Py3

tomato/analyzer.py

@@ -25,8 +25,7 @@
 # PhD thesis, Universitat Pompeu Fabra, Barcelona, Spain.
 
 from abc import ABCMeta, abstractmethod, abstractproperty
-from io import IO
-from six import iteritems
+from .io import IO
 import logging
 import warnings
 
@@ -73,7 +72,7 @@ def _parse_inputs(self, **kwargs):
         # initialize precomputed_features with the available analysis
         precomputed_features = dict((f, None)
                                     for f in self._inputs)
-        for feature, val in iteritems(kwargs):
+        for feature, val in kwargs.items():
             if feature not in self._inputs:
                 warn_str = u'Unrelated feature {0:s}: It will be kept, ' \
                            u'but it will not be used in the audio analysis.' \

tomato/audio/ahenk.py

@@ -25,7 +25,6 @@
 # PhD thesis, Universitat Pompeu Fabra, Barcelona, Spain.
 
 import numpy as np
-from future.utils import iteritems
 from ..io import IO
 
 
@@ -70,7 +69,7 @@ def identify(cls, tonic_freq, symbol_in):
                                         'unit': 'cent'}}
 
         # get the ahenk
-        for ahenk_slug, val in iteritems(ahenks):
+        for ahenk_slug, val in ahenks.items():
             if val['cent_transposition'] == mod_cent_approx:
                 ahenk_dict['name'] = val['name']
                 ahenk_dict['slug'] = ahenk_slug
@@ -95,7 +94,7 @@ def _get_tonic_symbol(cls, symbol_in, tonic_dict):
     @staticmethod
     def _get_tonic_symbol_from_makam(symbol_in, tonic_dict):
         tonic_bolahenk_freq = tonic_symbol = None
-        for sym, val in iteritems(tonic_dict):
+        for sym, val in tonic_dict.items():
             if symbol_in in val['makams']:
                 tonic_symbol = sym
                 tonic_bolahenk_freq = val['bolahenk_freq']

tomato/audio/audioanalyzer.py

@@ -156,7 +156,7 @@ def _call_audio_metadata(self, audio_meta, filepath):
             audio_meta = self.crawl_musicbrainz_metadata(audio_meta)
         elif not isinstance(audio_meta, dict):
             warn_str = 'The "metadata" input can be "False" (skipped), ' \
-                       '"basestring" (MBID input), "None" (attempt to get ' \
+                       '"str" (MBID input), "None" (attempt to get ' \
                        'the MBID from audio file tags) or "dict" (already ' \
                        'computed)'
             warnings.warn(warn_str, stacklevel=2)
@@ -183,7 +183,12 @@ def _get_makam_tonic_training():
             'models', 'makam_tonic_estimation',
             'training_model--pcd--7_5--15_0--dlfm2016.pkl')
 
-        return pickle.load(open(makam_tonic_training_file))
+        try:  # python 3
+            # the pickle was made via python 2, we have to specify the encoding
+            return pickle.load(open(makam_tonic_training_file, 'rb'),
+                               encoding='latin1')
+        except TypeError:  # python 2
+            return pickle.load(open(makam_tonic_training_file, 'rb'))
 
     def crawl_musicbrainz_metadata(self, rec_in):
         try:

tomato/audio/notemodel.py

@@ -76,7 +76,7 @@ def _stable_pitches_to_notes(self, stable_pitches_hz,
         for stable_pitch_cent, stable_pitch_hz in zip(stable_pitches_cent,
                                                       stable_pitches_hz):
             note_cent = TonicLastNote.find_nearest(
-                theoretical_intervals.values(), stable_pitch_cent)
+                list(theoretical_intervals.values()), stable_pitch_cent)
 
             if abs(stable_pitch_cent - note_cent) < self.pitch_threshold:
                 for key, val in theoretical_intervals.items():
@@ -124,7 +124,8 @@ def _get_theoretical_intervals_to_search(self, makam):
 
         # Remove the notes neighboring the scale notes
         keep_notes = natural_notes + key_signature
-        for note_name in note_dict.keys():
+        for note_name in list(note_dict.keys()):  # get a copy to avoid
+            # mutating note_dict during iteration
             if note_name in keep_notes:
                 if note_name[2:4] in ['b5', 'b4']:
                     rm_note_symbol = self.note_letters[
@@ -175,13 +176,14 @@ def _get_theoretical_intervals_to_search(self, makam):
     def _get_natural_notes(self, key_signature, note_dict):
         close_natural_notes = self._get_close_natural_notes(key_signature)
         # get the natural notes that are not close to the key signature
-        natural_notes = [n for n in note_dict.keys() if len(n) == 2 and n
+        natural_notes = [n for n in list(note_dict.keys()) if len(n) == 2 and n
                          not in close_natural_notes]
         return natural_notes
 
     def _get_extended_key_signature(self, key_sig_in, note_dict):
         key_signature = deepcopy(key_sig_in)
-        for note in note_dict.keys():  # extend the key signature to octaves
+        for note in list(note_dict.keys()):  # extend the key signature to
+            # octaves
             if any([self._is_same_pitch_class(note, ks)
                     for ks in key_signature]):
                 key_signature.append(note)
@@ -212,7 +214,7 @@ def plot(pitch_distribution, stable_notes):
             matplotlib.ticker.FormatStrFormatter('%d'))
         ax.set_xlim([min(pd_copy.bins), max(pd_copy.bins)])
         ax.set_xticks([note['stable_pitch']['value']
-                       for note in stable_notes.values()])
+                       for note in list(stable_notes.values())])
 
         # recording distribution
         ax.plot(pd_copy.bins, pd_copy.vals,

tomato/bincaller.py

@@ -28,7 +28,7 @@
 import os
 import subprocess
 
-from io import IO
+from .io import IO
 
 
 class BinCaller(object):

tomato/joint/jointanalyzer.py

@@ -26,7 +26,7 @@
 
 import json
 from scipy.io import savemat
-from six.moves import cStringIO
+from six import BytesIO
 import tempfile
 from copy import deepcopy
 import timeit
@@ -192,7 +192,7 @@ def extract_tonic_tempo(self, score_filename='', score_data=None,
             '.json', json.dumps(score_data))
 
         # matlab
-        matout = cStringIO()
+        matout = BytesIO()
         savemat(matout, audio_pitch)
 
         temp_pitch_file = IO.create_temp_file(
@@ -207,7 +207,7 @@ def extract_tonic_tempo(self, score_filename='', score_data=None,
         out, err = _mcr_caller.call(callstr)
 
         # check the MATLAB output
-        if "Tonic-Tempo-Tuning Extraction took" not in out:
+        if "Tonic-Tempo-Tuning Extraction took" not in out.decode('utf-8'):
             IO.remove_temp_files(
                 temp_score_data_file, temp_pitch_file, temp_out_folder)
             raise RuntimeError("Score-informed tonic, tonic and tuning "
@@ -278,7 +278,7 @@ def align_audio_score(self, score_filename='', score_data=None,
             '.json', json.dumps({'scoreInformed': audio_tempo_}))
 
         # matlab
-        matout = cStringIO()
+        matout = BytesIO()
         savemat(matout, audio_pitch)
 
         temp_pitch_file = IO.create_temp_file(
@@ -297,7 +297,7 @@ def align_audio_score(self, score_filename='', score_data=None,
         out, err = _mcr_caller.call(callstr)
 
         # check the MATLAB output
-        if "Audio-score alignment took" not in out:
+        if "Audio-score alignment took" not in out.decode('utf-8'):
             IO.remove_temp_files(
                 temp_score_data_file, temp_tonic_file, temp_tempo_file,
                 temp_pitch_file, temp_out_folder)

tomato/metadata/recording.py

@@ -29,7 +29,6 @@
 from . work import Work
 from . instrumentationvoicing import InstrumentationVoicing
 import musicbrainzngs as mb
-from six import iteritems
 import logging
 logging.basicConfig(level=logging.INFO)
 
@@ -60,9 +59,7 @@ def from_musicbrainz(self, audio_in):
         audio_meta['url'] = u'http://musicbrainz.org/recording/{}'.format(
             audio_meta['mbid'])
 
-        meta = mb.get_recording_by_id(
-            audio_meta['mbid'], includes=['artists', 'artist-rels', 'releases',
-                                          'tags', 'work-rels'])['recording']
+        meta = mb.get_recording_by_id(audio_meta['mbid'], includes=['artists', 'artist-rels', 'releases', 'tags', 'work-rels'])['recording']
         audio_meta['title'] = meta['title']
 
         # releases
@@ -106,7 +103,7 @@ def _get_attributes_from_works(self, audio_meta):
     @staticmethod
     def _get_recording_attribute_tags(audio_meta, meta):
         attributetags = Attribute.get_attrib_tags(meta)
-        for key, vals in iteritems(attributetags):
+        for key, vals in attributetags.items():
             for val in vals:  # add the source
                 val['source'] = 'http://musicbrainz.org/recording/' + \
                                 audio_meta['mbid']
@@ -121,7 +118,7 @@ def _get_recording_attribute_tags(audio_meta, meta):
     def get_file_metadata(filepath):
         audiofile = eyed3.load(filepath)
         mbid = audiofile.tag.unique_file_ids.get(
-            'http://musicbrainz.org').data[-36:]
+            'http://musicbrainz.org').data[-36:].decode('utf-8')
         duration = audiofile.info.time_secs
         sampling_frequency = audiofile.info.sample_freq
         bit_rate = audiofile.info.mp3_header.bit_rate

tomato/metadata/work.py

@@ -24,10 +24,10 @@
 # scores for the description and discovery of Ottoman-Turkish makam music.
 # PhD thesis, Universitat Pompeu Fabra, Barcelona, Spain.
 
-from . attribute import Attribute
 import json
 import warnings
-import urllib
+from six.moves.urllib.request import urlopen
+from . attribute import Attribute
 
 import musicbrainzngs as mb
 mb.set_useragent("Makam corpus metadata", "1.2.0", "compmusic.upf.edu")
@@ -71,7 +71,7 @@ def from_musicbrainz(self, mbid):
     def _add_scores(data, mbid):
         score_work_url = 'https://raw.githubusercontent.com/MTG/SymbTr/' \
                          'master/symbTr_mbid.json'
-        response = urllib.urlopen(score_work_url)
+        response = urlopen(score_work_url)
         score_work = json.loads(response.read())
         data['scores'] = []
         for sw in score_work:

tomato/plotter.py

@@ -28,7 +28,6 @@
 import copy
 import matplotlib.pyplot as plt
 import matplotlib.patches as patches
-from six import iteritems
 from matplotlib import gridspec
 from matplotlib.ticker import FixedLocator
 
@@ -312,7 +311,7 @@ def _plot_note_models(cls, ax2, note_models, pitch_distribution):
     @staticmethod
     def _plot_stable_pitches(ax, max_rel_occur, note_models, max_pd_height):
         ytick_vals = []
-        for note_symbol, note in iteritems(note_models):
+        for note_symbol, note in note_models.items():
             plot_bool = (note['stable_pitch']['value'] and
                          note['rel_occur'] > max_rel_occur * 0.15)
             if plot_bool:
@@ -346,7 +345,7 @@ def _plot_stable_pitches(ax, max_rel_occur, note_models, max_pd_height):
     @staticmethod
     def _get_relative_note_occurences(note_models, pitch_distribution):
         max_rel_occur = 0
-        for note_symbol, note in iteritems(note_models):
+        for note_symbol, note in note_models.items():
             try:
                 # get the relative occurrence of each note from the pitch
                 # distribution
@@ -370,7 +369,7 @@ def _get_relative_note_occurences(note_models, pitch_distribution):
 
     @staticmethod
     def _plot_note_distributions(ax, note_models):
-        for note_symbol, note in iteritems(note_models):
+        for note_symbol, note in note_models.items():
             try:
                 ax.plot(note_models[note_symbol]['distribution'].vals,
                         note_models[note_symbol]['distribution'].bins,

tomato/symbolic/musicxml2lilypond/musicxmlreader.py

@@ -1,8 +1,7 @@
 # coding=utf-8
 from __future__ import unicode_literals, division
 import warnings
-import xml.etree.ElementTree as eT
-from xml.etree.ElementTree import ParseError
+from lxml import etree
 
 __author__ = 'hsercanatli', 'burakuyar', 'andresferrero', 'sertansenturk'
 
@@ -25,15 +24,15 @@ def read(cls, xml_in):
         """
 
         # setting the xml tree
-        parser = _XMLCommentHandler()
+        parser = etree.XMLParser(remove_comments=False)
         try:
             try:  # document
-                tree = eT.parse(xml_in, parser)
+                tree = etree.parse(xml_in, parser)
                 root = tree.getroot()
             except IOError:  # string input
-                root = eT.fromstring(xml_in, parser)
-        except ParseError:
-            raise ParseError("Line 36(ish): Error parsing MusicXML file.")
+                root = etree.fromstring(xml_in, parser)
+        except etree.ParseError:
+            raise etree.ParseError("Line 36(?): Error parsing MusicXML file.")
 
         # getting key signatures
         keysig = cls._get_key_signature(root)
@@ -55,6 +54,7 @@ def read(cls, xml_in):
 
     @staticmethod
     def _get_key_signature(root):
+
         keysig = {}  # keys: notes, values: type of note accident
         for xx, key in enumerate(
                 root.findall('part/measure/attributes/key/key-step')):
@@ -185,10 +185,12 @@ def _get_pitchstep_octave(note, rest):
 
     @staticmethod
     def _get_symbtr_txt_id(note):
-        if note.find("symbtrid").text:
-            return int(note.find("symbtrid").text)
-        else:
-            return None
+        for el in note:
+            if isinstance(el, etree._Comment):
+                comment_txt = el.text
+                if 'symbtr_txt_note_index' in comment_txt:
+                    return int(comment_txt[len("symbtr_txt_note_index "):])
+        return None
 
     @staticmethod
     def _chk_rest(note):
@@ -225,17 +227,16 @@ def _get_lyrics(note):
         else:
             return ''
 
-
-class _XMLCommentHandler(eT.XMLTreeBuilder):
-    def __init__(self):
-        super(_XMLCommentHandler, self).__init__()
-
-        # assumes ElementTree 1.2.X
-        self._parser.CommentHandler = self.handle_comment
-
-    def handle_comment(self, data):
-        self._target.start("symbtrid", {})
-        if data and 'symbtr_txt_note_index' in data:
-            data = data.replace('symbtr_txt_note_index ', '')
-        self._target.data(data)
-        self._target.end("symbtrid")
+# class _XMLCommentHandler(XMLTreeBuilder):
+#     def __init__(self):
+#         super(_XMLCommentHandler, self).__init__()
+#
+#         # assumes ElementTree 1.2.X
+#         self._parser.CommentHandler = self.handle_comment
+#
+#     def handle_comment(self, data):
+#         self._target.start("symbtrid", {})
+#         if data and 'symbtr_txt_note_index' in data:
+#             data = data.replace('symbtr_txt_note_index ', '')
+#         self._target.data(data)
+#         self._target.end("symbtrid")

tomato/symbolic/musicxml2lilypond/scoreconverter.py

@@ -1,6 +1,6 @@
 # coding=utf-8
 from __future__ import unicode_literals, division
-from musicxmlreader import MusicXMLReader
+from .musicxmlreader import MusicXMLReader
 import os
 import sqlite3
 
@@ -256,6 +256,9 @@ def convert(cls, xml_in, ly_out=None, render_metadata=True):
         # save to file
         if ly_out is not None:
             with open(ly_out, 'w') as outfile:
-                outfile.write(ly_stream.encode('utf-8'))
+                try:  # python 2
+                    outfile.write(ly_stream.encode('utf-8'))
+                except TypeError:  # python 3
+                    outfile.write(ly_stream)
 
         return ly_stream, mapping