Re: New feature: automatically invert chords or drop/rise chord notes (issue 365840043 by address@hidden)

[v . villenave]

On 2019/01/28 21:53:04, dak wrote:
> Without any pointer to what you are having problems with, this is
essentially
> "do it yourself".

David, there seems to be some misunderstanding here (aside from your
usual snark :-)
My question was only addressing your comment with regard to (list-set! )
being unnecessary; prior to that, I had addressed all of your remarks
and uploaded an updated function that was entirely working, without any
need for additional work on your part.  So, you could hardly blame me
for letting you do all the work where I did, precisely, produce working
(if inelegant) code.

> Sigh.  I don't even understand what those interfaces are
> supposed to be good for

I’m not exactly sure myself (I wasn’t the OP). But I do know that I
sometimes use chordmode to produce basic accompaniment written on
staves, and there’s little doubt that enabling additional voicings will
be occasionally useful to me. Since we’ve always supported chordmode as
a shorthand for writing chords on staves (and not only chord names),
it’s a rather nice additional feature to offer.

> so I at least changed the internals to stop chaotically
> mixing directions and counts.  The user-level commands remain as
chaotic as I
> find them to be.

Well, now I have to use your code (how could I not, since it has the
same feature set than what I laboriously came up with, but with only
half the LOC count).  However, again, please note that from the
user-exposed side, you didn’t have to do that (even though I am thankful
for it).

> Well, considering the way invertChords is written, this should likely
just be
> (define (proper-pitch note) (ly:music-property note 'pitch))
> since "octavation" is only informational without affecting the pitch.
Thinko,
> sorry.

Oh, indeed. (I shall refrain myself from noticing out loud that _my_
last version didn’t have that bug -- of course then again, it probably
took me more time to write it than what you’d need to rewrite half our
codebase :-)

> By the way, I find it somewhat strange that the code discussion on the
user list
> as well as the patch progress here completely ignore this
contribution.  After
> all, it was solicited.

It is strange, through no fault of my own.  I did post a link to both
Allura and Rietveld:
https://lists.gnu.org/archive/html/lilypond-user/2019-01/msg00407.html
and I even asked on the list when my first draft produced unwanted
behavior.

V.

https://codereview.appspot.com/365840043/