Copperbox revision 1226.
I've added a typeclass ExtractBarImages to handle the difference between images of 'single' bars and images of bars with overlays.
Tuesday, April 6, 2010
Monday, April 5, 2010
Neume
Copperbox revision 1223.
I've corrected a long standing but previously unnoticed bug in the LilyPond duration to optional duration transformation.
I've also implemented some of the Show instances to print shorthand, rather than use the derived instances that print all the constructors.
I've corrected a long standing but previously unnoticed bug in the LilyPond duration to optional duration transformation.
I've also implemented some of the Show instances to print shorthand, rather than use the derived instances that print all the constructors.
Sunday, April 4, 2010
Neume
Copperbox revision 1222.
More work towards overlays in LilyPond - a prototype is in place that generates LilyPond code, however it generates bad code. The code has a trivial problem that it doesn't yet use the parallelMusic command which I discovered recently. It also has a serious problem that parallelMusic doesn't work well with repeat symbols - I didn't know this until I started hand editing the generated code to make it print.
More work towards overlays in LilyPond - a prototype is in place that generates LilyPond code, however it generates bad code. The code has a trivial problem that it doesn't yet use the parallelMusic command which I discovered recently. It also has a serious problem that parallelMusic doesn't work well with repeat symbols - I didn't know this until I started hand editing the generated code to make it print.
Neume
Copperbox revision 1221.
Work towards overlays (parallel music) for LilyPond.
LilyPond is significantly harder candidate for overlays than ABC. Rendering can be stateful - so I sometimes must use stmap rather than fmap. Also overlays can have different types i.e. one overlay line can be guitar chords and one normal notes. I'm committing the current work even though it is a way of working, as its in danger of getting complicated, ad hoc very quickly.
Work towards overlays (parallel music) for LilyPond.
LilyPond is significantly harder candidate for overlays than ABC. Rendering can be stateful - so I sometimes must use stmap rather than fmap. Also overlays can have different types i.e. one overlay line can be guitar chords and one normal notes. I'm committing the current work even though it is a way of working, as its in danger of getting complicated, ad hoc very quickly.
Saturday, April 3, 2010
Neume
Copperbox revision 1220.
Finally, a successful implementation that provides both overlays and score syntax (i.e straight sections, repeats and repeats with alternative endings). Currently ABC only and only overlays of size two.
I'm going to use an arity family of functions vis-a-vis zip2, zip3 etc. Trying something like a typed printf is too complicated given that I have to send quite a few parameters to the rendering function and the 'variable arity' represents the size of paired tuples within a functorial datatype.
Finally, a successful implementation that provides both overlays and score syntax (i.e straight sections, repeats and repeats with alternative endings). Currently ABC only and only overlays of size two.
I'm going to use an arity family of functions vis-a-vis zip2, zip3 etc. Trying something like a typed printf is too complicated given that I have to send quite a few parameters to the rendering function and the 'variable arity' represents the size of paired tuples within a functorial datatype.
Subscribe to:
Posts (Atom)
Blog Archive
About Me
- Stephen Tetley
- Disambiguating biog as there are a few Stephen Tetley's in the world. I'm neither a cage fighter or yachtsman. I studied Fine Art in the nineties (foundation Bradford 1992, degree Cheltenham 1992 - 95) then Computing part-time at Leeds Met graduating in 2003. I'm the Stephen Tetley on Haskell Cafe and Stackoverflow.