Re: Reasons why a LilyPond-to-MEI conversion should be developed

[David Kastrup]

Paul Morris <address@hidden> writes:

> I think the same considerations that apply for MusicXML also apply for
> MEI.  See the discussions about the Google summer of code project from
> last spring.
>
> Namely, LilyPond’s internal scheme data structure is a good target for
> import and export (better than LilyPond’s plain text input syntax),
> and it is easier to start with export to MEI which will give you a
> good sense of the mapping between the two formats, and then it will be
> easier to work on the import path.
>
> Also you can build on whatever work is done for MusicXML, which I
> assume has already begun through the summer of code work.  Hmmm…
> what’s the latest on that anyway?

The latest is that the project is over and I have a dump of stuff I need
to integrate into LilyPond before it may be useful.  My motivation to do
that is so-so as I've not yet seen a cent of the tutor's money yet (and
no mention whatsoever that there is any intent of or procedure for a
payout) for the considerable work I put in so far, and there does not
seem much of a point to ask for it when the code is not yet in LilyPond.

So it will likely arrive sometime next year.  It's just a starting
point, anyway, but it uses decidedly different techniques than current
converters.

Anybody wanting to get that dump and a sketch of how to integrate it is
welcome.  It's Scheme-only work requiring digging through LilyPond's own
Scheme internals for existing hooks in order to integrate it into a
command line utility.

-- 
David Kastrup