BAGF && % and so on or maybe, allowing for longer segments of chords with temporary voices, MVSTART x4 x2D2 ... & CDEF GABc ... & x>c BAGF ... MVEND (where MVSTART and MVEND are placeholders for start and end delimiters of the multivoice staff segment, e.g. !chordphrasestart! ... !chordphraseend!, and the "..." indicate omission of irrelevant matter). That way the meaning would be clear even in case voices crossed. Paulo E. Tibúrcio To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
