Re: Circular dependencies for beam-skip.ly

[address@hidden]

On Jul 13, 2011, at 12:19 PM, James Lowe wrote:

> Mike,
> 
> 
> )-----Original Message-----
> )From: address@hidden [mailto:address@hidden
> )Sent: 13 July 2011 11:14
> )To: James Lowe
> )Cc: address@hidden List
> )Subject: Re: Circular dependencies for beam-skip.ly
> )Importance: Low
> )
> )On Jul 13, 2011, at 12:12 PM, James Lowe wrote:
> )
> )> Mike,
> )>
> )> )-----Original Message-----
> )> )From: address@hidden
> )> )[mailto:address@hidden
> )On
> )> )Behalf Of address@hidden
> )> )Sent: 13 July 2011 07:58
> )> )To: address@hidden List
> )> )Subject: Circular dependencies for beam-skip.ly
> )> )
> )> )Hey all,
> )> )
> )> )I'm getting a lot of circular dependencies for input/regression/beam-
> )> )skip.ly from the current master.  I don't have time to do a full git
> )> bisect, )but is this standard behavior?
> )> )
> )>
> )> See
> )>
> )> http://code.google.com/p/lilypond/issues/detail?id=1563#c3
> )>
> )> I suddenly got a reg test difference with this patch and that .ly
> )>
> )> James
> )>
> )
> )I think the same problem exists on the current master - could you confirm
> )when you get a chance?#
> 
> I don't know what that means, but if you mean that the current git tree where 
> Jan N's checkin is the latest, well I used that last night.
> 
> Now if it had been in current master then the make test-baseline and make 
> check would not have shown up this reg test (they would have been the same) 
> right?
> 
> james

You're right - they would have been the same.  I think this error may have been 
introduced in a recent patch, in which case it is a regression.  I'd have to 
run git-bisect to figure it out, though, but all of my CPU is being eaten up by 
audio rendering at the moment (and just months ago I was bragging about my fast 
computer...serves me right...).

Does anyone have any intuition about what the culprit may be and/or if these 
circular dependencies are intended behavior?

Cheers,
MS