On 2024-11-07 12:06, Richard Gress - rgress(a)nd.edu wrote:
Jason, the new commands as presented in the 2.25 docs makes more sense than in 2.24, and doesn't require timing commands - bravo to you and whoever else worked on this.
Thank you!
The English instructions have a few grammatical errors that will need to be fixed for English readers. The example could perhaps be expanded, too. For example, users may wish to know about exactly what "respect" incomplete beams means. But I'm not familiar with software/documentation development processes.
The old algorithm in 2.24 operated as if respectIncompleteBeams is always true, which should be false by default (but can be configured to be true to preserve compatibility). I can clarify this in the documentation.
In 2.25, is it [yet] possible to set minimumBeamSubdivisionInterval to anything other than a \musiclength = X?
Generally, minimumBeamSubdivisionInterval is expected to be a power of 2 (e.g. 1/32, 1/8, 1/2, or even 2). Setting it otherwise will not produce special behavior as it is literal to its name. However, if maximumBeamSubdivisionInterval isn't (like the 2nd measure of the example in the documentation, w/ value of 3/8), then the beams are subdivided at every multiple of 3. This works best in simple meter; currently it may have some bugs if in compound meter. Supporting non-powers of 2 for both properties at the same time does not seem to make sense in simple meter, but may be expected when working with compound meter (as the numerators can reasonably be 3 in your case, but I believe that minimumBeamSubdivisionInterval should be then interpreted to the nearest power of 2 and not be special in its own way).
