Re: midi2ly: Fraction Reduction + 2 minor issues

[Simon Albrecht]

Hi Chris,

we use to keep communication on-list apart from special cases where 
privacy is required.


Am 16.04.2017 um 10:45 schrieb Christopher Heckman:
> On Sat, Apr 15, 2017 at 3:10 AM, Simon Albrecht <address@hidden> wrote:
> > Hi Chris,
> >
> > thanks for your analyses and code.
> >
> >
> > Am 14.04.2017 um 02:18 schrieb Christopher Heckman:
> > > (1) When midi2ly is run, it will print fractions with large numerators
> > > and denominators
> > >
> > > I have written a patch to fix this.
> >
> > Could you provide this as a git-formatted patch, please?
> > (In case of doubt, see
> > <http://lilypond.org/doc/v2.19/Documentation/contributor/working-with-source-code>)
> I tried to do it. I'm not 100% sure it was submitted; I haven't used
> git or anything like it before.

Well, there’s basically two options for how to proceed. Both require 
that you clone the git repository, add and commit your changes. (All of 
that should be explained in the CG <Contributor’s Guide>, ‘Basic Git 
usage’ or so).
The first option is to then use
git format-patch HEAD
(or HEAD~1, HEAD~2… if you have made two or more commits)
and send the patch to the bug list, to be appended to a tracker issue.
(<http://lilypond.org/doc/v2.19/Documentation/contributor/patches>)

Somewhat preferable is uploading the patch for review on Rietveld, as is 
done for all LilyPond development work. (see 
<http://lilypond.org/doc/v2.19/Documentation/contributor/uploading-a-patch-for-review>)

If getting acquainted with Git is too much overhead for you for now, you 
could use the lily-git user interface intended exactly for that. (CG 2.2)

> > > (3) At line 677, there's a cryptic comment
> > >
> > > # urg, this will barf at meter changes
> > >
> > > Is there a brief explanation of the problem available?
> > Well, apparently this particular code isn’t prepared to deal with meter
> > changes in the input. What exactly is unclear about that?
> >
> > Best, Simon
> First question: Is the exact problem known, or did midi2ly crash (or
> whatever -- that's another question), and someone said, "Oops, better
> not do that"? Or ... did someone recognize that the bug would occur,
> and put this here instead of trying to fix it?

I haven’t been involved in coding this at all, but to me the comment 
sounds like the author saw this limitation but didn’t want to fix it 
right away.

> Are there meter changes where midi2ly does not crash?

That would be up to the person tackling the limitation to figure out.

> Has anyone tried to fix it?

Probably not. If there is not already a tracker issue at 
<https://sourceforge.net/p/testlilyissues/issues/>, definitely not.

We’d certainly welcome your contributions and fixes to these bugs!

Best, Simon