On Dec 28, 2011, at 6:30 PM, n.putt...@gmail.com wrote: > > http://codereview.appspot.com/5504055/diff/3002/scm/define-grob-properties.scm > File scm/define-grob-properties.scm (right): > > http://codereview.appspot.com/5504055/diff/3002/scm/define-grob-properties.scm#newcode1101 > scm/define-grob-properties.scm:1101: (skyline-quantizing ,ly:vsize? "The > number of boxes to break a > Mike, why have you added this predicate? It has no analogue in Scheme. > I think integer? would be fine for this. You can silently prevent > negative numbers. >
I added it precisely to un-silently prevent negative numbers. It's true that it has no analogue in Scheme, but I didn't think this was a prerequisite for adding a predicate. In lily.scm, there are several predicates that don't have Scheme analogues (cheap-list, I believe, has no Scheme counterpart). > http://codereview.appspot.com/5504055/diff/3002/scm/define-grob-properties.scm#newcode1102 > scm/define-grob-properties.scm:1102: slur into when calculating its > skyline.") > I think this should be more generic. There are other objects which > would benefit from this such as hairpins and text spanners. Will do. If anyone feels like taking this on, please do - I have a few hard deadlines coming up and my lily development will be minimal for the next month-ish. Writing vertical-skyline approximations is not that bad - just use the slur model as a template and move code around to avoid duplication (for example, in the tuplet case, you'd want to change Tuplet_bracket::make_bracket to spit out a list of stencils and then harvest the boxes from the bounding boxes of the stencils in this list - then, move whatever logic is needed before calling Tuplet_bracket::make_bracket to a different callback that can be used by both the print callback and the vertical skylines callback). Cheers, MS _______________________________________________ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
