Re: [Lilypond-auto] Issue 2783 in lilypond: wrong placement of timesignature

[lilypond]

Comment #46 on issue 2783 by address@hidden: wrong placement of timesignature
http://code.google.com/p/lilypond/issues/detail?id=2783

I agree that the line_count change is correct in that returning an unused  
variable value makes no sense.  However, I have my doubts that any code  
relying on its value will actually do something useful with it in the case  
of irregular lines.  If this was Scheme code, it would likely make more  
sense to return #f in order to alert the calling code to the fact that any  
operation based on the number of lines rather than their actual offsets was  
not likely to lead to useful results.

So returning 0 in the C case would be defensible as well.  Except that the  
old code would have decided to shift when given 0.  Shrug.  At least this  
code no longer depends on any particular value of ...::line_count.  I doubt  
that much of the code that does will actually benefit from having the  
correct value returned.