Add modal transformations provided by Mike Ellis

Add modal transformations provided by Mike Ellis

- this is a continuation of 4126042 after rebasing

- and after responding to Neil's comments

[Trevor Daniels, 2011-02-05 10:17:48]

This is a continuation of Issue 4126042.  I had to start a new Rietfeld number
as I'd moved to a new branch after rebasing and resolving merge conflicts in
preparation for pushing ... then Neil's comments arrived.  Sorry Neil - I think
I've responded to all your comments correctly, but the direct comparison with
the previous upload is unavailable, so it won't be so easy for you to check.

Trevor

[Graham Percival (old account), 2011-02-05 18:20:48]

LGTM.

http://codereview.appspot.com/4079064/diff/1/Documentation/notation/pitches.i...
File Documentation/notation/pitches.itely (right):

http://codereview.appspot.com/4079064/diff/1/Documentation/notation/pitches.i...
Documentation/notation/pitches.itely:827: it may be converted to its
@notation{retrograde} (written backwards).
I'm not certain you need to spell out "written backwards".  You haven't said
what "inverted" or "transposed" means, after all!

If you're paranoid, then how about adding a stub to the music glossary, and a
@rgloss{} to the @seealso ?  (or actually add a full entry to the glossary if
you want)

http://codereview.appspot.com/4079064/diff/1/Documentation/notation/pitches.i...
Documentation/notation/pitches.itely:855: \new Staff {
I know that you don't want to hear this at this point, but...

You know, all these examples are beautifully written and are very clear.  I
think we could have capitalized on this by doing everything at once.

@lilypond
motif = ...
diatonicScale = ...
pentatonicscale = ...
chromaticscale = ...

\relative c' {
  \motif
  \modalTranspose \motif
  \modalInvert \motif
  \modalRetrograde \motif
@end lilypond

this isn't really a serious suggestion, and in any case, I'd need to look at the
output of a .ly example before being able to have a serious opinion on whether
this would actually help or not.  (the all-in-one example might be too long or
confusing or something)

I'm mainly just mentioning this so that people have it in the backs of their
minds when writing future doc stuff.


(BTW, do you actually need the \score and \new Staff in these?  My first guess
would be that a simple \relative c' would suffice, but maybe I'm wrong or maybe
the transposing complicates the \relative)

[t.daniels_treda.co.uk, 2011-02-15 21:04:52]

percival.music.ca@gmail.com> wrote Saturday, February 05, 2011 6:20 
PM

>
http://codereview.appspot.com/4079064/diff/1/Documentation/notation/pitches.i...
> Documentation/notation/pitches.itely:827: it may be converted to 
> its
> @notation{retrograde} (written backwards).
> I'm not certain you need to spell out "written backwards".  You 
> haven't
> said what "inverted" or "transposed" means, after all!
>
> If you're paranoid, then how about adding a stub to the music 
> glossary,
> and a @rgloss{} to the @seealso ?  (or actually add a full entry 
> to the
> glossary if you want)

I'll do this as a separate patch after this has been pushed.

>
http://codereview.appspot.com/4079064/diff/1/Documentation/notation/pitches.i...
> Documentation/notation/pitches.itely:855: \new Staff {
> I know that you don't want to hear this at this point, but...
>
> You know, all these examples are beautifully written and are very 
> clear.
>  I think we could have capitalized on this by doing everything at 
> once.

I think it would have been a little too long.

> (BTW, do you actually need the \score and \new Staff in these?  My 
> first
> guess would be that a simple \relative c' would suffice, but maybe 
> I'm
> wrong or maybe the transposing complicates the \relative)

\modalTranspose and the other functions have problems
inside \relative blocks.  Ordinary \transpose is similar.
I can get away with just a \new Staff, so I'll change to
this.

Trevor

[t.daniels_treda.co.uk, 2011-02-15 21:04:52]

percival.music.ca@gmail.com> wrote Saturday, February 05, 2011 6:20 
PM

>
http://codereview.appspot.com/4079064/diff/1/Documentation/notation/pitches.i...
> Documentation/notation/pitches.itely:827: it may be converted to 
> its
> @notation{retrograde} (written backwards).
> I'm not certain you need to spell out "written backwards".  You 
> haven't
> said what "inverted" or "transposed" means, after all!
>
> If you're paranoid, then how about adding a stub to the music 
> glossary,
> and a @rgloss{} to the @seealso ?  (or actually add a full entry 
> to the
> glossary if you want)

I'll do this as a separate patch after this has been pushed.

>
http://codereview.appspot.com/4079064/diff/1/Documentation/notation/pitches.i...
> Documentation/notation/pitches.itely:855: \new Staff {
> I know that you don't want to hear this at this point, but...
>
> You know, all these examples are beautifully written and are very 
> clear.
>  I think we could have capitalized on this by doing everything at 
> once.

I think it would have been a little too long.

> (BTW, do you actually need the \score and \new Staff in these?  My 
> first
> guess would be that a simple \relative c' would suffice, but maybe 
> I'm
> wrong or maybe the transposing complicates the \relative)

\modalTranspose and the other functions have problems
inside \relative blocks.  Ordinary \transpose is similar.
I can get away with just a \new Staff, so I'll change to
this.

Trevor

[Trevor Daniels, 2011-02-16 12:04:33]

I've just uploaded the latest changes from Mike.  These extend modalInversion to
permit inversions around a pivot between two notes in the scale.

There are also some minor changes suggested by Graham.

Please review.

Trevor

[benko.pal, 2011-02-16 12:18:26]

> I've just uploaded the latest changes from Mike.  These extend
> modalInversion to permit inversions around a pivot between two notes in
> the scale.
>
> There are also some minor changes suggested by Graham.
>
> Please review.

LGTM

> http://codereview.appspot.com/4079064/

[Graham Percival (old account), 2011-02-16 18:11:05]

LGTM

http://codereview.appspot.com/4079064/diff/7001/Documentation/notation/pitche...
File Documentation/notation/pitches.itely (right):

http://codereview.appspot.com/4079064/diff/7001/Documentation/notation/pitche...
Documentation/notation/pitches.itely:830: left untransformed and a warning
given.}
I'd omit the "and a warning given".  I mean, the warning will be obvious, right?
 Just end with "... left untransformed."

http://codereview.appspot.com/4079064/diff/7001/input/regression/modal-transf...
File input/regression/modal-transforms.ly (right):

http://codereview.appspot.com/4079064/diff/7001/input/regression/modal-transf...
input/regression/modal-transforms.ly:11: motive = {
I'm amused by the discrepancy between "motif" (in the docs) and "motive" (in
this regtest).

[Keith, 2011-02-17 03:46:51]

Still works well for me, including the new bits.

Some late suggestions on the docs that you can take or leave.

http://codereview.appspot.com/4079064/diff/7001/Documentation/notation/pitche...
File Documentation/notation/pitches.itely (right):

http://codereview.appspot.com/4079064/diff/7001/Documentation/notation/pitche...
Documentation/notation/pitches.itely:861: A scale of any length and with any
intervals may be specified:
"An ascending scale of any length ..."
The octaves are linked smoothly if the scale is ascending.

Jumps of an octave in the input scale will be a common first-time mistake, so
the \relative command in the example is very good idea. It might be worth ten
more words.
"The @code{relative} command is useful for creating a continuous scale:"

[t.daniels_treda.co.uk, 2011-02-17 11:22:32]

>
http://codereview.appspot.com/4079064/diff/7001/Documentation/notation/pitche...
> Documentation/notation/pitches.itely:830: left untransformed and a
> warning given.}
> I'd omit the "and a warning given".  I mean, the warning will be
> obvious, right?  Just end with "... left untransformed."

Done

>
http://codereview.appspot.com/4079064/diff/7001/input/regression/modal-transf...
> input/regression/modal-transforms.ly:11: motive = {
> I'm amused by the discrepancy between "motif" (in the docs) and 
> "motive"
> (in this regtest).

Missed this one!  Fixed. 

[t.daniels_treda.co.uk, 2011-02-17 11:27:36]

>
http://codereview.appspot.com/4079064/diff/7001/Documentation/notation/pitche...
> Documentation/notation/pitches.itely:861: A scale of any length 
> and with
> any intervals may be specified:
> "An ascending scale of any length ..."
> The octaves are linked smoothly if the scale is ascending.

Done

> Jumps of an octave in the input scale will be a common first-time
> mistake, so the \relative command in the example is very good 
> idea. It
> might be worth ten more words.
> "The @code{relative} command is useful for creating a continuous 
> scale:"

I didn't take this.  I don't fully understand the
interaction between \relative, \transpose and the
new modal transforms, so just adding a simple
recommendation might be misleading.  In any case
the examples show this clearly enough.

> http://codereview.appspot.com/4079064/

[Trevor Daniels, 2011-02-17 11:30:05]

Now pushed.
fd7ad42f98d7e6117958a41e74cd736e396fb933
Will appear in 2.13.51.

Many thanks, Mike.