Re: PATCH: Consolidate autobeaming to one property that controls it

[Carl Sorensen]

On 7/12/09 9:43 AM, "Neil Puttock" <address@hidden> wrote:

> 2009/7/12 Carl Sorensen <address@hidden>:
> 
>> I've created a second patch set with the whitespace removed (along with
>> automatic removal of whitespace from the lines of the files that weren't
>> part of the patch).
> 
> Brilliant work, Carl!

Thanks for the compliment.   It was considerably harder than I expected it
to be, and I ended up running down several dead ends, but I think I finally
got to something that works well and integrates nicely.
> 
> I'll comment on individual issues in Rietveld shortly.

Thanks for your comments.  I've read them and will respond to them (probably
not until tomorrow).

> 
> I notice you've changed the autobeaming for 4/8; is this deliberate?
>

No.  It was a mistake.  I did it because scm/auto-beam.scm said "Use
beatLength", and I missed the setting in scm/music-functions.scm that said
to group it (2 2).

Thanks for the catch.  I've fixed it.

> Is it possible to fix the output from \displayLilyMusic for \time and
> set-time-signature?
> 
> \displayLilyMusic { \time 3/4 }
> 
> { \context Score { \set Timing . timeSignatureFraction = #(cons 3 4)
>     \set Timing . beatLength = #(ly:make-moment 1 4 0 1)
>     \set Timing . measureLength = #(ly:make-moment 3 4 0 1)
>     } }
> 
> should be
> 
> { \time 3/4 }
> 

I don't know.  I didn't write displayLilyMusic; that was Nicolas IIRC.  I do
know that \time 3/4 executes those functions.  It's really easy to go
forward in the parse tree (i.e. from \time 3/4 to \set Timing....), but I
don't know any way to reliably go in reverse.

I guess I'm not *too* concerned that \displayLilyMusic return exactly what
what entered into it.  As long as it returns valid music expressions, then I
think it's probably OK.

>> 
>> I've also fixed up the mistakes I made in the headers of the lsr snippets,
>> so there won't be double translations.
> 
> Where are the input/new snippets these are generated from (there's
> only one in the patch set)?
> 

Oh -- I see.  I did things wrong.

I copied the snippets from the lsr into input/new, and edited them.  Then I
used makelsr.py to copy them back into input/lsr.  But I neglected to do git
add for the new snippets.  I'll rectify that soon.  Thanks for that catch,
as well.

Thanks,

Carl