On Aug 18, 2011, at 2:51 PM, Han-Wen Nienhuys wrote:
> On Thu, Aug 18, 2011 at 9:27 AM, <mts...@gmail.com> wrote:
>> I figured not making a regtest just cuz there will already be little
>> changes to several regtests. But, I can certainly add one.
>>
>> Cheers,
>> MS
>>
>>
>> http://codereview.appspot.com/4898060/diff/1/lily/ledger-line-spanner.cc
>> File lily/ledger-line-spanner.cc (right):
>>
>> http://codereview.appspot.com/4898060/diff/1/lily/ledger-line-spanner.cc#newcode89
>> lily/ledger-line-spanner.cc:89: continue;
>> On 2011/08/18 12:16:50, hanwenn wrote:
>>>
>>> can't you check poss.find(max(0, lincount - *it)) ?
>>
>>> I don't understand what this is supposed to do ; in which cases do you
>>
>> want to
>>>
>>> skip ledger creation?
>>
>> Good call - I can do this.
>>
>> It skips over all ledger lines that already have a note on them, thus
>> avoiding duplicates.
>
> are you sure this works with chords that have seconds in them? It
> sounds like an optimization that doesn't really improve output at all,
I agree that the improvement is kinda meh - I was more optimistic about it when
writing it then when seeing the actual results, as there are collisions with
lines above and below depending on the accidental.
> and performance not by much. Why not skip it altogether and avoid the
> complexity?
>
I am more or less indifferent to how it gets done, but I do think these
collisions should be avoided.
One easy fix is simply making the default whiteout property of accidentals to
true somewhere in ly/ and then pushing their layer way down low and push ledger
lines one notch lower.
{
\override Staff.Accidental #'layer = #-100
\override Staff.LedgerLineSpanner #'layer = #-101
\override Staff.Accidental #'whiteout = ##t
<des ces'>
}
Thoughts?
Cheers,
MS
_______________________________________________
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel
