Compare commits

...

5 Commits

Author SHA1 Message Date
Adam Malcontenti-Wilson
0145f4fd5f
Merge dc179db221 into 1036478d13 2025-01-10 11:28:13 +00:00
dirkf
1036478d13 [YouTube] Endure subtitle URLs are complete
* WEB URLs are, MWEB not
* resolves #33017
2025-01-06 01:39:04 +00:00
dirkf
00ad2b8ca1 [YouTube] Refactor subtitle processing
* move to internal function
* use `traverse-obj()`
2025-01-06 01:24:30 +00:00
dirkf
ab7c61ca29 [YouTube] Apply code style changes, trailing commas, etc 2025-01-06 01:22:16 +00:00
Adam Malcontenti-Wilson
dc179db221 Add concat-playlist option 2021-02-13 20:09:31 +11:00
7 changed files with 120 additions and 65 deletions

View File

@ -122,6 +122,7 @@ from .postprocessor import (
FFmpegFixupM4aPP, FFmpegFixupM4aPP,
FFmpegFixupStretchedPP, FFmpegFixupStretchedPP,
FFmpegMergerPP, FFmpegMergerPP,
FFmpegConcatPP,
FFmpegPostProcessor, FFmpegPostProcessor,
get_postprocessor, get_postprocessor,
) )
@ -1206,6 +1207,20 @@ class YoutubeDL(object):
entry_result = self.__process_iterable_entry(entry, download, extra) entry_result = self.__process_iterable_entry(entry, download, extra)
# TODO: skip failed (empty) entries? # TODO: skip failed (empty) entries?
playlist_results.append(entry_result) playlist_results.append(entry_result)
if self.params.get('concat_playlist', False):
concat_pp = FFmpegConcatPP(self)
ie_result['__postprocessors'] = [concat_pp]
ie_result['__files_to_concat'] = list(map(lambda e: e['_filename'], entries))
filename_wo_ext = self.prepare_filename(ie_result)
# Ensure filename always has a correct extension for successful merge
ie_result['ext'] = self.params.get('merge_output_format') or entries[0]['ext']
ie_result['_filename'] = filename = '%s.%s' % (filename_wo_ext, ie_result['ext'])
self.post_process(filename, ie_result)
ie_result['entries'] = playlist_results ie_result['entries'] = playlist_results
self.to_screen('[download] Finished downloading playlist: %s' % playlist) self.to_screen('[download] Finished downloading playlist: %s' % playlist)
return ie_result return ie_result
@ -1858,6 +1873,7 @@ class YoutubeDL(object):
new_info = dict(info_dict) new_info = dict(info_dict)
new_info.update(format) new_info.update(format)
self.process_info(new_info) self.process_info(new_info)
format.update(new_info)
# We update the info dict with the best quality format (backwards compatibility) # We update the info dict with the best quality format (backwards compatibility)
info_dict.update(formats_to_download[-1]) info_dict.update(formats_to_download[-1])
return info_dict return info_dict

View File

@ -411,6 +411,7 @@ def _real_main(argv=None):
'youtube_include_dash_manifest': opts.youtube_include_dash_manifest, 'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
'encoding': opts.encoding, 'encoding': opts.encoding,
'extract_flat': opts.extract_flat, 'extract_flat': opts.extract_flat,
'concat_playlist': opts.concat_playlist,
'mark_watched': opts.mark_watched, 'mark_watched': opts.mark_watched,
'merge_output_format': opts.merge_output_format, 'merge_output_format': opts.merge_output_format,
'postprocessors': postprocessors, 'postprocessors': postprocessors,

View File

@ -9,6 +9,7 @@ import json
import os.path import os.path
import random import random
import re import re
import string
import time import time
import traceback import traceback
@ -67,6 +68,7 @@ from ..utils import (
class YoutubeBaseInfoExtractor(InfoExtractor): class YoutubeBaseInfoExtractor(InfoExtractor):
"""Provide base functions for Youtube extractors""" """Provide base functions for Youtube extractors"""
_LOGIN_URL = 'https://accounts.google.com/ServiceLogin' _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
_TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge' _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
@ -138,7 +140,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
[2, 1, None, 1, [2, 1, None, 1,
'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn',
None, [], 4], None, [], 4],
1, [None, None, []], None, None, None, True 1, [None, None, []], None, None, None, True,
], ],
username, username,
] ]
@ -160,7 +162,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
None, 1, None, [1, None, None, None, [password, None, True]], None, 1, None, [1, None, None, None, [password, None, True]],
[ [
None, None, [2, 1, None, 1, 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', None, [], 4], None, None, [2, 1, None, 1, 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', None, [], 4],
1, [None, None, []], None, None, None, True 1, [None, None, []], None, None, None, True,
]] ]]
challenge_results = req( challenge_results = req(
@ -213,7 +215,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
user_hash, None, 2, None, user_hash, None, 2, None,
[ [
9, None, None, None, None, None, None, None, 9, None, None, None, None, None, None, None,
[None, tfa_code, True, 2] [None, tfa_code, True, 2],
]] ]]
tfa_results = req( tfa_results = req(
@ -284,7 +286,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
'client': { 'client': {
'clientName': 'WEB', 'clientName': 'WEB',
'clientVersion': '2.20201021.03.00', 'clientVersion': '2.20201021.03.00',
} },
}, },
} }
@ -385,7 +387,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
'client': { 'client': {
'clientName': 'WEB', 'clientName': 'WEB',
'clientVersion': '2.20201021.03.00', 'clientVersion': '2.20201021.03.00',
} },
}, },
'query': query, 'query': query,
} }
@ -462,7 +464,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
# (HTML, videodetails, metadata, renderers) # (HTML, videodetails, metadata, renderers)
'name': ('content', 'author', (('ownerChannelName', None), 'title'), ['text']), 'name': ('content', 'author', (('ownerChannelName', None), 'title'), ['text']),
'url': ('href', 'ownerProfileUrl', 'vanityChannelUrl', 'url': ('href', 'ownerProfileUrl', 'vanityChannelUrl',
['navigationEndpoint', 'browseEndpoint', 'canonicalBaseUrl']) ['navigationEndpoint', 'browseEndpoint', 'canonicalBaseUrl']),
} }
if any((videodetails, metadata, renderers)): if any((videodetails, metadata, renderers)):
result = ( result = (
@ -671,7 +673,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO', 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
'description': '', 'description': '',
'uploader': '8KVIDEO', 'uploader': '8KVIDEO',
'title': 'UHDTV TEST 8K VIDEO.mp4' 'title': 'UHDTV TEST 8K VIDEO.mp4',
}, },
'params': { 'params': {
'youtube_include_dash_manifest': True, 'youtube_include_dash_manifest': True,
@ -711,7 +713,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'uploader_url': r're:https?://(?:www\.)?youtube\.com/@theamazingatheist', 'uploader_url': r're:https?://(?:www\.)?youtube\.com/@theamazingatheist',
'title': 'Burning Everyone\'s Koran', 'title': 'Burning Everyone\'s Koran',
'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms \r\n\r\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html', 'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms \r\n\r\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html',
} },
}, },
# Age-gated videos # Age-gated videos
{ {
@ -839,7 +841,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
}, },
'expected_warnings': [ 'expected_warnings': [
'DASH manifest missing', 'DASH manifest missing',
] ],
}, },
# Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431) # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
{ {
@ -1820,8 +1822,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# cpn generation algorithm is reverse engineered from base.js. # cpn generation algorithm is reverse engineered from base.js.
# In fact it works even with dummy cpn. # In fact it works even with dummy cpn.
CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_' CPN_ALPHABET = string.ascii_letters + string.digits + '-_'
cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)) cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(16))
# more consistent results setting it to right before the end # more consistent results setting it to right before the end
qs = parse_qs(playback_url) qs = parse_qs(playback_url)
@ -1881,8 +1883,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
mobj = re.match(cls._VALID_URL, url, re.VERBOSE) mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
if mobj is None: if mobj is None:
raise ExtractorError('Invalid URL: %s' % url) raise ExtractorError('Invalid URL: %s' % url)
video_id = mobj.group(2) return mobj.group(2)
return video_id
def _extract_chapters_from_json(self, data, video_id, duration): def _extract_chapters_from_json(self, data, video_id, duration):
chapters_list = try_get( chapters_list = try_get(
@ -2035,7 +2036,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
headers = { headers = {
'X-YouTube-Client-Name': '85', 'X-YouTube-Client-Name': '85',
'X-YouTube-Client-Version': '2.0', 'X-YouTube-Client-Version': '2.0',
'Origin': 'https://www.youtube.com' 'Origin': 'https://www.youtube.com',
} }
video_info = self._call_api('player', query, video_id, fatal=False, headers=headers) video_info = self._call_api('player', query, video_id, fatal=False, headers=headers)
@ -2064,8 +2065,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
return ''.join([r['text'] for r in runs if isinstance(r.get('text'), compat_str)]) return ''.join([r['text'] for r in runs if isinstance(r.get('text'), compat_str)])
search_meta = ( search_meta = (
lambda x: self._html_search_meta(x, webpage, default=None)) \ (lambda x: self._html_search_meta(x, webpage, default=None))
if webpage else lambda x: None if webpage else lambda _: None)
video_details = player_response.get('videoDetails') or {} video_details = player_response.get('videoDetails') or {}
microformat = try_get( microformat = try_get(
@ -2137,7 +2138,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
def build_fragments(f): def build_fragments(f):
return LazyList({ return LazyList({
'url': update_url_query(f['url'], { 'url': update_url_query(f['url'], {
'range': '{0}-{1}'.format(range_start, min(range_start + CHUNK_SIZE - 1, f['filesize'])) 'range': '{0}-{1}'.format(range_start, min(range_start + CHUNK_SIZE - 1, f['filesize'])),
}) })
} for range_start in range(0, f['filesize'], CHUNK_SIZE)) } for range_start in range(0, f['filesize'], CHUNK_SIZE))
@ -2236,7 +2237,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'protocol': 'http_dash_segments', 'protocol': 'http_dash_segments',
'fragments': build_fragments(dct), 'fragments': build_fragments(dct),
} if dct['filesize'] else { } if dct['filesize'] else {
'downloader_options': {'http_chunk_size': CHUNK_SIZE} # No longer useful? 'downloader_options': {'http_chunk_size': CHUNK_SIZE}, # No longer useful?
}) })
formats.append(dct) formats.append(dct)
@ -2414,9 +2415,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'is_live': is_live, 'is_live': is_live,
} }
pctr = try_get( pctr = traverse_obj(
player_response, player_response,
lambda x: x['captions']['playerCaptionsTracklistRenderer'], dict) ('captions', 'playerCaptionsTracklistRenderer', T(dict)))
if pctr: if pctr:
def process_language(container, base_url, lang_code, query): def process_language(container, base_url, lang_code, query):
lang_subs = [] lang_subs = []
@ -2430,31 +2431,34 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
}) })
container[lang_code] = lang_subs container[lang_code] = lang_subs
subtitles = {} def process_subtitles():
for caption_track in (pctr.get('captionTracks') or []): subtitles = {}
base_url = caption_track.get('baseUrl') for caption_track in traverse_obj(pctr, (
if not base_url: 'captionTracks', lambda _, v: v.get('baseUrl'))):
continue base_url = self._yt_urljoin(caption_track['baseUrl'])
if caption_track.get('kind') != 'asr': if not base_url:
lang_code = caption_track.get('languageCode')
if not lang_code:
continue continue
process_language( if caption_track.get('kind') != 'asr':
subtitles, base_url, lang_code, {}) lang_code = caption_track.get('languageCode')
continue if not lang_code:
automatic_captions = {} continue
for translation_language in (pctr.get('translationLanguages') or []): process_language(
translation_language_code = translation_language.get('languageCode') subtitles, base_url, lang_code, {})
if not translation_language_code:
continue continue
process_language( automatic_captions = {}
automatic_captions, base_url, translation_language_code, for translation_language in traverse_obj(pctr, (
{'tlang': translation_language_code}) 'translationLanguages', lambda _, v: v.get('languageCode'))):
info['automatic_captions'] = automatic_captions translation_language_code = translation_language['languageCode']
info['subtitles'] = subtitles process_language(
automatic_captions, base_url, translation_language_code,
{'tlang': translation_language_code})
info['automatic_captions'] = automatic_captions
info['subtitles'] = subtitles
process_subtitles()
parsed_url = compat_urllib_parse_urlparse(url) parsed_url = compat_urllib_parse_urlparse(url)
for component in [parsed_url.fragment, parsed_url.query]: for component in (parsed_url.fragment, parsed_url.query):
query = compat_parse_qs(component) query = compat_parse_qs(component)
for k, v in query.items(): for k, v in query.items():
for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]: for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
@ -2684,7 +2688,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
'title': 'Super Cooper Shorts - Shorts', 'title': 'Super Cooper Shorts - Shorts',
'uploader': 'Super Cooper Shorts', 'uploader': 'Super Cooper Shorts',
'uploader_id': '@SuperCooperShorts', 'uploader_id': '@SuperCooperShorts',
} },
}, { }, {
# Channel that does not have a Shorts tab. Test should just download videos on Home tab instead # Channel that does not have a Shorts tab. Test should just download videos on Home tab instead
'url': 'https://www.youtube.com/@emergencyawesome/shorts', 'url': 'https://www.youtube.com/@emergencyawesome/shorts',
@ -2738,7 +2742,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
'description': 'md5:609399d937ea957b0f53cbffb747a14c', 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
'uploader': 'ThirstForScience', 'uploader': 'ThirstForScience',
'uploader_id': '@ThirstForScience', 'uploader_id': '@ThirstForScience',
} },
}, { }, {
'url': 'https://www.youtube.com/c/ChristophLaimer/playlists', 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
'only_matching': True, 'only_matching': True,
@ -3037,7 +3041,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
'uploader': '3Blue1Brown', 'uploader': '3Blue1Brown',
'uploader_id': '@3blue1brown', 'uploader_id': '@3blue1brown',
'channel_id': 'UCYO_jab_esuFRV4b17AJtAw', 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
} },
}] }]
@classmethod @classmethod
@ -3335,7 +3339,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
'client': { 'client': {
'clientName': 'WEB', 'clientName': 'WEB',
'clientVersion': client_version, 'clientVersion': client_version,
} },
} }
visitor_data = try_get(context, lambda x: x['client']['visitorData'], compat_str) visitor_data = try_get(context, lambda x: x['client']['visitorData'], compat_str)
@ -3354,7 +3358,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
headers['x-goog-visitor-id'] = visitor_data headers['x-goog-visitor-id'] = visitor_data
data['continuation'] = continuation['continuation'] data['continuation'] = continuation['continuation']
data['clickTracking'] = { data['clickTracking'] = {
'clickTrackingParams': continuation['itct'] 'clickTrackingParams': continuation['itct'],
} }
count = 0 count = 0
retries = 3 retries = 3
@ -3613,7 +3617,7 @@ class YoutubePlaylistIE(InfoExtractor):
'uploader': 'milan', 'uploader': 'milan',
'uploader_id': '@milan5503', 'uploader_id': '@milan5503',
'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw', 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
} },
}, { }, {
'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl', 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
'playlist_mincount': 455, 'playlist_mincount': 455,
@ -3623,7 +3627,7 @@ class YoutubePlaylistIE(InfoExtractor):
'uploader': 'LBK', 'uploader': 'LBK',
'uploader_id': '@music_king', 'uploader_id': '@music_king',
'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA', 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
} },
}, { }, {
'url': 'TLGGrESM50VT6acwMjAyMjAxNw', 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
'only_matching': True, 'only_matching': True,
@ -3734,7 +3738,7 @@ class YoutubeSearchIE(SearchInfoExtractor, YoutubeBaseInfoExtractor):
'info_dict': { 'info_dict': {
'id': 'youtube-dl test video', 'id': 'youtube-dl test video',
'title': 'youtube-dl test video', 'title': 'youtube-dl test video',
} },
}] }]
def _get_n_results(self, query, n): def _get_n_results(self, query, n):
@ -3754,7 +3758,7 @@ class YoutubeSearchDateIE(YoutubeSearchIE):
'info_dict': { 'info_dict': {
'id': 'youtube-dl test video', 'id': 'youtube-dl test video',
'title': 'youtube-dl test video', 'title': 'youtube-dl test video',
} },
}] }]
@ -3769,7 +3773,7 @@ class YoutubeSearchURLIE(YoutubeBaseInfoExtractor):
'id': 'youtube-dl test video', 'id': 'youtube-dl test video',
'title': 'youtube-dl test video', 'title': 'youtube-dl test video',
}, },
'params': {'playlistend': 5} 'params': {'playlistend': 5},
}, { }, {
'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB', 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
'only_matching': True, 'only_matching': True,
@ -3785,6 +3789,7 @@ class YoutubeSearchURLIE(YoutubeBaseInfoExtractor):
class YoutubeFeedsInfoExtractor(YoutubeTabIE): class YoutubeFeedsInfoExtractor(YoutubeTabIE):
""" """
Base class for feed extractors Base class for feed extractors
Subclasses must define the _FEED_NAME property. Subclasses must define the _FEED_NAME property.
""" """
_LOGIN_REQUIRED = True _LOGIN_REQUIRED = True

View File

@ -183,6 +183,11 @@ def parseOpts(overrideArguments=None):
action='store_const', dest='extract_flat', const='in_playlist', action='store_const', dest='extract_flat', const='in_playlist',
default=False, default=False,
help='Do not extract the videos of a playlist, only list them.') help='Do not extract the videos of a playlist, only list them.')
general.add_option(
'--concat-playlist',
action='store_true', dest='concat_playlist', default=False,
help='Concatenate all videos in a playlist into a single video file. Useful for services which split up full '
'episodes into multiple segments.')
general.add_option( general.add_option(
'--mark-watched', '--mark-watched',
action='store_true', dest='mark_watched', default=False, action='store_true', dest='mark_watched', default=False,

View File

@ -9,6 +9,7 @@ from .ffmpeg import (
FFmpegFixupM3u8PP, FFmpegFixupM3u8PP,
FFmpegFixupM4aPP, FFmpegFixupM4aPP,
FFmpegMergerPP, FFmpegMergerPP,
FFmpegConcatPP,
FFmpegMetadataPP, FFmpegMetadataPP,
FFmpegVideoConvertorPP, FFmpegVideoConvertorPP,
FFmpegSubtitlesConvertorPP, FFmpegSubtitlesConvertorPP,
@ -31,6 +32,7 @@ __all__ = [
'FFmpegFixupM4aPP', 'FFmpegFixupM4aPP',
'FFmpegFixupStretchedPP', 'FFmpegFixupStretchedPP',
'FFmpegMergerPP', 'FFmpegMergerPP',
'FFmpegConcatPP',
'FFmpegMetadataPP', 'FFmpegMetadataPP',
'FFmpegPostProcessor', 'FFmpegPostProcessor',
'FFmpegSubtitlesConvertorPP', 'FFmpegSubtitlesConvertorPP',

View File

@ -84,7 +84,7 @@ class EmbedThumbnailPP(FFmpegPostProcessor):
self._downloader.to_screen('[ffmpeg] Adding thumbnail to "%s"' % filename) self._downloader.to_screen('[ffmpeg] Adding thumbnail to "%s"' % filename)
self.run_ffmpeg_multiple_files([filename, thumbnail_filename], temp_filename, options) self.run_ffmpeg_multiple_files([filename, thumbnail_filename], temp_filename, options, [])
if not self._already_have_thumbnail: if not self._already_have_thumbnail:
os.remove(encodeFilename(thumbnail_filename)) os.remove(encodeFilename(thumbnail_filename))

View File

@ -5,7 +5,6 @@ import subprocess
import time import time
import re import re
from .common import AudioConversionError, PostProcessor from .common import AudioConversionError, PostProcessor
from ..compat import compat_open as open from ..compat import compat_open as open
@ -24,7 +23,6 @@ from ..utils import (
replace_extension, replace_extension,
) )
EXT_TO_OUT_FORMATS = { EXT_TO_OUT_FORMATS = {
'aac': 'adts', 'aac': 'adts',
'flac': 'flac', 'flac': 'flac',
@ -196,7 +194,7 @@ class FFmpegPostProcessor(PostProcessor):
return mobj.group(1) return mobj.group(1)
return None return None
def run_ffmpeg_multiple_files(self, input_paths, out_path, opts): def run_ffmpeg_multiple_files(self, input_paths, out_path, opts, file_opts):
self.check_version() self.check_version()
oldest_mtime = min( oldest_mtime = min(
@ -207,6 +205,7 @@ class FFmpegPostProcessor(PostProcessor):
files_cmd = [] files_cmd = []
for path in input_paths: for path in input_paths:
files_cmd.extend([ files_cmd.extend([
*file_opts,
encodeArgument('-i'), encodeArgument('-i'),
encodeFilename(self._ffmpeg_filename_argument(path), True) encodeFilename(self._ffmpeg_filename_argument(path), True)
]) ])
@ -231,8 +230,8 @@ class FFmpegPostProcessor(PostProcessor):
raise FFmpegPostProcessorError(msg) raise FFmpegPostProcessorError(msg)
self.try_utime(out_path, oldest_mtime, oldest_mtime) self.try_utime(out_path, oldest_mtime, oldest_mtime)
def run_ffmpeg(self, path, out_path, opts): def run_ffmpeg(self, path, out_path, opts, file_opts):
self.run_ffmpeg_multiple_files([path], out_path, opts) self.run_ffmpeg_multiple_files([path], out_path, opts, file_opts)
def _ffmpeg_filename_argument(self, fn): def _ffmpeg_filename_argument(self, fn):
# Always use 'file:' because the filename may contain ':' (ffmpeg # Always use 'file:' because the filename may contain ':' (ffmpeg
@ -270,7 +269,8 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor):
raise PostProcessingError('WARNING: unable to obtain file audio codec with ffprobe') raise PostProcessingError('WARNING: unable to obtain file audio codec with ffprobe')
more_opts = [] more_opts = []
if self._preferredcodec == 'best' or self._preferredcodec == filecodec or (self._preferredcodec == 'm4a' and filecodec == 'aac'): if self._preferredcodec == 'best' or self._preferredcodec == filecodec or (
self._preferredcodec == 'm4a' and filecodec == 'aac'):
if filecodec == 'aac' and self._preferredcodec in ['m4a', 'best']: if filecodec == 'aac' and self._preferredcodec in ['m4a', 'best']:
# Lossless, but in another container # Lossless, but in another container
acodec = 'copy' acodec = 'copy'
@ -315,7 +315,8 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor):
extension = 'wav' extension = 'wav'
more_opts += ['-f', 'wav'] more_opts += ['-f', 'wav']
prefix, sep, ext = path.rpartition('.') # not os.path.splitext, since the latter does not work on unicode in all setups prefix, sep, ext = path.rpartition(
'.') # not os.path.splitext, since the latter does not work on unicode in all setups
new_path = prefix + sep + extension new_path = prefix + sep + extension
information['filepath'] = new_path information['filepath'] = new_path
@ -353,14 +354,16 @@ class FFmpegVideoConvertorPP(FFmpegPostProcessor):
def run(self, information): def run(self, information):
path = information['filepath'] path = information['filepath']
if information['ext'] == self._preferedformat: if information['ext'] == self._preferedformat:
self._downloader.to_screen('[ffmpeg] Not converting video file %s - already is in target format %s' % (path, self._preferedformat)) self._downloader.to_screen(
'[ffmpeg] Not converting video file %s - already is in target format %s' % (path, self._preferedformat))
return [], information return [], information
options = [] options = []
if self._preferedformat == 'avi': if self._preferedformat == 'avi':
options.extend(['-c:v', 'libxvid', '-vtag', 'XVID']) options.extend(['-c:v', 'libxvid', '-vtag', 'XVID'])
prefix, sep, ext = path.rpartition('.') prefix, sep, ext = path.rpartition('.')
outpath = prefix + sep + self._preferedformat outpath = prefix + sep + self._preferedformat
self._downloader.to_screen('[' + 'ffmpeg' + '] Converting video from %s to %s, Destination: ' % (information['ext'], self._preferedformat) + outpath) self._downloader.to_screen('[' + 'ffmpeg' + '] Converting video from %s to %s, Destination: ' % (
information['ext'], self._preferedformat) + outpath)
self.run_ffmpeg(path, outpath, options) self.run_ffmpeg(path, outpath, options)
information['filepath'] = outpath information['filepath'] = outpath
information['format'] = self._preferedformat information['format'] = self._preferedformat
@ -419,7 +422,7 @@ class FFmpegEmbedSubtitlePP(FFmpegPostProcessor):
temp_filename = prepend_extension(filename, 'temp') temp_filename = prepend_extension(filename, 'temp')
self._downloader.to_screen('[ffmpeg] Embedding subtitles in \'%s\'' % filename) self._downloader.to_screen('[ffmpeg] Embedding subtitles in \'%s\'' % filename)
self.run_ffmpeg_multiple_files(input_files, temp_filename, opts) self.run_ffmpeg_multiple_files(input_files, temp_filename, opts, [])
os.remove(encodeFilename(filename)) os.remove(encodeFilename(filename))
os.rename(encodeFilename(temp_filename), encodeFilename(filename)) os.rename(encodeFilename(temp_filename), encodeFilename(filename))
@ -502,7 +505,7 @@ class FFmpegMetadataPP(FFmpegPostProcessor):
options.extend(['-map_metadata', '1']) options.extend(['-map_metadata', '1'])
self._downloader.to_screen('[ffmpeg] Adding metadata to \'%s\'' % filename) self._downloader.to_screen('[ffmpeg] Adding metadata to \'%s\'' % filename)
self.run_ffmpeg_multiple_files(in_filenames, temp_filename, options) self.run_ffmpeg_multiple_files(in_filenames, temp_filename, options, [])
if chapters: if chapters:
os.remove(metadata_filename) os.remove(metadata_filename)
os.remove(encodeFilename(filename)) os.remove(encodeFilename(filename))
@ -516,7 +519,7 @@ class FFmpegMergerPP(FFmpegPostProcessor):
temp_filename = prepend_extension(filename, 'temp') temp_filename = prepend_extension(filename, 'temp')
args = ['-c', 'copy', '-map', '0:v:0', '-map', '1:a:0'] args = ['-c', 'copy', '-map', '0:v:0', '-map', '1:a:0']
self._downloader.to_screen('[ffmpeg] Merging formats into "%s"' % filename) self._downloader.to_screen('[ffmpeg] Merging formats into "%s"' % filename)
self.run_ffmpeg_multiple_files(info['__files_to_merge'], temp_filename, args) self.run_ffmpeg_multiple_files(info['__files_to_merge'], temp_filename, args, [])
os.rename(encodeFilename(temp_filename), encodeFilename(filename)) os.rename(encodeFilename(temp_filename), encodeFilename(filename))
return info['__files_to_merge'], info return info['__files_to_merge'], info
@ -538,6 +541,29 @@ class FFmpegMergerPP(FFmpegPostProcessor):
return True return True
class FFmpegConcatPP(FFmpegPostProcessor):
def run(self, info):
filename = info['filepath']
list_filename = prepend_extension(filename, 'list')
temp_filename = prepend_extension(filename, 'temp')
with open(list_filename, 'wt') as f:
f.write('ffconcat version 1.0\n')
for file in info['__files_to_concat']:
f.write("file '%s'\n" % self._ffmpeg_filename_argument(file))
file_opts = ['-f', 'concat', '-safe', '0']
opts = ['-c', 'copy']
self._downloader.to_screen('[ffmpeg] Concatenating files into "%s"' % filename)
self.run_ffmpeg(list_filename, temp_filename, opts, file_opts)
os.rename(encodeFilename(temp_filename), encodeFilename(filename))
return info['__files_to_concat'], info
def can_merge(self):
return True
class FFmpegFixupStretchedPP(FFmpegPostProcessor): class FFmpegFixupStretchedPP(FFmpegPostProcessor):
def run(self, info): def run(self, info):
stretched_ratio = info.get('stretched_ratio') stretched_ratio = info.get('stretched_ratio')
@ -583,7 +609,7 @@ class FFmpegFixupM3u8PP(FFmpegPostProcessor):
options = ['-c', 'copy', '-f', 'mp4', '-bsf:a', 'aac_adtstoasc'] options = ['-c', 'copy', '-f', 'mp4', '-bsf:a', 'aac_adtstoasc']
self._downloader.to_screen('[ffmpeg] Fixing malformed AAC bitstream in "%s"' % filename) self._downloader.to_screen('[ffmpeg] Fixing malformed AAC bitstream in "%s"' % filename)
self.run_ffmpeg(filename, temp_filename, options) self.run_ffmpeg(filename, temp_filename, options, [])
os.remove(encodeFilename(filename)) os.remove(encodeFilename(filename))
os.rename(encodeFilename(temp_filename), encodeFilename(filename)) os.rename(encodeFilename(temp_filename), encodeFilename(filename))