Re: Charles Winston's chord-semantics GSOC work (issue 568650043 by address@hidden)

[Carl Sorensen]

On 4/8/19, 1:59 PM, "address@hidden" <address@hidden> wrote:

    On 2019/04/08 18:47:13, lilypond-pkx wrote:
    > Are we sure all the reg tests are OK (see tracker for download link)?
    
    > For example
    
    > regression/chord-name-major7.ly
    
    > This looks completely broken with this patch

Interesting question.  The semantics that Charles coded says that c:maj7 is the 
way to indicate the C major 7 semantics.  This regtest had c:7+, which gives 
the same pitches as c:maj7, but has a different semantic spelling.  In my 
opinion, c:7+ should probably be named C#7, not C triangle.  Of course, the 
code is broken because it loses the alteration on the extension (but not the 
addition).
    
    
    input/regression/chord-voicings.ly
    #11 always becomes 11
    On purpose? Not sure.

Same problem as before: the alteration on the extension is not typeset.  Bug 
that needs fixing.

    
    input/regression/spacing-non-adjacent-columns3.ly
    F9add13add15 becomes F13add15!?
    On purpose? Not sure.

Definitely not on purpose.  Need to investigate.
    
    input/regression/one-staff.ly
    B7b5 should be printed as before, imho.
    Thus, broken.

The current semantics is entered as b:m7.5-, which is Bm7b5.  The half-dim 
symbol was part of the ignatzek exceptions.  Probably need to add it to 
semantic-exceptions.
    
    input/regression/chord-name-override-text.ly
    Shouldn't there be a "foo"?
    Likely broken

Yes -- this needs investigation.

    
    input/regression/chord-names-languages.ly
    Broken, but germanChords are often insufficient anyway...
    
    input/regression/predefined-fretboards.ly
    Absolutely broken.

I agree.  This is probably due to the use of chord shapes --- need to figure 
out how best to handle it (probably drop semantics from chord shapes).
    
    
    Carl, many thanks working on it, but I'm sorry, I think there are some
    show-stopper for now.
    
You're absolutely right.  I've made some progress, but it will probably be next 
weekend before I can finish it.

Thanks,

Carl

    
    
    https://codereview.appspot.com/568650043/