Re: FretDiagrams Architecture

[Carl Sorensen]

On Sun, 2004-11-21 at 05:25, Han-Wen Nienhuys wrote:

> I think that
> 
> a. we are storing chords in the wrong way. In retrospect, I find
> 
>    root + list of (step, alteration) tuples
> 
> more logical

So rather than using the pitches as the key for the hash table, it ought
to be the chord structure, as is inferred in ChordNames, but not the
chord name itself.  Is that what you have in mind?

> 
> b. using the octave to store different variants is misuse of that
> less-than-perfect storage model.  I think that we should have some
> kind of ChordAlternativeEvent to store the extra information.

Something along the lines of 

  \chordmode {
       d \alternative 1
      }

?

Carl