Re: Ledger-line-spanner: symmetric extents; issue 2493 (issue 6490043)

[Keith OHara]

On Wed, 29 Aug 2012 11:23:42 -0700, <address@hidden> wrote:

> http://codereview.appspot.com/6490043/diff/7001/lily/ledger-line-spanner.cc#newcode50
> lily/ledger-line-spanner.cc:50: < Paper_column::get_rank
> (previous_column)))
> It seems that previous_column should have a rank falling before current column,
> not after.
This is a helper function for the loop down on line 100, which runs backwards.
There is already a comment just below "we go from right to left."
The loop can now run forwards; maybe I should un-reverse it.

> Also, it seems like if the variables are called this, then the order
> should be correct when they are passed into the function (meaning the
> one supposed to come before the other comes before the other).
> Otherwise a programming error should be raised.  What would be the case
> where the ranks of the columns would be reversed and what would justify
> that happening?

Master does raise a programming for ambitus note heads
  \new Staff \with { \consists "Ambitus_engraver" } { c''' d'' e' f }
The ambitus engraver creates the heads when it first starts, and adds them 
using add_grob(), but somehow that puts them between the c''' and d'' in the 
array.

The array contains every note-head in the Staff (that is, thousands) so I would 
rather not take the time to sort it.  Creating an add_front() through all the 
layers of classes, for use in ambitus_engraver.cc, did not seem wise.

Given at least one known exception to ordering, I added a test for ordering.  
Now I think it better to test for the known exception: skip over any 
ambitus-note-heads.  Or, we could continue to issue a programming-error for 
this case.

> http://codereview.appspot.com/6490043/

My motivation is to replace the non-looping loop
  Direction d=DOWN; do {...} while (flip (&d) != DOWN);