Monday, November 28, 2011

pdss

Copperbox revision 2661.

I've added some more Glue objects - select, route and pack. This has highlighted some things I've overlooked like types in Pd and "function" arities.

Sunday, November 27, 2011

pdss

Copperbox revision 2660.

More work typing ports, plus I've added bounding box calculation for float and symbol atoms.

With the changes to ports, I've also added a simpler constructor for objects so now all objects in libraries like Glue and Math should be two-liners - one line for the type signature and one of the function definition.

Saturday, November 26, 2011

pdss

Copperbox revision 2559.

I've changed the object representation to used sized types to count the number of ports - this will cut down quite a bit of boilerplate.

Wednesday, November 23, 2011

pdss

Copperbox revision 2658.

A bit more work replacing Bool flags with enumerated types.

Tuesday, November 22, 2011

pdss

Copperbox revision 2657.

I've replaced so uses of Bool in function argument lists with more descriptive enumerated types.

This will also make things clearer if I need to add them to the context so I can have default values.

Monday, November 21, 2011

pdss

Copperbox revision 2656.

I've added a canvas object (unfortunately Pd calls two different things canvas).

Most of the Pd objects are now tentatively implemented, I'll have to work out what to do with them...

Sunday, November 20, 2011

pdss

Copperbox revision 2655.

I've added vertical and horizontal slider objects, and implemented bounding box calculation for the messages.

pdss

Copperbox revision 2654.

I've improved some more objects by adding properly calculated bounding boxes.

Saturday, November 19, 2011

pdss

Copperbox revision 2653.

I've added a correct (contextual) bounding box to some of the simple objects (print and int) and started to put object definitions into libraries to match the Pd layout - only Glue so far and only a few members of it...

Embedding Pd will have a lot of name clashes with Haskell's Prelude that I will have to do something about.

Friday, November 18, 2011

pdss

Copperbox revision 2652.

Initial work on calculating bounding boxes. I've worked out the sizes of objects boxes - this required counting the pixels which was a bit laborious, but at least I've worked out an algorithm for it.

Wednesday, November 16, 2011

pdss

Copperbox revision 2651.

I've decided the generic object wasn't a good idea so I've removed it. Also added a few more objects.

Tuesday, November 15, 2011

pdss

Copperbox revision 2650.

More work on objects.

Many objects are generic and are just drawn with a box and its ports (print, integer literals, random etc.). I'm not sure if these are worth the boilerplate of a newtype for every object, so for the time being I've added a generic object so I can avoid the newtype wrapping.

Monday, November 14, 2011

pdss

Copperbox revision 2649.

I've added label offsets to the Context as a number of objects need them. Added a toggle object.

Sunday, November 13, 2011

pdss

Copperbox revision 2648.

I've added a lot of abstractions from Wumpus and now PDSS can generate working Pd patches:

Saturday, November 12, 2011

pdss

Copperbox revision 2647.

I've added PdContext to monad - the PdContext will work like the DrawingContext in Wumpus-Basic so objects like canvas can take graphic parameters like fill colour, font etc. from the implicit context rather than need them supplied as arguments to the constructor.

PDSS will probably get quite a bit of the functionality of Wumpus-Basic, so pdss-core won't be directly analogue to Wumpus-Core.

Friday, November 11, 2011

pdss

Copperbox revision 2646.

More work on the Doc builders.

This code is easy to do without thinking about it, hopefully I should be able to tackle something more ambitious tomorrow as it's the weekend.

Thursday, November 10, 2011

pdss

Copperbox revision 2645.

I've tidied up the Colour code and the Doc code so it should be more readable. Particularly, the code has been formatted so it is easier to count the arguments supplied to records - PD doesn't reject files with bad record arities, instead they lead to obscure bugs. 

Wednesday, November 9, 2011

pdss

Copperbox revision 2644.

Finally I've got working RGB colours. The code needs tidying up, but I wanted to commit it now while it works (I don't what to lose the working coding tidying it up...).

Tuesday, November 8, 2011

pdss

Copperbox revision 2643.

More work on colour handling - still buggy, but somewhat nearer. The PD file format reference on the web is wrong, colours are 18-bit in PD. There are a couple of descriptions of a formula to convert to 18-bit on the web but I haven't got one working yet. I actually had a bigger problem where I was generating the wrong argument list for cnv objects.

Monday, November 7, 2011

pdss

Copperbox revision 2642.

Work on RGB colour, which is currently buggy as I can't fathom the conversion formula given in the guide to the PD file format.

Sunday, November 6, 2011

pdss

Copperbox revision 2641.

I've expanded PDSS so the internals follow the example of Wumpus-Core. A number of abstractions in Wumpus are going to be pertinent to PDSS, so adopting the module layout of Wumpus-Core seems wise.

Saturday, November 5, 2011

pdss

Copperbox revision 2640.

I've added the Format Combinators from Wumpus rather than depend on PrettyPJ.

Wumpus-Basic exposes the Format Combinators which seems wrong now I think about it.

Friday, November 4, 2011

pdss

Copperbox revision 2639.

I've started a new project PDSS - this intends to be an analogue to FMSS but it generates PD code rather than Csound.

Thursday, November 3, 2011

new-neume

Copperbox revision 2638.

I've changed the Concat combinators to use re-defined versions the regular pretty print names. This makes the example look like they are just pretty printing even though they are implemented with a monad.

Wednesday, November 2, 2011

new-neume

Copperbox revision 2637.

Work on the Bulgarian 6 example - the first four bars now print.

Tuesday, November 1, 2011

new-neume

Copperbox revision 2636.

More work adding pretty printers for LilyPond syntax elements.

Blog Archive

About Me

My photo
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.