Skip to main content

a python framework for programming music applications

Project description

kord

kord is a python framework that provides programmers with a simple api for the creation of music-based applications. While it's mainly focused for theoretical purposes, some of it's more visually oriented features are well-suited for the generation of plucked-string instruments.

installation

The only dependency for kord is the pip package bestia, my own library for creating command-line applications. It will be automatically downloaded when you install using pip:

$   python3 -m pip install kord

api reference:

kord.notes module:

Note() class

Note instances are the lowest-level objects of the framework and have 3 main attributes:

* chr: str   ('C', 'D', 'E', 'F', 'G', 'A', 'B')
* alt: str   ('b', 'bb', '', '#', '##')
* oct: int   (0, 1, 2, 3, 4, 5, 6, 7, 8, 9)

Only the chr argument is required to create an instance. Arguments alt and oct will default to '' and 3 respectively.

>>> from kord.notes import Note
>>> e3, f3 = Note('e'), Note('f')
>>> e3, f3
(E³, F³) 
>>> Note('B', 'b', 7)
B♭⁷
>>> Note('C', '#', 0)
C♯⁰

Notes with double alterations are supported but Notes with triple (or more) alterations raise InvalidAlteration Exceptions:

>>> n3 = Note('A', 'bb', 1)
>>> n3
A𝄫¹
>>> n4 = Note('F', '##', 1)
>>> n4
F𝄪¹
>>> Note('G', '###')
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  ...
kord.errors.InvalidAlteration: ###

Intervals between note objects can be evaluated using the following operators:

-   < >   <= >=   == !=   << >>

These allow calculation of semitone deltas between notes as well as insights into their enharmonic relationships. Let's take a quick look at each operator separately:

- operator

This operator allows you to calculate the difference in semitones between two notes:

>>> f3 - e3
1
>>> Note('a', 8) - Note('c', 4)
57
>>> Note('a', 8) - Note('c', '##', 4)
55

< > operators

>>> f3
F³
>>> f3 > e3
True

!= operator

>>> n1 = Note('F', '#', 5)
>>> n2 = Note('G', 'b', 5)
>>> n1, n2
(F♯⁵, G♭⁵)
>>> n1 != n2
False

Given that python lacks a === operator, Notes can be compared for a "stricter" equality using their is_a() method:

>>> n3 == n4
True
>>> n3.is_a(n4)
False

This method directly compares Note attributes instead of their semitone interval. Set the ignore_oct argument appropriately for a less strict comparison:

>>> n1
F♯⁵
>>> n1.is_a(Note('F', '#', 3))
False
>>> n1.is_a(Note('F', '#', 3), ignore_oct=False)
True

kord.keys module:

class TonalKey(object):

The best way to describe a TonalKey object is basically as a Note object generator. You can use this class to create any theoretical arrangement of musical notes (ie. chords, scales).

Let us take a look at 2 examples using the 2 main categories of child class that inherit from TonalKey class and how they are defined:

class ChromaticKey(TonalKey):

The ChromaticKey class uses the TonalKey class as an interface while implementing it's own structure of intervals.

class ChromaticKey(TonalKey):

    root_intervals = (
        UNISON,
        MINOR_SECOND,
        MAJOR_SECOND,
        MINOR_THIRD,
        MAJOR_THIRD,
        PERFECT_FOURTH,
        AUGMENTED_FOURTH,
        PERFECT_FIFTH,
        MINOR_SIXTH,
        MAJOR_SIXTH,
        MINOR_SEVENTH,
        MAJOR_SEVENTH,
    )

We can initialize ChromaticKey objects on any given note and use the degree() method to obtain one of it's degrees. Using list index notation will achieve a similar result:

>>> from kord.keys import ChromaticKey
>>> c_chromatic = ChromaticKey('C')
>>> c_chromatic.degree(2)
C♯⁰
>>> c_chromatic[12]
B⁰

Perhaps the most interesting aspect of any TonalKey sub-class is it's ability to iterate over Note objects using one of their several generator methods. As an example, let's take a quick look at the scale() method:

>>> for note in c_chromatic.scale()
...   print(note, end=' ')
...
C⁰ C♯⁰ D⁰ D♯⁰ E⁰ F⁰ F♯⁰ G⁰ G♯⁰ A⁰ A♯⁰ B⁰ C¹ 

We can use the note_count= argument to specify to the scale generator the amount of notes to yield:

>>> for note in c_chromatic.scale(note_count=4):
...   print(note, end=' ')
...
C⁰ C♯⁰ D⁰ D♯⁰ 

The start_note= argument can be used to to start yielding from a specific note. This can be done even if the note is not part of the scale:

>>> from kord.notes import Note
>>> Ab = Note('A', 'b', 0)
>>> for note in c_chromatic.scale(note_count=6, start_note=Ab):
...   print(note, end=' ')
...
G♯⁰ A⁰ A♯⁰ B⁰ C¹ C♯¹ 

class DiatonicKey(TonalKey):

class MajorKey(DiatonicKey):

    root_intervals = (
        UNISON,
        MAJOR_SECOND,
        MAJOR_THIRD,
        PERFECT_FOURTH,
        PERFECT_FIFTH,
        MAJOR_SIXTH,
        MAJOR_SEVENTH,
    )

fretboard sample application

A sample application fretboard.py comes built-in with the kord framework and gives some insight into it's possibilities. It displays a representation of the fretboard of your instrument, tuned to your preference along as where to find the notes for any given mode for any given root note.

usage: fretboard.py [-h] [-m] [-i] [-t] [-f] [-v] ROOT

<<< Fretboard visualizer sample tool for the kord music framework >>>

positional arguments:
  ROOT                select a root note

optional arguments:
  -h, --help          show this help message and exit
  -m , --mode         mode to visualize: ['major', 'minor', 'natural_minor', 'melodic_minor', 'harmonic_minor', 'ionian', 'lydian', 'mixo', 'aeolian', 'dorian', 'phrygian', 'chromatic']
  -i , --instrument   instrument fretboard to visualize: ['banjo', 'guitar', 'bass', 'ukulele']
  -t , --tuning       instrument tuning: check your .json files for available options
  -f , --frets        number of displayed frets: [0, 1, 2, .. , 36]
  -v , --verbosity    application verbosity: [0, 1, 2]

You will find some pre-defined instrument tunings in the tunings directory in the form of .json files. Feel free to modify these or add your own following the correct syntax and they will immediately become available to the fretboard tool.

I'm pretty sure guitar/bass players may find it very handy as a means to develop their knowledge of their instrument. Have fun!

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

kord-1.6.tar.gz (19.0 kB view hashes)

Uploaded Source

Built Distribution

kord-1.6-py3-none-any.whl (18.0 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page