On 06.07.2013, at 23:18, Pavel Roskin <pro...@gnu.org> wrote:

> Quoting pls <p.l.schm...@gmx.de>:
> 
>>> I don't care how the faulty markup looks.  But MuseScore got the  second 
>>> measure right.  It could recover and musicxml2ly could not.
> 
>> that's because of the faulty MusicXML markup you provided. the first  
>> measure contained the duration of five quarter notes instead of  four. 
>> that's why all the following measures were affected.
> 
> Oh well, this puts me between the rock and the hard place.  musicxml2ly 
> cannot deal with MusicXML errors reliably.  On the other hand, MuseScore can 
> read broken MusicXML but makes broken Lilypond output (missing braces).
> 
> One possible approach is to open and save MusicXML in MuseScore and then use 
> musicxml2ly on the saved file.  I checked it quickly, the chords are not 
> lost, but I need to see if anything else is preserved.
> 
> I don't know how to persuade you that musicxml2ly could and should do better.
You don't need to persuade me that musicxml2ly is far from perfect.  (I know 
the script very well.)   The problem you describe is not really a bug in 
musixml2ly but an ill-defined feature request.  You are basically asking for an 
automated correction of false values in <duration> elements in combination with 
an automated corrective assignment of notes to voices.  This is non-trivial – 
especially in polyphonic parts with more than two voices.
> 
> -- 
> Regards,
> Pavel Roskin


_______________________________________________
bug-lilypond mailing list
bug-lilypond@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-lilypond

Reply via email to