Merge pull request #117 from sertansenturk/linting

Linting

.travis.yml

@@ -1,7 +1,7 @@
 language: python
 matrix:
   include:
-    - name: "Docker build"
+    - name: "docker build"
       services: docker
       python: 3.6
       env: TOX_ENV=docker
@@ -12,6 +12,24 @@ matrix:
       install: # installed by tox
       script: tox -e $TOX_ENV
       after_success: # do nothing
+    - name: "pylint"
+      python: 3.6
+      env: TOX_ENV=pylint
+      before_install:
+        - pip install -U pip
+        - pip install tox-travis
+      install: # installed by tox
+      script: tox -e $TOX_ENV
+      after_success: # do nothing
+    - name: "flake8"
+      python: 3.6
+      env: TOX_ENV=flake8
+      before_install:
+        - pip install -U pip
+        - pip install tox-travis
+      install: # installed by tox
+      script: tox -e $TOX_ENV
+      after_success: # do nothing
     - name: "Python 3.5"
       python: 3.5
       env: TOX_PYTHON_ENV=py35
@@ -30,6 +48,6 @@ install:
   - python setup.py install
   - pip show tomato
 script:
-  - tox -e $TOX_PYTHON_ENV, flake8, pylint
+  - tox -e $TOX_PYTHON_ENV
 after_success:
   - codecov # submit coverage to https://codecov.io/gh/sertansenturk/tomato/

setup.py

@@ -133,14 +133,14 @@ def _download_binary(fpath, bin_url, sys_os):
           "computational-analysis"),
       packages=find_packages(exclude=['docs', 'tests']),
       include_package_data=True,
-      python_requires='>=3.5, <3.8',
+      python_requires='>=3.5,<3.8',
       install_requires=[
           "numpy>=1.9.0",  # numerical operations
           "scipy>=0.17.0",  # temporary mat file saving for MCR binary inputs
           "pandas>=0.18.0,<=0.24.2",  # tabular data processing
           "matplotlib>=1.5.1,<=3.0.3",  # plotting
           "json_tricks>=3.12.1",  # saving json files with classes and numpy
-          "eyeD3>=0.7.5,<=0.8.11",  # reading metadata embedded in the audio recordings
+          "eyeD3>=0.7.5,<=0.8.11",  # reading metadata embedded in recordings
           "python-Levenshtein>=0.12.0",  # semiotic structure labeling
           "networkx>=1.11",  # semiotic structure labeling clique computation
           "lxml>=3.6.0",  # musicxml conversion

tomato/analyzer.py

@@ -27,7 +27,8 @@
 
 from .io import IO
 
-logger = logging.Logger(__name__, level=logging.INFO)
+logger = logging.Logger(  # pylint: disable-msg=C0103
+    __name__, level=logging.INFO)
 
 
 class Analyzer:

tomato/audio/ahenk.py

@@ -32,8 +32,9 @@ class Ahenk:
 
     @classmethod
     def identify(cls, tonic_freq, symbol_in):
-        assert 20.0 <= tonic_freq <= 20000.0, "The input tonic frequency " \
-                                              "must be between and 20000 Hz"
+        if (tonic_freq < 20) or (tonic_freq > 20000):
+            raise ValueError("The input tonic frequency must be between "
+                             "and 20000 Hz.")
 
         tonic_dict = IO.load_music_data('tonic')
         ahenks = IO.load_music_data('ahenk')
@@ -74,6 +75,8 @@ def identify(cls, tonic_freq, symbol_in):
                 ahenk_dict['slug'] = ahenk_slug
                 return ahenk_dict
 
+        raise ValueError("Unknown error: could not identify the ahenk.")
+
     @classmethod
     def _get_tonic_symbol(cls, symbol_in, tonic_dict):
         if symbol_in in tonic_dict.keys():  # tonic symbol given

tomato/audio/audioanalyzer.py

@@ -43,7 +43,8 @@
 from .predominantmelody import PredominantMelody
 from .seyir import Seyir
 
-logger = logging.Logger(__name__, level=logging.INFO)
+logger = logging.Logger(  # pylint: disable-msg=C0103
+    __name__, level=logging.INFO)
 
 
 class AudioAnalyzer(Analyzer):
@@ -348,7 +349,6 @@ def set_melody_progression_params(self, **kwargs):
     def set_note_modeler_params(self, **kwargs):
         self._set_params('_note_modeler', **kwargs)
 
-    # plot
     @staticmethod
     def plot(audio_features):
         pitch = audio_features['pitch_filtered']['pitch']

tomato/audio/makamtonic/inputparser.py

@@ -30,7 +30,8 @@
 from ..pitchdistribution import PitchDistribution
 from ...converter import Converter
 
-logger = logging.Logger(__name__, level=logging.INFO)
+logger = logging.Logger(  # pylint: disable-msg=C0103
+    __name__, level=logging.INFO)
 
 
 class InputParser:

tomato/audio/makamtonic/knn.py

@@ -165,4 +165,4 @@ def classify(cand_pairs, sorted_pair):
                 sorted_pair = [pp for pp in sorted_pair if pp[0] != p[0]]
                 return estimated_pair, sorted_pair
 
-        assert False, 'No pair selected, this should be impossible!'
+        raise ValueError('No pair selected, this should be impossible!')

tomato/audio/makamtonic/knnclassifier.py

@@ -58,11 +58,12 @@ def train(self, pitches, tonics, modes, sources=None, model_type='multi'):
         if model_type == 'single':
             return self._train_single_distrib_per_mode(
                 pitches, tonics, modes, sources=sources)
-        elif model_type == 'multi':
+
+        if model_type == 'multi':
             return self._train_multi_distrib_per_mode(
                 pitches, tonics, modes, sources=sources)
-        else:
-            raise ValueError("Unknown training model")
+
+        raise ValueError("Unknown training model")
 
     def _train_single_distrib_per_mode(self, pitches, tonics, modes,
                                        sources=None):
@@ -346,8 +347,8 @@ def model_from_pickle(self, input_str):
     def model_to_pickle(model, file_name=None):
         if file_name is None:
             return pickle.dumps(model)
-        else:
-            pickle.dump(model, open(file_name, 'wb'))
+
+        return pickle.dump(model, open(file_name, 'wb'))
 
     def model_from_json(self, file_name):
         """--------------------------------------------------------------------
@@ -387,5 +388,5 @@ def model_to_json(model, file_name=None):
 
         if file_name is None:
             return json.dumps(temp_model, indent=4)
-        else:
-            json.dump(temp_model, open(file_name, 'w'), indent=4)
+
+        return json.dump(temp_model, open(file_name, 'w'), indent=4)

tomato/audio/makamtonic/toniclastnote.py

@@ -24,11 +24,13 @@
 
 from copy import deepcopy
 
+import matplotlib.pyplot as plt
+import matplotlib.ticker
 import numpy as np
 
+from ...converter import Converter
 from ..pitchdistribution import PitchDistribution
 from ..pitchfilter import PitchFilter
-from ...converter import Converter
 
 
 class TonicLastNote:
@@ -156,9 +158,6 @@ def check_tonic_w_octave_cor(self, tonic, distribution):
 
     @staticmethod
     def plot(pitch, tonic, pitch_chunks, distribution):
-        import matplotlib.pyplot as plt
-        import matplotlib.ticker
-
         _, (ax1, ax2, ax3) = plt.subplots(3, num=None, figsize=(18, 8), dpi=80)
         plt.subplots_adjust(left=None, bottom=None, right=None, top=None,
                             wspace=0, hspace=0.4)

tomato/audio/notemodel.py

@@ -23,6 +23,8 @@
 
 from copy import deepcopy
 
+import matplotlib.pyplot as plt
+import matplotlib.ticker
 import numpy as np
 
 from ..converter import Converter
@@ -53,7 +55,7 @@ def calculate_notes(self, pitch_distribution, tonic_hz, makam,
             pass
 
         # Calculate stable pitches
-        peak_idx, peak_heights = pd_copy.detect_peaks(
+        peak_idx, _ = pd_copy.detect_peaks(
             min_peak_ratio=min_peak_ratio)
 
         stable_pitches_hz = pd_copy.bins[peak_idx]
@@ -187,10 +189,7 @@ def _get_extended_key_signature(self, key_sig_in, note_dict):
 
     @staticmethod
     def plot(pitch_distribution, stable_notes):
-        import matplotlib.pyplot as plt
-        import matplotlib.ticker
-
-        fig, ax = plt.subplots()
+        _, ax = plt.subplots()
         plt.subplots_adjust(left=None, bottom=None, right=None, top=None,
                             wspace=0, hspace=0.4)
 

tomato/audio/pitchdistribution.py

@@ -29,14 +29,16 @@
 
 import essentia
 import essentia.standard as std
+import matplotlib.pyplot as plt
 import numpy as np
 import scipy.integrate
 import scipy.stats
 
 from ..converter import Converter
 from ..io import IO
 
-logger = logging.Logger(__name__, level=logging.INFO)
+logger = logging.Logger(  # pylint: disable-msg=C0103
+    __name__, level=logging.INFO)
 
 
 class PitchDistribution:
@@ -84,11 +86,12 @@ def bin_unit(self):
                   '(bin unit is Hz) or a number greater than 0.'
         if self.ref_freq is None:
             return 'Hz'
-        elif isinstance(self.ref_freq, (numbers.Number, np.ndarray)):
+
+        if isinstance(self.ref_freq, (numbers.Number, np.ndarray)):
             assert self.ref_freq > 0, err_str
             return 'cent'
-        else:
-            return ValueError(err_str)
+
+        return ValueError(err_str)
 
     @staticmethod
     def from_cent_pitch(cent_track, ref_freq=440.0, kernel_width=7.5,
@@ -227,13 +230,11 @@ def is_pcd(self):
         if self.has_cent_bin():  # cent bins; compare directly
             return np.isclose(max(self.bins) - min(self.bins),
                               1200 - self.step_size)
-        else:
-            dummy_d = copy.deepcopy(self)
-
-            dummy_d.hz_to_cent(dummy_d.bins[0])
 
-            return np.isclose(max(dummy_d.bins) - min(dummy_d.bins),
-                              1200 - dummy_d.step_size)
+        dummy_d = copy.deepcopy(self)
+        dummy_d.hz_to_cent(dummy_d.bins[0])
+        return np.isclose(max(dummy_d.bins) - min(dummy_d.bins),
+                          1200 - dummy_d.step_size)
 
     def is_pdf(self):
         return np.isclose(np.sum(self.vals), 1)
@@ -273,7 +274,7 @@ def detect_peaks(self, min_peak_ratio=0.15):
             '1 (keep only the highest peak)'
 
         # Peak detection is handled by Essentia
-        detector = std.PeakDetection()
+        detector = std.PeakDetection()  # pylint: disable-msg=E1101
         peak_bins, peak_vals = detector(essentia.array(self.vals))
 
         # Essentia normalizes the positions to 1, they are converted here
@@ -413,23 +414,17 @@ def merge(self, distrib):
             self.normalize()
 
     def plot(self):
-        import matplotlib.pyplot as plt
-
         plt.plot(self.bins, self.vals)
         self._label_figure()
 
-    def bar(self):
-        import matplotlib.pyplot as plt
-
+    def bar(self):  # pylint: disable-msg=C0102
         bars = plt.bar(self.bins, self.vals, width=self.step_size,
                        align='center')
         self._label_figure()
 
         return bars
 
     def _label_figure(self):
-        import matplotlib.pyplot as plt
-
         if self.is_pcd():
             plt.title('Pitch class distribution')
             ref_freq_str = 'Hz x 2^n'
@@ -477,8 +472,8 @@ def to_json(self, file_name=None):
 
         if file_name is None:
             return json.dumps(dist_json, indent=4)
-        else:
-            json.dump(dist_json, open(file_name, 'w'), indent=4)
+
+        return json.dump(dist_json, open(file_name, 'w'), indent=4)
 
     @staticmethod
     def from_dict(distrib_dict):
@@ -488,10 +483,10 @@ def from_dict(distrib_dict):
 
     def to_dict(self):
         pdict = self.__dict__
-        for key in pdict.keys():
+        for key, val in pdict.items():
             try:
                 # convert to list from np array
-                pdict[key] = pdict[key].tolist()
+                pdict[key] = val.tolist()
             except AttributeError:
                 pass