Re: Autobeaming patch rev 2

[Carl D. Sorensen]

On 8/21/08 2:25 PM, "Trevor Daniels" <address@hidden> wrote:

> Carl
>
> I'm not competent to judge your patch but this sounds excellent!

Thanks.

>
> When this arrived I was part-way through replying to your
> previous email, which was to make the point that not all
> combinations of the several parameters which control beaming
> were necessarily compatible.  There are many other combinations
> of properties in LilyPond which do not make sense.  I too was
> coming to the conclusion that beatGrouping and beatLength
> were essentially incompatible, and that beatLength was more
> intended for controlling the subdivision of manual beams,
> as I suggested earlier.

I think with this current architecture, we can use beatLength to subdivide
beams and either beatLength or beatGrouping to automatically divide beams.

>
> Do all the existing beam-related regression tests run
> correctly through your code?

They didn't as I had missed an improper setting of the default.

They do now -- there are no changed files in all of the regression tests.

See revised patch.

Carl

auto-beam-3.patch
Description: auto-beam-3.patch