Hi Vaughan,

thank you for trying!

> All the notes and lyrics were in the right places. Recently I got excited
> by Lilypond’s automatic placement of lyric extenders, and deleted all the
> "__" from my score, so the XML version has no lyric extenders. A
> previous version of the file has the extenders so I could use that, but
> future scores would be less likely to include explicit extenders.

Actually that is not an issue of you removing __. I designed the system the way 
I did so that it is 
actually compatible with what engravers are doing, such as 
Completion_heads_engraver and 
stuff (which is the reason why instead of interpreting the music I react to the 
grobs created by 
the engravers). This means this could work with automatic syllabic lines, just 
that these I have 
not implemented yet (because I consider them somewhat of a lower priority. They 
are very easy 
to implement, and you can do so yourself. Essentially:

LyricExtender stores spanned heads in (ly:grob-object grob) heads. Via these we 
get access to the 
note-column xml-reference, and thus access to the base note. We can then add to 
each such 
note a <lyric><extend/></lyric> node.

So it is ~ 15 lines of code or something. But as I said, I did not consider it 
that important and 
rather implemented some other things. If it is important to you I can quickly 
add it.

> Text scripts were great, but one came out as "mezza voce &dagger;", so the
> ASCII alias wasn’t replaced.

Currently the whole thing is yet missing proper xml-escaping. The only thing 
currently 
implemented it that " in attributes is replaced by \". This is something on my 
list, but I wanted to 
check if there is already a scheme module for this that can be used. If not I 
will add a general 
replacement for <, >, &, " and ' by the correct xml-replaces. That will also be 
quite fast to 
implement, in any case.

> An anacrusis bar was achieved with hidden rests after the notes, which
> didn’t look great.

This is probably caused by the importing software. I’m missing still some small 
logic to (in case of 
anacruses) set the `implicit="yes"` attribute, which probably causes your 
software to interpret 
this as underful measure and fills it with rests. Will also be fixed.

> So overall it exported this piece excellently. It would require very little
> cleaning up to be publishable. Well done!

Thank you very much, and thank you for trying and all of the input as well!

Before adding new features though I want to make sure the existing ones are 
working fully 
correctly and well, which means I’ll need to fix things like

https://codeberg.org/tinapetzel/ly-mxml/issues/3[1]

first.

Here it is really important for me to get your support, since I do not have 
access to every 
notation software and musicxml workflow out there, so I cannot test everything 
myself. So again, 
thank you very much.

Cheers,
Tina



--------
[1] https://codeberg.org/tinapetzel/ly-mxml/issues/3

Attachment: signature.asc
Description: This is a digitally signed message part.

Reply via email to