def testComplexMode100(self):
m = MIDITime(base_octave=5, octave_range=3)
pct = m.linear_scale_pct(0, 5.7, 5.7, reverse=False)
self.assertEqual(m.scale_to_note(pct, ['D', 'E', 'F', 'G', 'A', 'Bb', 'C']), 'Bb7')
def testSimpleMode70(self):
m = MIDITime(base_octave=5, octave_range=3)
pct = m.linear_scale_pct(0, 5.7, 4.0, reverse=False)
self.assertEqual(m.scale_to_note(pct, ['C', 'D', 'E', 'F', 'G', 'A', 'B']), 'C7')
def testComplexMode100(self):
m = MIDITime(base_octave=5, octave_range=3)
pct = m.linear_scale_pct(0, 5.7, 5.7, reverse=False)
self.assertEqual(
m.scale_to_note(pct, ['D', 'E', 'F', 'G', 'A', 'Bb', 'C']), 'Bb7')
def testSimpleMode70(self):
m = MIDITime(base_octave=5, octave_range=3)
pct = m.linear_scale_pct(0, 5.7, 4.0, reverse=False)
self.assertEqual(
m.scale_to_note(pct, ['C', 'D', 'E', 'F', 'G', 'A', 'B']), 'C7')
class bomb2midi(object):
''' Submitted by Jennifer LaFleur. '''
epoch = datetime(
1945, 1, 1) # Not actually necessary, but optional to specify your own
mymidi = None
min_value = 0
max_value = 5.7
tempo = 120
min_attack = 30
max_attack = 255
min_duration = 1
max_duration = 5
seconds_per_year = 3
c_major = ['C', 'D', 'E', 'F', 'G', 'A', 'B']
c_minor = ['C', 'D', 'Eb', 'F', 'G', 'Ab', 'Bb']
a_minor = ['A', 'B', 'C', 'D', 'E', 'F', 'F#', 'G', 'G#']
c_blues_minor = ['C', 'Eb', 'F', 'F#', 'G', 'Bb']
d_minor = ['D', 'E', 'F', 'G', 'A', 'Bb', 'C']
c_gregorian = ['C', 'D', 'Eb', 'F', 'G', 'Ab', 'A', 'Bb']
current_key = c_major
base_octave = 2
octave_range = 5
def __init__(self):
self.csv_to_miditime()
def read_csv(self, filepath):
csv_file = open(filepath, 'rU')
return csv.DictReader(csv_file, delimiter=',', quotechar='"')
def remove_weeks(self, csv_obj):
return [r for r in csv_obj if r['Date'] not in ['']]
def round_to_quarter_beat(self, input):
return round(input * 4) / 4
def make_notes(self, data_timed, data_key):
note_list = []
start_time = data_timed[0]['beat']
for d in data_timed:
note_list.append([
self.round_to_quarter_beat(d['beat'] - start_time),
self.data_to_pitch_tuned(d[data_key]),
100,
#mag_to_attack(d['magnitude']), # attack
1 # duration, in beats
])
return note_list
def csv_to_miditime(self):
raw_data = list(self.read_csv('data/bombs.csv'))
filtered_data = self.remove_weeks(raw_data)
self.mymidi = MIDITime(self.tempo, 'bombtest_log.mid',
self.seconds_per_year, self.base_octave,
self.octave_range, self.epoch)
self.minimum = self.mymidi.get_data_range(filtered_data, 'Yieldnum')[0]
self.maximum = self.mymidi.get_data_range(filtered_data, 'Yieldnum')[1]
timed_data = []
for r in filtered_data:
python_date = datetime.strptime(r["Date"], "%m/%d/%Y")
days_since_epoch = self.mymidi.days_since_epoch(python_date)
beat = self.mymidi.beat(days_since_epoch)
timed_data.append({
'days_since_epoch': days_since_epoch,
'beat': beat,
'BombYieldMillions': float(r['Yieldnum'])
})
note_list = self.make_notes(timed_data, 'BombYieldMillions')
# Add a track with those notes
self.mymidi.add_track(note_list)
# Output the .mid file
self.mymidi.save_midi()
def data_to_pitch_tuned(self, datapoint):
# Where does this data point sit in the domain of your data? (I.E. the min magnitude is 3, the max in 5.6). In this case the optional 'True' means the scale is reversed, so the highest value will return the lowest percentage.
#scale_pct = self.mymidi.linear_scale_pct(0, self.maximum, datapoint)
# Another option: Linear scale, reverse order
# scale_pct = self.mymidi.linear_scale_pct(0, self.maximum, datapoint, True)
# print 10**self.maximum
# Another option: Logarithmic scale, reverse order
scale_pct = self.mymidi.log_scale_pct(0, self.maximum, datapoint, True,
'log')
# Pick a range of notes. This allows you to play in a key.
mode = self.current_key
#Find the note that matches your data point
note = self.mymidi.scale_to_note(scale_pct, mode)
#Translate that note to a MIDI pitch
midi_pitch = self.mymidi.note_to_midi_pitch(note)
print scale_pct, note
return midi_pitch
def mag_to_attack(self, datapoint):
# Where does this data point sit in the domain of your data? (I.E. the min magnitude is 3, the max in 5.6). In this case the optional 'True' means the scale is reversed, so the highest value will return the lowest percentage.
scale_pct = self.mymidi.linear_scale_pct(0, self.maximum, datapoint)
#max_attack = 10
adj_attack = (1 - scale_pct) * max_attack + 70
#adj_attack = 100
return adj_attack
class bomb2midi(object):
''' Submitted by Jennifer LaFleur. '''
epoch = datetime(1945, 1, 1) # Not actually necessary, but optional to specify your own
mymidi = None
min_value = 0
max_value = 5.7
tempo = 120
min_attack = 30
max_attack = 255
min_duration = 1
max_duration = 5
seconds_per_year = 3
c_major = ['C', 'D', 'E', 'F', 'G', 'A', 'B']
c_minor = ['C', 'D', 'Eb', 'F', 'G', 'Ab', 'Bb']
a_minor = ['A', 'B', 'C', 'D', 'E', 'F', 'F#', 'G', 'G#']
c_blues_minor = ['C', 'Eb', 'F', 'F#', 'G', 'Bb']
d_minor = ['D', 'E', 'F', 'G', 'A', 'Bb', 'C']
c_gregorian = ['C', 'D', 'Eb', 'F', 'G', 'Ab', 'A', 'Bb']
current_key = c_major
base_octave = 2
octave_range = 5
def __init__(self):
self.csv_to_miditime()
def read_csv(self, filepath):
csv_file = open(filepath, 'rU')
return csv.DictReader(csv_file, delimiter=',', quotechar='"')
def remove_weeks(self, csv_obj):
return [r for r in csv_obj if r['Date'] not in ['']]
def round_to_quarter_beat(self, input):
return round(input * 4) / 4
def make_notes(self, data_timed, data_key):
note_list = []
start_time = data_timed[0]['beat']
for d in data_timed:
note_list.append([
self.round_to_quarter_beat(d['beat'] - start_time),
self.data_to_pitch_tuned(d[data_key]),
100,
#mag_to_attack(d['magnitude']), # attack
1 # duration, in beats
])
return note_list
def csv_to_miditime(self):
raw_data = list(self.read_csv('data/bombs.csv'))
filtered_data = self.remove_weeks(raw_data)
self.mymidi = MIDITime(self.tempo, 'bombtest_log.mid', self.seconds_per_year, self.base_octave, self.octave_range, self.epoch)
self.minimum = self.mymidi.get_data_range(filtered_data, 'Yieldnum')[0]
self.maximum = self.mymidi.get_data_range(filtered_data, 'Yieldnum')[1]
timed_data = []
for r in filtered_data:
python_date = datetime.strptime(r["Date"], "%m/%d/%Y")
days_since_epoch = self.mymidi.days_since_epoch(python_date)
beat = self.mymidi.beat(days_since_epoch)
timed_data.append({
'days_since_epoch': days_since_epoch,
'beat': beat,
'BombYieldMillions': float(r['Yieldnum'])
})
note_list = self.make_notes(timed_data, 'BombYieldMillions')
# Add a track with those notes
self.mymidi.add_track(note_list)
# Output the .mid file
self.mymidi.save_midi()
def data_to_pitch_tuned(self, datapoint):
# Where does this data point sit in the domain of your data? (I.E. the min magnitude is 3, the max in 5.6). In this case the optional 'True' means the scale is reversed, so the highest value will return the lowest percentage.
#scale_pct = self.mymidi.linear_scale_pct(0, self.maximum, datapoint)
# Another option: Linear scale, reverse order
# scale_pct = self.mymidi.linear_scale_pct(0, self.maximum, datapoint, True)
# print 10**self.maximum
# Another option: Logarithmic scale, reverse order
scale_pct = self.mymidi.log_scale_pct(0, self.maximum, datapoint, True, 'log')
# Pick a range of notes. This allows you to play in a key.
mode = self.current_key
#Find the note that matches your data point
note = self.mymidi.scale_to_note(scale_pct, mode)
#Translate that note to a MIDI pitch
midi_pitch = self.mymidi.note_to_midi_pitch(note)
print scale_pct, note
return midi_pitch
def mag_to_attack(self, datapoint):
# Where does this data point sit in the domain of your data? (I.E. the min magnitude is 3, the max in 5.6). In this case the optional 'True' means the scale is reversed, so the highest value will return the lowest percentage.
scale_pct = self.mymidi.linear_scale_pct(0, self.maximum, datapoint)
#max_attack = 10
adj_attack = (1 - scale_pct) * max_attack + 70
#adj_attack = 100
return adj_attack