These files contain Lisp code that works with the pitch-class sets of post-tonal music theory. Specifically, imset-code-only-functions provides an "imset-decision-maker," which is roughly described in various pieces of documentation throughout the files.
The overall project is in rough form, but is in the process of being refactored into more logical pieces and documented more conventionally.
Ultimately, an interactive front-end, probably to be used via PWGL or a similar graphical environment, allowing the user to hear and explore the relationships would be ideal.
There is now (<2015-04-28 Tue>) a new direction to this project, which was originally written entirely in Common Lisp. Take a look at the .org files for the most recent modifications and development.
A suite of functions for dealing with a new, relatively-esoteric concept in the relatively esoteric world of post-tonal music theory: the interval multiset (or imset), which is derived from the concept of an 'interval function' (or IFUNC) as described by influential music theorist David Lewin in his classic Generalized Musical Intervals and Transformations.
The following code is being used to provide the computational underpinning for research to be published soon in a scholarly music theory journal and, as such, is not quite ready for public consumption. There is a considerable amount of commenting that exists mostly to keep a kind of running note-taking for me as I consider the issues of design and how to adjust to the changing needs of the data which is being produced for the different structures being inputted. the commenting style needs to be adjusted so more of the useful information there appears in a documentation string (most functions lack such documentation). Consider also providing consistent appearances of examples, cf. the highly-structured model used in How To Design Programs
While there is a very rough attempt to group functions that belong together in the same sections, consider a more standardized approach to semi-colon commenting. for some reason, hash-pipe commenting blocks seems to be glitchy in Aquamacs, but is otherwise quite useful.
For now, the top-level function imset-decision-maker can be
called with any two sets of any cardinality and the user will receive
the a multiset. This multiset represents the prime form of the
interval multiset formed by applying Lewin's interval function (or
IFUNC) to the given sets.
Examples to follow.
Currently, the data for all combinations of a vast number of sets already exists in a static form elsewhere. It will eventually all be posted, though further analysis remains to be done to tease out interesting or noteworthy patterns.
Begins by loading various pitch-class sets, and then initializing globals that will hold pairs of the various pitch-class sets as laid out according to the Forte-system. The list is, unsurprisingly, the primary data structure for pitch classes rather than strings or vectors with no whitespace, largely because of Lispy propensities. The data could probably be re-formatted if performance ever became an issue.
As it is, compiling this whole file populates the necessary databases in decent enough time, despite reliance on inefficient data structures and some inevitably sloppy coding (still not bad for my first big Lisp program, I must say).