Thanks David for the explanation! JM
> Le 20 avr. 2017 à 08:51, David Kastrup <d...@gnu.org> a écrit : > > Jacques Menu Muzhic <imj-muz...@bluewin.ch> writes: > >> Hello David, >> >>> LilyPond 3 could be a rework of the optimization engine doing away with >>> the hard pure/unpure distinction. >> >> Not sure I understand this, can you tell us more? > > It's not a user-level distinction. LilyPond currently makes line breaks > first, page breaks afterwards. Page break decisions do not influence > the line break decisions. Grobs/callbacks basically have to declare > whether they are dependent on line break decisions or not and use > different mechanisms and callbacks if they are. That's awkward because > there often are very rare circumstances where this makes a difference in > results. > > It's the main source of unexpected interactions when refining > typesetting stuff and consequently a major roadblock for > extending/fixing LilyPond: quite often a change has consequences where > other stuff is falling apart unexpectedly. That makes a lot of the > typesetting a mine field mostly navigatable by "backend experts" since > they have experience patching up the things falling apart as they go. > > -- > David Kastrup _______________________________________________ lilypond-user mailing list lilypond-user@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-user
