Merge branch 'master' of https://github.com/jiaaro/pydub

Author: jiaaro

pydub/audio_segment.py

@@ -71,6 +71,7 @@ def setter(self, func):
         self.fset = func
         return self
 
+
 def classproperty(func):
     if not isinstance(func, (classmethod, staticmethod)):
         func = classmethod(func)
@@ -83,7 +84,6 @@ def classproperty(func):
     "wave": "wav",
 }
 
-
 WavSubChunk = namedtuple('WavSubChunk', ['id', 'position', 'size'])
 WavData = namedtuple('WavData', ['audio_format', 'channels', 'sample_rate',
                                  'bits_per_sample', 'raw_data'])
@@ -209,10 +209,10 @@ def __init__(self, data=None, *args, **kwargs):
                 data = data if isinstance(data, (basestring, bytes)) else data.read()
             except(OSError):
                 d = b''
-                reader = data.read(2**31-1)
+                reader = data.read(2 ** 31 - 1)
                 while reader:
                     d += reader
-                    reader = data.read(2**31-1)
+                    reader = data.read(2 ** 31 - 1)
                 data = d
 
             wav_data = read_wav_audio(data)
@@ -257,7 +257,6 @@ def raw_data(self):
         """
         return self._data
 
-
     def get_array_of_samples(self):
         """
         returns the raw_data as an array of samples
@@ -290,7 +289,7 @@ def __getitem__(self, millisecond):
         if isinstance(millisecond, slice):
             if millisecond.step:
                 return (
-                    self[i:i+millisecond.step]
+                    self[i:i + millisecond.step]
                     for i in xrange(*millisecond.indices(len(self)))
                 )
 
@@ -468,7 +467,8 @@ def from_mono_audiosegments(cls, *mono_segments):
         segs = cls._sync(*mono_segments)
 
         if segs[0].channels != 1:
-            raise ValueError("AudioSegment.from_mono_audiosegments requires all arguments are mono AudioSegment instances")
+            raise ValueError(
+                "AudioSegment.from_mono_audiosegments requires all arguments are mono AudioSegment instances")
 
         channels = len(segs)
         sample_width = segs[0].sample_width
@@ -536,13 +536,13 @@ def is_format(f):
         except(OSError):
             input_file.flush()
             input_file.close()
-            input_file = NamedTemporaryFile(mode='wb', delete=False, buffering=2**31-1)
+            input_file = NamedTemporaryFile(mode='wb', delete=False, buffering=2 ** 31 - 1)
             file.close()
-            file = open(orig_file, buffering=2**13-1, mode='rb')
-            reader = file.read(2**31-1)
+            file = open(orig_file, buffering=2 ** 13 - 1, mode='rb')
+            reader = file.read(2 ** 31 - 1)
             while reader:
                 input_file.write(reader)
-                reader = file.read(2**31-1)
+                reader = file.read(2 ** 31 - 1)
         input_file.flush()
         file.close()
 
@@ -583,7 +583,9 @@ def is_format(f):
 
         try:
             if p.returncode != 0:
-                raise CouldntDecodeError("Decoding failed. ffmpeg returned error code: {0}\n\nOutput from ffmpeg/avlib:\n\n{1}".format(p.returncode, p_err))
+                raise CouldntDecodeError(
+                    "Decoding failed. ffmpeg returned error code: {0}\n\nOutput from ffmpeg/avlib:\n\n{1}".format(
+                        p.returncode, p_err))
             obj = cls._from_safe_wav(output)
         finally:
             input_file.close()
@@ -661,9 +663,9 @@ def is_format(f):
                              if x['codec_type'] == 'audio']
             # This is a workaround for some ffprobe versions that always say
             # that mp3/mp4/aac/webm/ogg files contain fltp samples
-            if (audio_streams[0]['sample_fmt'] == 'fltp' and
-                (is_format("mp3") or is_format("mp4") or is_format("aac") or
-                 is_format("webm") or is_format("ogg"))):
+            if (audio_streams[0].get('sample_fmt') == 'fltp' and
+                    (is_format("mp3") or is_format("mp4") or is_format("aac") or
+                     is_format("webm") or is_format("ogg"))):
                 bits_per_sample = 16
             else:
                 bits_per_sample = audio_streams[0]['bits_per_sample']
@@ -688,7 +690,9 @@ def is_format(f):
 
         if p.returncode != 0 or len(p_out) == 0:
             file.close()
-            raise CouldntDecodeError("Decoding failed. ffmpeg returned error code: {0}\n\nOutput from ffmpeg/avlib:\n\n{1}".format(p.returncode, p_err))
+            raise CouldntDecodeError(
+                "Decoding failed. ffmpeg returned error code: {0}\n\nOutput from ffmpeg/avlib:\n\n{1}".format(
+                    p.returncode, p_err))
 
         p_out = bytearray(p_out)
         fix_wav_headers(p_out)
@@ -716,17 +720,19 @@ def from_wav(cls, file, parameters=None):
 
     @classmethod
     def from_raw(cls, file, **kwargs):
-        return cls.from_file(file, 'raw', sample_width=kwargs['sample_width'], frame_rate=kwargs['frame_rate'], channels=kwargs['channels'])
+        return cls.from_file(file, 'raw', sample_width=kwargs['sample_width'], frame_rate=kwargs['frame_rate'],
+                             channels=kwargs['channels'])
 
     @classmethod
     def _from_safe_wav(cls, file):
         file = _fd_or_path_or_tempfile(file, 'rb', tempfile=False)
         file.seek(0)
-        obj = cls(data=file);
+        obj = cls(data=file)
         file.close()
         return obj
 
-    def export(self, out_f=None, format='mp3', codec=None, bitrate=None, parameters=None, tags=None, id3v2_version='4', cover=None):
+    def export(self, out_f=None, format='mp3', codec=None, bitrate=None, parameters=None, tags=None, id3v2_version='4',
+               cover=None):
         """
         Export an AudioSegment to a file with given options
 
@@ -804,9 +810,10 @@ def export(self, out_f=None, format='mp3', codec=None, bitrate=None, parameters=
 
         if cover is not None:
             if cover.lower().endswith(('.png', '.jpg', '.jpeg', '.bmp', '.tif', '.tiff')) and format == "mp3":
-                conversion_command.extend(["-i" , cover, "-map", "0", "-map", "1", "-c:v", "mjpeg"])
+                conversion_command.extend(["-i", cover, "-map", "0", "-map", "1", "-c:v", "mjpeg"])
             else:
-                raise AttributeError("Currently cover images are only supported by MP3 files. The allowed image formats are: .tif, .jpg, .bmp, .jpeg and .png.")
+                raise AttributeError(
+                    "Currently cover images are only supported by MP3 files. The allowed image formats are: .tif, .jpg, .bmp, .jpeg and .png.")
 
         if codec is not None:
             # force audio encoder
@@ -835,7 +842,7 @@ def export(self, out_f=None, format='mp3', codec=None, bitrate=None, parameters=
                         raise InvalidID3TagVersion(
                             "id3v2_version not allowed, allowed versions: %s" % id3v2_allowed_versions)
                     conversion_command.extend([
-                        "-id3v2_version",  id3v2_version
+                        "-id3v2_version", id3v2_version
                     ])
 
         if sys.platform == 'darwin':
@@ -856,7 +863,9 @@ def export(self, out_f=None, format='mp3', codec=None, bitrate=None, parameters=
         log_subprocess_output(p_err)
 
         if p.returncode != 0:
-            raise CouldntEncodeError("Encoding failed. ffmpeg/avlib returned error code: {0}\n\nCommand:{1}\n\nOutput from ffmpeg/avlib:\n\n{2}".format(p.returncode, conversion_command, p_err))
+            raise CouldntEncodeError(
+                "Encoding failed. ffmpeg/avlib returned error code: {0}\n\nCommand:{1}\n\nOutput from ffmpeg/avlib:\n\n{2}".format(
+                    p.returncode, conversion_command, p_err))
 
         output.seek(0)
         out_f.write(output.read())
@@ -1114,11 +1123,11 @@ def overlay(self, seg, position=0, loop=False, times=None, gain_during_overlay=N
             if gain_during_overlay:
                 seg1_overlaid = seg1[pos:pos + seg2_len]
                 seg1_adjusted_gain = audioop.mul(seg1_overlaid, self.sample_width,
-                                               db_to_float(float(gain_during_overlay)))
+                                                 db_to_float(float(gain_during_overlay)))
                 output.write(audioop.add(seg1_adjusted_gain, seg2, sample_width))
             else:
                 output.write(audioop.add(seg1[pos:pos + seg2_len], seg2,
-                                     sample_width))
+                                         sample_width))
             pos += seg2_len
 
             # dec times to break our while loop (eventually)
@@ -1219,7 +1228,7 @@ def fade(self, to_gain=0, from_gain=0, start=None, end=None,
 
         # fades longer than 100ms can use coarse fading (one gain step per ms),
         # shorter fades will have audible clicks so they use precise fading
-        #(one gain step per sample)
+        # (one gain step per sample)
         if duration > 100:
             scale_step = gain_delta / duration
 
@@ -1266,14 +1275,15 @@ def reverse(self):
         )
 
     def _repr_html_(self):
-            src = """
+        src = """
                     <audio controls>
                         <source src="data:audio/mpeg;base64,{base64}" type="audio/mpeg"/>
                         Your browser does not support the audio element.
                     </audio>
                   """
-            fh = self.export()
-            data = base64.b64encode(fh.read()).decode('ascii')
-            return src.format(base64=data)
+        fh = self.export()
+        data = base64.b64encode(fh.read()).decode('ascii')
+        return src.format(base64=data)
+
 
 from . import effects

pydub/utils.py

@@ -1,32 +1,29 @@
 from __future__ import division
 
-from math import log, ceil, floor
+import json
 import os
 import re
-from subprocess import Popen, PIPE
 import sys
+from subprocess import Popen, PIPE
+from math import log, ceil
 from tempfile import TemporaryFile
 from warnings import warn
-import json
 
 try:
     import audioop
 except ImportError:
     import pyaudioop as audioop
 
-
 if sys.version_info >= (3, 0):
     basestring = str
 
-
-
 FRAME_WIDTHS = {
     8: 1,
     16: 2,
     32: 4,
 }
 ARRAY_TYPES = {
-    8:  "b",
+    8: "b",
     16: "h",
     32: "i",
 }
@@ -78,7 +75,7 @@ def db_to_float(db, using_amplitude=True):
     db = float(db)
     if using_amplitude:
         return 10 ** (db / 20)
-    else: # using power
+    else:  # using power
         return 10 ** (db / 10)
 
 
@@ -92,33 +89,28 @@ def ratio_to_db(ratio, val2=None, using_amplitude=True):
     # accept 2 values and use the ratio of val1 to val2
     if val2 is not None:
         ratio = ratio / val2
-    
+
     # special case for multiply-by-zero (convert to silence)
     if ratio == 0:
         return -float('inf')
 
     if using_amplitude:
         return 20 * log(ratio, 10)
-    else: # using power
+    else:  # using power
         return 10 * log(ratio, 10)
-    
+
 
 def register_pydub_effect(fn, name=None):
     """
     decorator for adding pydub effects to the AudioSegment objects.
-
     example use:
-
         @register_pydub_effect
         def normalize(audio_segment):
             ...
-
     or you can specify a name:
-
         @register_pydub_effect("normalize")
         def normalize_audio_segment(audio_segment):
             ...
-
     """
     if isinstance(fn, basestring):
         name = fn
@@ -136,7 +128,6 @@ def make_chunks(audio_segment, chunk_length):
     """
     Breaks an AudioSegment into chunks that are <chunk_length> milliseconds
     long.
-
     if chunk_length is 50 then you'll get a list of 50 millisecond long audio
     segments back (except the last one, which can be shorter)
     """
@@ -149,7 +140,7 @@ def which(program):
     """
     Mimics behavior of UNIX which command.
     """
-    #Add .exe program extension for windows support
+    # Add .exe program extension for windows support
     if os.name == "nt" and not program.endswith(".exe"):
         program += ".exe"
 
@@ -174,6 +165,7 @@ def get_encoder_name():
         warn("Couldn't find ffmpeg or avconv - defaulting to ffmpeg, but may not work", RuntimeWarning)
         return "ffmpeg"
 
+
 def get_player_name():
     """
     Return enconder default application for system, either avconv or ffmpeg
@@ -220,6 +212,33 @@ def fsdecode(filename):
     raise TypeError("type {0} not accepted by fsdecode".format(type(filename)))
 
 
+def get_extra_info(stderr):
+    """
+    avprobe sometimes gives more information on stderr than
+    on the json output. The information has to be extracted
+    from stderr of the format of:
+    '    Stream #0:0: Audio: flac, 88200 Hz, stereo, s32 (24 bit)'
+    or (macOS version):
+    '    Stream #0:0: Audio: vorbis'
+    '      44100 Hz, stereo, fltp, 320 kb/s'
+
+    :type stderr: str
+    :rtype: list of dict
+    """
+    extra_info = {}
+
+    re_stream = r'(?P<space_start> +)Stream #0[:\.](?P<stream_id>([0-9]+))(?P<content_0>.+)\n?((?P<space_end> +)(?P<content_1>.+))?'
+    for i in re.finditer(re_stream, stderr):
+        if i.group('space_end') is not None and len(i.group('space_start')) <= len(
+                i.group('space_end')):
+            content_line = ','.join([i.group('content_0'), i.group('content_1')])
+        else:
+            content_line = i.group('content_0')
+        tokens = [x.strip() for x in re.split('[:,]', content_line) if x]
+        extra_info[int(i.group('stream_id'))] = tokens
+    return extra_info
+
+
 def mediainfo_json(filepath):
     """Return json dictionary with media info(codec, duration, size, bitrate...) from filepath
     """
@@ -253,18 +272,7 @@ def mediainfo_json(filepath):
         # (for example, because the file doesn't exist)
         return info
 
-    # avprobe sometimes gives more information on stderr than
-    # on the json output. The information has to be extracted
-    # from lines of the format of:
-    # '    Stream #0:0: Audio: flac, 88200 Hz, stereo, s32 (24 bit)'
-    extra_info = {}
-    for line in stderr.split("\n"):
-        match = re.match(' *Stream #0[:\.]([0-9]+)(\(\w+\))?', line)
-        if match:
-            stream_id = int(match.group(1))
-            tokens = [x.strip()
-                      for x in re.split('[:,]', line[match.end():]) if x]
-            extra_info[stream_id] = tokens
+    extra_info = get_extra_info(stderr)
 
     audio_streams = [x for x in info['streams'] if x['codec_type'] == 'audio']
     if len(audio_streams) == 0:
@@ -296,7 +304,6 @@ def set_property(stream, prop, value):
             set_property(stream, 'sample_fmt', token)
             set_property(stream, 'bits_per_sample', 64)
             set_property(stream, 'bits_per_raw_sample', 64)
-
     return info