def ffprobe_get_media_duration(file):
print('Getting {} duration'.format(file))
params = [FFPROBE]
params.extend(['-i', file])
params.extend(['-show_entries', 'format=duration'])
params.extend(['-v', 'quiet'])
params.extend(['-of', 'csv=p=0'])
return subprocess.check_output(
params, stdin=STDIN, stderr=subprocess.STDOUT
).decode().strip()
def ffmpeg_download_stream(
files, title, ext, params={}, output_dir='.', stream=True, **kwargs
):
"""str, str->True
WARNING: NOT THE SAME PARMS AS OTHER FUNCTIONS!!!!!!
You can basicly download anything with this function
but better leave it alone with
"""
# https://ffmpeg.org/ffmpeg.html#Main-options
output = '{}.{}'.format(title, ext)
if not (output_dir == '.'):
output = '{}/{}'.format(output_dir, output)
print(
'Downloading streaming content with FFmpeg, press q to stop '
'recording...'
)
if stream:
ffmpeg_params = [FFMPEG] + ['-y', '-re', '-i']
else:
ffmpeg_params = [FFMPEG] + ['-y', '-i']
ffmpeg_params.append(files) # not the same here!!!!
if FFMPEG == 'avconv': # who cares?
ffmpeg_params += ['-c', 'copy', output]
elif kwargs.get('override'):
pass
else:
ffmpeg_params += ['-c', 'copy', '-bsf:a', 'aac_adtstoasc']
if params is not None:
if len(params) > 0:
for k, v in params.items():
ffmpeg_params.append(k)
ffmpeg_params.append(v)
ffmpeg_params.append(output)
print(' '.join(ffmpeg_params))
try:
a = subprocess.Popen(ffmpeg_params, stdin=subprocess.PIPE)
a.communicate()
except KeyboardInterrupt:
try:
a.stdin.write('q'.encode('utf-8'))
except Exception:
pass
return True
def ffmpeg_concat_av(files, output, ext):
print('Merging video parts... ', end="", flush=True)
params = [FFMPEG] + LOGLEVEL
for file in files:
if os.path.isfile(file):
params.extend(['-i', file])
params.extend(['-c:v', 'copy'])
if ext == 'mp4':
params.extend(['-c:a', 'aac'])
elif ext == 'webm':
params.extend(['-c:a', 'vorbis'])
params.extend(['-strict', 'experimental'])
params.append(output)
return subprocess.call(params, stdin=STDIN)
def ffmpeg_concat_audio_and_video(files, output, ext):
print('Merging video and audio parts... ', end='', flush=True)
if has_ffmpeg_installed:
params = [FFMPEG] + LOGLEVEL
params.extend(['-f', 'concat'])
for file in files:
if os.path.isfile(file):
params.extend(['-i', file])
params.extend(['-c:v', 'copy'])
params.extend(['-c:a', 'aac'])
params.extend(['-strict', 'experimental'])
params.append('{}.{}'.format(output, ext))
return subprocess.call(params, stdin=STDIN)
else:
raise EnvironmentError('No ffmpeg found')
def ffmpeg_concat_ts_to_mkv(files, output='output.mkv'):
print('Merging video parts... ', end='', flush=True)
params = [FFMPEG] + LOGLEVEL + ['-isync', '-y', '-i']
params.append('concat:')
for file in files:
if os.path.isfile(file):
params[-1] += file + '|'
params += ['-f', 'matroska', '-c', 'copy', output]
try:
if subprocess.call(params, stdin=STDIN) == 0:
return True
else:
return False
except Exception:
return False
def ffmpeg_concat_flv_to_mp4(files, output='output.mp4'):
print('Merging video parts... ', end='', flush=True)
# Use concat demuxer on FFmpeg >= 1.1
if FFMPEG == 'ffmpeg' and (
FFMPEG_VERSION[0] >= 2 or (
FFMPEG_VERSION[0] == 1 and FFMPEG_VERSION[1] >= 1
)
):
concat_list = generate_concat_list(files, output)
params = [FFMPEG] + LOGLEVEL + [
'-y', '-f', 'concat', '-safe', '-1', '-i', concat_list, '-c',
'copy', '-bsf:a', 'aac_adtstoasc', output
]
subprocess.check_call(params, stdin=STDIN)
os.remove(output + '.txt')
return True
for file in files:
if os.path.isfile(file):
params = [FFMPEG] + LOGLEVEL + ['-y', '-i']
params.append(file)
params += [
'-map', '0', '-c', 'copy', '-f', 'mpegts', '-bsf:v',
'h264_mp4toannexb'
]
params.append(file + '.ts')
subprocess.call(params, stdin=STDIN)
params = [FFMPEG] + LOGLEVEL + ['-y', '-i']
params.append('concat:')
for file in files:
f = file + '.ts'
if os.path.isfile(f):
params[-1] += f + '|'
if FFMPEG == 'avconv':
params += ['-c', 'copy', output]
else:
params += ['-c', 'copy', '-absf', 'aac_adtstoasc', output]
if subprocess.call(params, stdin=STDIN) == 0:
for file in files:
os.remove(file + '.ts')
return True
else:
raise
def get_usable_ffmpeg(cmd):
try:
p = subprocess.Popen(
[cmd, '-version'], stdin=DEVNULL, stdout=subprocess.PIPE,
stderr=subprocess.PIPE
)
out, err = p.communicate()
vers = str(out, 'utf-8').split('\n')[0].split()
assert (vers[0] == 'ffmpeg' and vers[2][0] > '0') or \
(vers[0] == 'avconv')
# set version to 1.0 for nightly build and print warning
try:
version = [int(i) for i in vers[2].split('.')]
except Exception:
print('It seems that your ffmpeg is a nightly build.')
print('Please switch to the latest stable if merging failed.')
version = [1, 0]
return cmd, 'ffprobe', version
except Exception:
return None
def netease_lyric_download(self,
song,
output_dir,
info_only,
playlist_prefix='',
**kwargs):
if info_only or not kwargs.get('caption'):
return
data = json.loads(
get_content('http://music.163.com/api/song/lyric/?'
'id={}&lv=-1&csrf_token='.format(song['id']),
headers={'Referer': 'http://music.163.com/'}))
title = '{}{}. {}'.format(playlist_prefix, song['position'],
song['name'])
filename = '{}.lrc'.format(get_filename(title))
print('Saving {} ...'.format(filename), end='', flush=True)
with open(os.path.join(output_dir, filename), 'w',
encoding='utf-8') as x:
x.write(data['lrc']['lyric'])
print('Done.')
def p_i(self, stream_id):
if stream_id in self.streams:
stream = self.streams[stream_id]
else:
stream = self.dash_streams[stream_id]
maybe_print(" - title: %s" % self.title)
print(" size: {} MiB ({} bytes)".format(
round(stream['size'] / 1048576, 1), stream['size']))
print(" url: %s" % self.url)
print()
def download(self, **kwargs):
if 'json_output' in kwargs and kwargs['json_output']:
json_output.output(self)
elif 'info_only' in kwargs and kwargs['info_only']:
if 'stream_id' in kwargs and kwargs['stream_id']:
# Display the stream
stream_id = kwargs['stream_id']
if 'index' not in kwargs:
self.p(stream_id)
else:
self.p_i(stream_id)
else:
# Display all available streams
if 'index' not in kwargs:
self.p([])
else:
stream_id = self.streams_sorted[0]['id'] \
if 'id' in self.streams_sorted[0] \
else self.streams_sorted[0]['itag']
self.p_i(stream_id)
else:
if 'stream_id' in kwargs and kwargs['stream_id']:
# Download the stream
stream_id = kwargs['stream_id']
else:
# Download stream with the best quality
stream_id = self.streams_sorted[0]['id'] \
if 'id' in self.streams_sorted[0] \
else self.streams_sorted[0]['itag']
if 'index' not in kwargs:
self.p(stream_id)
else:
self.p_i(stream_id)
if stream_id in self.streams:
urls = self.streams[stream_id]['src']
ext = self.streams[stream_id]['container']
total_size = self.streams[stream_id]['size']
else:
urls = self.dash_streams[stream_id]['src']
ext = self.dash_streams[stream_id]['container']
total_size = self.dash_streams[stream_id]['size']
if not urls:
log.wtf('[Failed] Cannot extract video source.')
if ext == 'm3u8':
ffmpeg_kwargs = {}
if 'iqiyi' in self.name:
# ffmpeg_kwargs['override'] = True
# ffmpeg_kwargs['params'] = {
# '-c:a': 'copy', '-bsf:a': 'aac_adtstoasc'
# }
m3u8_urls = general_m3u8_extractor(urls[0])
# FIXME(iawia002): 如果要计算大小的话需要消耗太多时间
if len(m3u8_urls) <= 100:
size = urls_size(m3u8_urls)
else:
size = float('inf')
download_urls(m3u8_urls, self.title, 'mp4', size, **kwargs)
else:
download_url_ffmpeg(urls[0],
self.title,
'mp4',
output_dir=kwargs['output_dir'],
merge=kwargs['merge'],
stream=False,
**ffmpeg_kwargs)
else:
headers = copy(config.FAKE_HEADERS)
if self.ua is not None:
headers['User-Agent'] = self.ua
if self.referer is not None:
headers['Referer'] = self.referer
download_urls(urls,
self.title,
ext,
total_size,
headers=headers,
output_dir=kwargs['output_dir'],
merge=kwargs['merge'],
av=stream_id in self.dash_streams)
if 'caption' not in kwargs or not kwargs['caption']:
print('Skipping captions or danmuku.')
return
for lang in self.caption_tracks:
filename = '%s.%s.srt' % (get_filename(self.title), lang)
print('Saving %s ... ' % filename, end="", flush=True)
srt = self.caption_tracks[lang]
with open(os.path.join(kwargs['output_dir'], filename),
'w',
encoding='utf-8') as x:
x.write(srt)
print('Done.')
if self.danmuku is not None and not dry_run:
filename = '{}.cmt.xml'.format(get_filename(self.title))
print('Downloading {} ...\n'.format(filename))
with open(os.path.join(kwargs['output_dir'], filename),
'w',
encoding='utf8') as fp:
fp.write(self.danmuku)
keep_obj = kwargs.get('keep_obj', False)
if not keep_obj:
self.__init__()
def p_playlist(self, stream_id=None):
maybe_print("site: %s" % self.__class__.name)
print("playlist: %s" % self.title)
print("videos:")
def p(self, stream_id=None):
maybe_print("site: %s" % self.__class__.name)
maybe_print("title: %s" % self.title)
if stream_id:
# Print the stream
print("stream:")
self.p_stream(stream_id)
elif stream_id is None:
# Print stream with best quality
print("stream: # Best quality")
stream_id = self.streams_sorted[0]['id'] \
if 'id' in self.streams_sorted[0] \
else self.streams_sorted[0]['itag']
self.p_stream(stream_id)
elif stream_id == []:
print("streams: # Available quality and codecs")
# Print DASH streams
if self.dash_streams:
print(" [ DASH ] %s" % ('_' * 36))
itags = sorted(self.dash_streams,
key=lambda i: -self.dash_streams[i]['size'])
for stream in itags:
self.p_stream(stream)
# Print all other available streams
print(" [ DEFAULT ] %s" % ('_' * 33))
for stream in self.streams_sorted:
self.p_stream(stream['id'] if 'id' in
stream else stream['itag'])
if self.audiolang:
print("audio-languages:")
for i in self.audiolang:
print(" - lang: {}".format(i['lang']))
print(" download-url: {}\n".format(i['url']))
def p_stream(self, stream_id):
if stream_id in self.streams:
stream = self.streams[stream_id]
else:
stream = self.dash_streams[stream_id]
if 'itag' in stream:
print(" - itag: %s" %
log.sprint(stream_id, log.NEGATIVE))
else:
print(" - format: %s" %
log.sprint(stream_id, log.NEGATIVE))
if 'container' in stream:
print(" container: %s" % stream['container'])
if 'video_profile' in stream:
maybe_print(" video-profile: %s" % stream['video_profile'])
if 'quality' in stream:
print(" quality: %s" % stream['quality'])
if 'size' in stream and stream['container'].lower() != 'm3u8':
if stream['size'] != float('inf') and stream['size'] != 0:
print(" size: {} MiB ({} bytes)".format(
round(stream['size'] / 1048576, 1), stream['size']))
if 'm3u8_url' in stream:
print(" m3u8_url: {}".format(stream['m3u8_url']))
if 'itag' in stream:
print(
" # download-with: %s" %
log.sprint("lulu --itag=%s [URL]" % stream_id, log.UNDERLINE))
else:
print(" # download-with: %s" % log.sprint(
"lulu --format=%s [URL]" % stream_id, log.UNDERLINE))
print()
def download(self, **kwargs):
if 'json_output' in kwargs and kwargs['json_output']:
json_output.output(self)
elif 'info_only' in kwargs and kwargs['info_only']:
if 'stream_id' in kwargs and kwargs['stream_id']:
# Display the stream
stream_id = kwargs['stream_id']
if 'index' not in kwargs:
self.p(stream_id)
else:
self.p_i(stream_id)
else:
# Display all available streams
if 'index' not in kwargs:
self.p([])
else:
stream_id = self.streams_sorted[0]['id'] \
if 'id' in self.streams_sorted[0] \
else self.streams_sorted[0]['itag']
self.p_i(stream_id)
else:
if 'stream_id' in kwargs and kwargs['stream_id']:
# Download the stream
stream_id = kwargs['stream_id']
else:
# Download stream with the best quality
stream_id = self.streams_sorted[0]['id'] \
if 'id' in self.streams_sorted[0] \
else self.streams_sorted[0]['itag']
if 'index' not in kwargs:
self.p(stream_id)
else:
self.p_i(stream_id)
if stream_id in self.streams:
urls = self.streams[stream_id]['src']
ext = self.streams[stream_id]['container']
total_size = self.streams[stream_id]['size']
else:
urls = self.dash_streams[stream_id]['src']
ext = self.dash_streams[stream_id]['container']
total_size = self.dash_streams[stream_id]['size']
if ext == 'm3u8':
ext = 'mp4'
if not urls:
log.wtf('[Failed] Cannot extract video source.')
# For legacy main()
headers = copy(config.FAKE_HEADERS)
if self.ua is not None:
headers['User-Agent'] = self.ua
if self.referer is not None:
headers['Referer'] = self.referer
download_urls(
urls,
self.title,
ext,
total_size,
headers=headers,
output_dir=kwargs['output_dir'],
merge=kwargs['merge'],
av=stream_id in self.dash_streams
)
if 'caption' not in kwargs or not kwargs['caption']:
print('Skipping captions or danmuku.')
return
for lang in self.caption_tracks:
filename = '%s.%s.srt' % (get_filename(self.title), lang)
print('Saving %s ... ' % filename, end="", flush=True)
srt = self.caption_tracks[lang]
with open(
os.path.join(kwargs['output_dir'], filename),
'w',
encoding='utf-8'
) as x:
x.write(srt)
print('Done.')
if self.danmuku is not None and not dry_run:
filename = '{}.cmt.xml'.format(get_filename(self.title))
print('Downloading {} ...\n'.format(filename))
with open(
os.path.join(kwargs['output_dir'], filename),
'w',
encoding='utf8'
) as fp:
fp.write(self.danmuku)
# For main_dev()
# download_urls(
# urls, self.title, self.streams[stream_id]['container'],
# self.streams[stream_id]['size']
# )
keep_obj = kwargs.get('keep_obj', False)
if not keep_obj:
self.__init__()
def download(self, **kwargs):
"""Override the original one
Ugly ugly dirty hack
"""
if 'json_output' in kwargs and kwargs['json_output']:
json_output.output(self)
elif 'info_only' in kwargs and kwargs['info_only']:
if 'stream_id' in kwargs and kwargs['stream_id']:
# Display the stream
stream_id = kwargs['stream_id']
if 'index' not in kwargs:
self.p(stream_id)
else:
self.p_i(stream_id)
else:
# Display all available streams
if 'index' not in kwargs:
self.p([])
else:
stream_id = self.streams_sorted[0]['id'] \
if 'id' in self.streams_sorted[0] \
else self.streams_sorted[0]['itag']
self.p_i(stream_id)
else:
if 'stream_id' in kwargs and kwargs['stream_id']:
# Download the stream
stream_id = kwargs['stream_id']
else:
# Download stream with the best quality
stream_id = self.streams_sorted[0]['id'] \
if 'id' in self.streams_sorted[0] \
else self.streams_sorted[0]['itag']
if 'index' not in kwargs:
self.p(stream_id)
else:
self.p_i(stream_id)
if stream_id in self.streams:
urls = self.streams[stream_id]['src']
# ext = self.streams[stream_id]['container']
# total_size = self.streams[stream_id]['size']
else:
urls = self.dash_streams[stream_id]['src']
# ext = self.dash_streams[stream_id]['container']
# total_size = self.dash_streams[stream_id]['size']
if not urls:
log.wtf('[Failed] Cannot extract video source.')
# Here's the change!
download_url_ffmpeg(urls[0],
self.title,
'mp4',
output_dir=kwargs['output_dir'],
merge=kwargs['merge'],
stream=False)
if not kwargs['caption']:
print('Skipping captions.')
return
for lang in self.caption_tracks:
filename = '{}.{}.srt'.format(get_filename(self.title), lang)
print('Saving {} ... '.format(filename), end='', flush=True)
srt = self.caption_tracks[lang]
with open(os.path.join(kwargs['output_dir'], filename),
'w',
encoding='utf-8') as x:
x.write(srt)
print('Done.')
