Sep19_2017.py

#

Music Computation Code from September 19, 2017
Original code here

#

from music21 import *

#

Assignment thoughts

#

Be careful using music21 reserved terms like note, pitch, and duration as variable names.

#

The first line redefines note as a music21 note object instead of the music21 note module, resulting in an error when trying to create new note objects.

note = note.Note('e-4') #Don't do this!
print(note)
note2 = note.Note('f4') #This line will result in an error.

#

Don't assume that parts always begin at index 1 in the stream hierarchy

#

In the file for BWV 348, the soprano part is at index 3 in the hierarchy. Better is to use the .parts property.

sBach = corpus.parse('bach/bwv348.mxl')
print('Elements at the highest level of the sBach stream hierarchy:')
for i, el in enumerate(sBach):
    print('sBach[' + str(i) + '] is ' + str(el))
print()
print('sBach[1] is', sBach[1])
print('sBach[3] is', sBach[3])
print('sBach.parts[0] is', sBach.parts[0])

#

Aim for the clearest and most efficient code.

#

Building a music21 stream note by note by note ...

myStream = stream.Stream()
myStream.append(note.Note('D4', quarterLength=1.5))
myStream.append(note.Note('F4', quarterLength=0.5))
myStream.append(note.Note('A4', quarterLength=1/3))
myStream.append(note.Note('G#4', quarterLength=1/3))
myStream.append(note.Note('A4', quarterLength=1/3))
myStream.append(note.Note('E-4', quarterLength=0.5))
myStream.append(note.Note('E-4', quarterLength=0.5))

#

Using lists of pitches and durations

pitches = ['d4', 'f4', 'a4', 'g#4', 'a4', 'e-4', 'e-4']
durations = [3/2, 1/2, 1/3, 1/3, 1/3, 1/2, 1/2]
s = stream.Stream()

#

and iterating through list indices

for i in range(len(pitches)):
    s.append(note.Note(pitches[i], quarterLength=durations[i]))

#

or iterating through tuples created by the built-in function zip.

for p, dur in zip(pitches, durations):
    s.append(note.Note(p, quarterLength=dur))

s.show('text')

#

Accessing music for parsing

#

Using shortened filenames in the music21 local corpus

path = 'ciconia/quod_jactatur' ##you can omit the file extension
sCiconia = corpus.parse(path)
sCiconia.show('text')
sCiconia.show()

#

Getting all paths to a composer in the music21 corpus

paths = corpus.getComposer('Schoenberg')
for path in paths:
    print(path)
    print()
    corpus.parse(path).show('text')
    print()

#

Converting a file on the hard drive; note the use of converter.parse rather than corpus.parse for files that are not in the music21 local corpus.

path = '/Users/ccallender/Desktop/waldstein.krn'
sWaldstein = converter.parse(path)
sWaldstein.show()

#

Getting a file from a url

url = 'http://kern.ccarh.org/cgi-bin/ksdata?location=users/craig/classical/beethoven/piano/sonata&file=sonata21-1.krn&format=kern'
sWaldstein = converter.parse(url)
sWaldstein.show()

#

Use of the os library for navigating file system/directory structure

#

The os library is a basic operating system interface. You need to import this library for the examples below.

import os

#

(In order to run the other blocks of code below you will need to download the europa subdirectory of the Essen folksong collection, unzip the file, modify the file paths as necessary depending on where the folder is installed on your machine.)

#

Retrieving all files of a given type in a single directory

dir_name = '/Users/ccallender/Desktop/europa/deutschl/altdeu1/'
file_names = os.listdir(dir_name)
songs = [name for name in file_names if name.endswith('.krn')]
print(len(songs))
for song in songs:
    path = dir_name + song
    sSong = converter.parse(path)
    sSong.show('text')
    print()

#

Retrieving all files of a given type in a directory and all its subdirectories

dir_name = '/Users/ccallender/Desktop/europa'
paths = []
for root, dirs, files in os.walk(dir_name):
    for file_name in files:
        if file_name.endswith('.krn'):
            paths.append(os.path.join(root, file_name))
print(len(paths)) #should be 6213, the number of songs in the europa directory

#

Helper functions for recreating "pitch promixity" finding from Huron ch. 5

#

helper function to count the frequency of items in a list

def update_count(keyList, count):

#

keyList is a list of keys; count is dictionary.

Returns an updated dictionary in which the value of any key is the number of times that key occurs in keyList added to the previous value for this key in count.

For example:

>>> count = {} # count is an empty dictionary
>>> mylist = [1, 2, 1, 2, 3, 1]
>>> update_count(mylist, count)
>>> count
{1: 3, 2: 2, 3: 1}

    for key in keyList:
        if key in count:
            count[key] += 1
        else:
            count[key] = 1

#

helper function for printing a dictionary with keys sorted

def print_dict(my_dict):
    result = ''
    for key in sorted(my_dict, key=lambda key: key):
        print(str(key) + " : " + str(my_dict[key]))