Re: Autobeaming and cadenzas (was: bug-lilypond Digest, Vol 95, Issue 11)

[Patrick McCarty]

On Wed, Oct 6, 2010 at 10:14 AM, Carl Sorensen <address@hidden> wrote:
>
> On 10/5/10 11:44 PM, "David Kastrup" <address@hidden> wrote:
>
>> Carl Sorensen <address@hidden> writes:
>>
>>> On 10/5/10 5:50 PM, "Keith E OHara" <address@hidden> wrote:
>>>>
>>>> If \cadenzaOff turns autobeaming on, then what about scores that turn off
>>>> autobeaming for the whole piece, but then (mis-)use short cadenzas?
>>>
>>> The CHANGES file informs the user that they will need to turn off
>>> autobeaming manually after the cadenza, as does the notation reference.
>>>
>>> I think this is the right thing to do.
>>
>> I think a revert would be more appropriate.
>
> I agree that a revert would be more appropriate.  However, a revert would
> be very difficult to accomplish, because there is no easy way to save the
> "original" value.  The value is only known in the translation stage as a
> context property, so we can't really save the value for use in the parsing
> stage.
>
> Right now, the benefit/time ration isn't high enough for me to code the
> revert.
>
> I'm fine to have an enhancement request to implement it as a revert,
> however.

Can't we just use \unset instead?  Like

diff --git a/ly/property-init.ly b/ly/property-init.ly
index 5724004..9f7876d 100644
--- a/ly/property-init.ly
+++ b/ly/property-init.ly
@@ -101,7 +101,7 @@ cadenzaOn  = {
 cadenzaOff = {
   \set Timing.timing = ##t
   \set Timing.measurePosition = #ZERO-MOMENT
-  \set Timing.autoBeaming = ##t
+  \unset Timing.autoBeaming
 }


Thanks,
Patrick