Re: Grand PartCombine Rewrite Project

[Anthony W. Youngman]

In message <address@hidden>, Kieren 
MacMillan <address@hidden> writes
> Hello all,
>
> I've decided this is going to be my pet project during this, my 
> "Autumn of Lilypond".
> I've started a separate thread here, and would like to get moving on 
> the project as soon as possible. [n.b. I am not intending to cross- 
> post to -user after this "announcement" email.]
>
> 1. Team members. Several people (Wol, etc.) expressed an interest in 
> helping; if you are one of these, please re-confirm your interest 
> here, for thread completeness. If anyone else wants to help, just 
> shout out.

Yes, I'm still interested.
> 2. Lead(s). I don't mind taking the lead on this if nobody else wants 
> that job; however, if anyone *does* want it, I can guarantee you'll  do 
> better than I, and am happy to defer.
>
> 3. Programming. In order to get this done before, say, 2020, we'll 
> need at least one lead developer with good Scheme-fu to (1) write  most 
> of the heavy-lifting code, and/or (2) vet my/our Froggie work.  In 
> either case, my goal is to have "clean" patches to give to Carl/ 
> whomever for final merging, as opposed to banging on his/their door 
> with sketchy code.

Well, I want to learn good Scheme-fu, so I'm quite happy to tackle the 
heavy lifting, but at the moment my tools are made of balsa wood ... :-)
> 4. Here's my first, naive attempt at listing the areas where a good 
> partcombine must Do The Right Thing:
>  — \relative
>  — \transpose
Hopefully, partcombine comes after these two have done their thing (that 
makes logical sense to me) so if I'm right we can ignore these two.

>  — \times
>  — lyrics
>  — merging/discarding notes & rests
>  — merging/discarding text (e.g., markup)
>  — merging/discarding articulations, etc.
>  — adding "editorial" texts (e.g., "a2")
>  — unlimited number of parts

Please note, I think it would be a good idea to control a lot of this 
with switches. Again, I don't know how it's currently done, and this MAY 
be a stupid idea, but it seems to me that the basic lifting engine 
should be capable of combining parts, voices, etc etc.

It feels to me that combining two parts (each a music expression) into 
one music expression is logically exactly the same thing as converting 
two voices (each a music expression) into one "music on a staff" which 
feels like it, too, should be a music expression. Quite how practical 
this is is entirely another matter ...
> 5. My 30000-foot plan on how to proceed is to:
>   — dissect the current implementation to see if anything can be 
> saved/rewritten;
>   — plan the implementation revision/rewrite; and;
>   — do it.
Count me in. I want to learn. Just be warned, I do tend to get a grand 
vision, get a good grip of the detail of what needs to be done, then 
need a shove to actually get it done. I'm very good for bouncing ideas 
off, just not so good at actually getting them implemented.

Cheers,
Wol
--
Anthony W. Youngman - address@hidden