def test_get_prev_voice_indices():
settingsdict = {"num_voices": 2, "tet": 12}
er = er_preprocess.preprocess_settings(settingsdict)
# voice, pitch, onset, dur, evaluates_to
notes1 = [
(0, 60, 1.5, 0.5, []),
(1, 64, 1.75, 0.25, [0]),
(0, 64, 2, 0.25, []),
(1, 64, 2, 0.5, [0]),
]
notes2 = [
(1, 64, 1.75, 0.25, []),
(0, 60, 1.5, 0.5, [1]),
(1, 64, 2, 0.5, []),
(0, 64, 2, 0.25, [1]),
]
for notes in (notes1, notes2):
score = er_classes.Score(num_voices=er.num_voices, tet=er.tet)
for (v, p, a, d, l) in notes: # pylint: disable=invalid-name
try:
assert (
er_misc_funcs.get_prev_voice_indices(score, a, d) == l
), "er_misc_funcs.get_prev_voice_indices(score, a, d) != l"
except: # pylint: disable=bare-except
exc_type, exc_value, exc_traceback = sys.exc_info()
traceback.print_exception(
exc_type, exc_value, exc_traceback, file=sys.stdout
)
breakpoint()
score.add_note(v, p, a, d)
def test_get_kern():
# A case that was failing:
notes = (
((60, 0, 1.75), (61, 1.75, 1.5), (62, 3.25, 0.75), (63, 4, 1)),
((48, 0, 1), (49, 1, 1), (50, 2, 1), (51, 3, 1), (52, 4, 1)),
)
score = er_classes.Score(
num_voices=len(notes),
tet=12,
time_sig=(2, 4),
harmony_len=(2,),
total_len=6,
)
for voice_i, voice in enumerate(notes):
for pitch, onset, dur in voice:
score.add_note(voice_i, pitch, onset, dur)
kern = er_output_notation.get_kern(score)
_, temp_path = tempfile.mkstemp()
# If the input is incorrect, verovio should fail
try:
er_misc_funcs.silently_run_process(
["verovio", "-o", temp_path, "-"], stdin=kern
)
except er_misc_funcs.ProcError:
os.remove(temp_path)
raise
os.remove(temp_path)
def test_get_sounding_pitches():
settingsdict = {"num_voices": 2, "tet": 12}
er = er_preprocess.preprocess_settings(settingsdict, silent=True)
score = er_classes.Score(num_voices=er.num_voices, tet=er.tet)
# voice, pitch, onset, dur
notes = [
(0, 57, 0, 0.75),
(1, 58, 0.25, 0.25),
(1, 59, 0.5, 0.75),
(0, 60, 0.75, 0.75),
]
for (v, p, a, d) in notes: # pylint: disable=invalid-name
score.add_note(v, p, a, d)
# onset, dur, min_onset, min_dur, sounding_pitches
test_items = [
(0.5, 0, 0, 0, [57, 59]),
(0.5, 1, 0, 0, [57, 59, 60]),
(0.5, 1, 0, 1, []),
(0.5, 1, 0.5, 0, [59, 60]),
]
for (a, d, m1, m2, sp) in test_items: # pylint: disable=invalid-name
assert (
score.get_sounding_pitches(a,
end_time=a + d,
min_onset=m1,
min_dur=m2) == sp
), f"score.get_sounding_pitches({a}, dur={d}, min_onset={m1}, min_dur={m2}) != {sp}"
def test_generic_transpose():
settingsdict = {
"num_voices": 1,
"tet": 12,
"harmony_len": 2,
"foot_pcs": [0],
"interval_cycle": 3,
"num_harmonies": 4,
}
er = er_preprocess.preprocess_settings(settingsdict, silent=True)
score = er_classes.Score(
num_voices=er.num_voices,
tet=er.tet,
harmony_len=er.harmony_len,
total_len=16,
)
notes = [
er_classes.Note(p + 60, i * 1, 1)
for i, p in enumerate([0, 2, 3, 5, 6, 8, 9, 11])
]
for note in notes: # pylint: disable=invalid-name
score.add_note(0, note)
# TODO review the rationale behind max_interval
# should it have a default value?
# what is its source in the script?
score.transpose(2, er=er, max_interval=7)
for i, p in enumerate([4, 5, 7, 8, 10, 11, 13, 14]):
assert notes[i].pitch == 60 + p, "notes[i].pitch != 60 + p"
score.transpose(-2, er=er, max_interval=7)
for i, p in enumerate([0, 2, 3, 5, 6, 8, 9, 11]):
assert notes[i].pitch == 60 + p, "notes[i].pitch != 60 + p"
score.transpose(7, er=er, max_interval=5)
for i, p in enumerate([0, 2, 3, 5, 6, 8, 9, 11]):
assert notes[i].pitch == 60 + p, "notes[i].pitch != 60 + p"
score.transpose(-7, er=er, max_interval=6)
for i, p in enumerate([0, 2, 3, 5, 6, 8, 9, 11]):
assert notes[i].pitch == 60 + p, "notes[i].pitch != 60 + p"
score.transpose(9, er=er, max_interval=7)
for i, p in enumerate([4, 5, 7, 8, 10, 11, 13, 14]):
assert notes[i].pitch == 60 + p, "notes[i].pitch != 60 + p"
score.transpose(-9, er=er, max_interval=6)
for i, p in enumerate([0, 2, 3, 5, 6, 8, 9, 11]):
assert notes[i].pitch == 60 + p, "notes[i].pitch != 60 + p"
score.transpose(2)
for i, p in enumerate([0, 2, 3, 5, 6, 8, 9, 11]):
assert notes[i].pitch == 62 + p, "notes[i].pitch != 60 + p"
def test_score_misc_attrs():
settingsdict = {"num_voices": 2, "tet": 12}
er = er_preprocess.preprocess_settings(settingsdict, silent=True)
score = er_classes.Score(num_voices=er.num_voices, tet=er.tet)
# voice, pitch, onset, dur
notes = [
(0, 57, 0, 0.75),
(1, 58, 0.25, 0.25),
(1, 59, 0.5, 0.75),
(0, 60, 0.75, 0.75),
]
assert score.total_dur == 0
for (v, p, a, d) in notes: # pylint: disable=invalid-name
score.add_note(v, p, a, d)
assert score.total_dur == 1.5
for note in score.voices[0][0.75]:
score.voices[0].remove_note(note)
assert score.total_dur == 1.25
score.add_note(1, 42, 250, 17)
assert score.total_dur == 267
def test_get_harmony_i():
settingsdict = {"harmony_len": 4, "tet": 12}
er = er_preprocess.preprocess_settings(settingsdict, silent=True)
score = er_classes.Score(
num_voices=er.num_voices,
tet=er.tet,
harmony_len=er.harmony_len,
total_len=16,
)
assert score.get_harmony_i(0) == 0
assert score.get_harmony_i(3.75) == 0
assert score.get_harmony_i(4.0) == 1
assert score.get_harmony_i(4.01) == 1
assert score.get_harmony_i(16) == 3
assert score.get_harmony_i(27) == 3
ht = score.get_harmony_times(0)
assert (ht.start_time, ht.end_time, ht.i) == (0, 4, 0)
ht = score.get_harmony_times(3)
# end time of final harmony should be None
assert (ht.start_time, ht.end_time, ht.i) == (12, None, 3)
def test_get_prev_and_last_notes():
settingsdict = {"num_voices": 2, "tet": 12}
er = er_preprocess.preprocess_settings(settingsdict, silent=True)
# voice, pitch, onset, dur, evaluates_to
notes = [
(0, 57, 0, 0.75, None),
(1, 58, 0.25, 0.25, None),
(1, 59, 0.5, 0.75, 58),
(0, 60, 0.75, 0.75, 57),
(0, 61, 3.0, 1.0, 60),
]
score = er_classes.Score(num_voices=er.num_voices, tet=er.tet)
for (v, p, a, d, ev) in notes: ##pylint: disable=invalid-name
if ev is None:
assert (score.voices[v].get_prev_note(a) is
None), "score.voices[v].get_prev_note(a) is not None"
assert (score.voices[v].get_prev_pitch(a) == -1
), "score.voices[v].get_prev_pitch(a) != -1"
else:
assert (score.voices[v].get_prev_note(a).pitch == ev
), "score.voices[v].get_prev_note(a).pitch != ev"
assert (score.voices[v].get_prev_pitch(a) == ev
), "score.voices[v].get_prev_pitch(a) != ev"
score.add_note(v, p, a, d)
assert (score.voices[v].get_prev_note(a + 1).pitch == p
), "score.voices[v].get_prev_note(a + 1).pitch != p"
if ev is None:
assert (score.voices[v].get_prev_note(a - 0.1) is
None), "score.voices[v].get_prev_note(a - .1) is not None"
assert (score.voices[v].get_prev_pitch(a - 0.1) == -1
), "score.voices[v].get_prev_pitch(a - .1) != -1"
else:
assert (score.voices[v].get_prev_note(a - 0.1).pitch == ev
), "score.voices[v].get_prev_note(a - .1).pitch != ev"
assert (score.voices[v].get_prev_pitch(a - 0.1) == ev
), "score.voices[v].get_prev_pitch(a - .1) != ev"
assert (score.voices[v].get_last_n_notes(1, a)[0].pitch == p
), "score.voices[v].get_last_n_notes(1, a)[0].pitch != p"
assert (score.voices[v].get_last_n_pitches(
1, a)[0] == p), "score.voices[v].get_last_n_pitches(1, a)[0] != p"
def test_voices_to_tracks():
base_settings = {
"randomly_distribute_between_choirs": False,
"num_voices": 3,
}
er = er_preprocess.preprocess_settings(base_settings)
# voice, pitch, onset, dur
notes = [
(0, 60, 0, 1),
(1, 65, 0, 1),
(2, 67, 0, 1),
]
score = er_classes.Score(num_voices=er.num_voices, tet=er.tet)
for (v, p, a, d) in notes:
score.add_note(v, p, a, d)
er_choirs.assign_choirs(er, score)
# note that by default reverse_tracks would be False, which would mean that
# the choir assignments would be reversed (e.g., with three voices, track
# 0 would become track 2). To avoid that hassle, we pass
# reverse_tracks=False
mf = er_midi.write_er_midi(er,
score,
"test_voices_to_tracks",
reverse_tracks=False,
return_mf=True)
try:
for track_i, track in enumerate(mf.tracks):
for msg in track:
if msg.type in ("note_on", "note_off", "program_change"):
# the meta track is track 0, so we need to subtract 1
# to get the right element from er.choir_assignments
assert msg.channel == er.choir_assignments[track_i - 1]
except: # pylint: disable=bare-except
exc_type, exc_value, exc_traceback = sys.exc_info()
traceback.print_exception(exc_type,
exc_value,
exc_traceback,
file=sys.stdout)
breakpoint()
def test_check_harmonic_intervals():
settingsdict = {
"num_voices": 2,
"tet": 12,
"forbidden_interval_classes": [
0,
],
}
er = er_preprocess.preprocess_settings(settingsdict)
# voice, pitch, onset, dur, evaluates_to
notes1 = [
(0, 60, 1.5, 0.5, True),
(1, 64, 1.75, 0.25, True),
(0, 64, 2, 0.25, True),
(1, 76, 2, 0.5, False),
]
notes2 = [
(1, 52, 1.75, 0.25, True),
(0, 60, 1.5, 0.5, True),
(1, 64, 2, 0.5, True),
(0, 64, 2, 0.25, False),
]
for notes in (notes1, notes2):
score = er_classes.Score(num_voices=er.num_voices, tet=er.tet)
for (v, p, a, d, b) in notes: # pylint: disable=invalid-name
try:
assert (er_make2.check_harmonic_intervals(
er, score, p, a, d, v) is
b), ("er_make2.check_harmonic_intervals"
f"(er, score, {p}, {a}, {d}, {v}) is not {b}")
except: # pylint: disable=bare-except
exc_type, exc_value, exc_traceback = sys.exc_info()
traceback.print_exception(exc_type,
exc_value,
exc_traceback,
file=sys.stdout)
breakpoint()
score.add_note(v, p, a, d)
settingsdict = {
"num_voices": 3,
"tet": 12,
"forbidden_intervals": [
0,
],
"forbidden_interval_classes": [],
}
er = er_preprocess.preprocess_settings(settingsdict)
notes1 = [
(0, 60, 0, 1.0, True),
(1, 72, 0, 1.0, True),
(2, 60, 0, 0.5, False),
(2, 72, 0, 0.5, False),
]
for notes in (notes1, ):
score = er_classes.Score(num_voices=er.num_voices, tet=er.tet)
for (v, p, a, d, b) in notes: # pylint: disable=invalid-name
try:
assert (er_make2.check_harmonic_intervals(
er, score, p, a, d, v) is
b), ("er_make2.check_harmonic_intervals"
f"(er, score, {p}, {a}, {d}, {v}) is not {b}")
except: # pylint: disable=bare-except
exc_type, exc_value, exc_traceback = sys.exc_info()
traceback.print_exception(exc_type,
exc_value,
exc_traceback,
file=sys.stdout)
breakpoint()
score.add_note(v, p, a, d)
def test_between():
settingsdict = {"num_voices": 2, "tet": 12}
er = er_preprocess.preprocess_settings(settingsdict, silent=True)
score = er_classes.Score(num_voices=er.num_voices, tet=er.tet)
# voice, pitch, onset, dur
notes = [
(0, 57, 0, 0.75),
(1, 58, 0.25, 0.25),
(1, 59, 0.5, 0.75),
(0, 60, 0.75, 0.5),
(0, 61, 0.75, 0.75),
(0, 62, 1.5, 0.5),
(1, 63, 1.5, 0.5),
(0, 64, 2.0, 1.0),
]
note_objs = [
er_classes.Note(pitch, onset, dur) for _, pitch, onset, dur in notes
]
voice_is = [v for v, _, _, _ in notes]
for v, n in zip(voice_is, note_objs):
score.add_note(v, n)
more_notes = [n for n in score.between()]
assert note_objs == more_notes
more_notes = [n for n in score.between(start_time=0.5)]
assert note_objs[2:] == more_notes
more_notes = [n for n in score.between(start_time=0.4)]
assert note_objs[2:] == more_notes
more_notes = [n for n in score.between(end_time=2.0)]
assert note_objs[:7] == more_notes
more_notes = [n for n in score.between(end_time=1.75)]
assert note_objs[:7] == more_notes
more_notes = [n for n in score.between(start_time=0.75, end_time=1.5)]
assert note_objs[3:5] == more_notes
more_notes = [n for n in score.between(start_time=0.8, end_time=1.2)]
assert more_notes == []
note_objs_by_onset = [
[note_objs[0]],
[note_objs[1]],
[note_objs[2]],
note_objs[3:5],
note_objs[5:7],
[note_objs[7]],
]
more_onsets = [o for o in score.notes_by_onset_between()]
assert more_onsets == note_objs_by_onset
more_onsets = [n for n in score.notes_by_onset_between(start_time=0.5)]
assert note_objs_by_onset[2:] == more_onsets
more_onsets = [n for n in score.notes_by_onset_between(start_time=0.4)]
assert note_objs_by_onset[2:] == more_onsets
more_onsets = [n for n in score.notes_by_onset_between(end_time=2.0)]
assert note_objs_by_onset[:5] == more_onsets
more_onsets = [n for n in score.notes_by_onset_between(end_time=1.75)]
assert note_objs_by_onset[:5] == more_onsets
more_onsets = [
n for n in score.notes_by_onset_between(start_time=0.75, end_time=1.5)
]
assert note_objs_by_onset[3:4] == more_onsets
more_onsets = [
n for n in score.notes_by_onset_between(start_time=0.8, end_time=1.2)
]
assert more_onsets == []
def _init_score_with_voices_and_notes(voices_and_notes, er):
score = er_classes.Score(num_voices=er.num_voices, tet=er.tet)
for voice_i, voice in enumerate(voices_and_notes):
for note_i, note in enumerate(voice):
score.add_note(voice_i, note, 1 * note_i, note)
return score