X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/3f03cde7c013b3ba884e1968702d9bdf8d987f95..9d6a3ba754469c85b7f0f88714203f99ac69087a:/youtube_dl/extractor/youtube.py?ds=sidebyside
diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index e7bd1f1..89c8b7f 100644
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -37,6 +37,7 @@ from ..utils import (
orderedSet,
parse_codecs,
parse_duration,
+ qualities,
remove_quotes,
remove_start,
smuggle_url,
@@ -84,7 +85,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
"""
- (username, password) = self._get_login_info()
+ username, password = self._get_login_info()
# No authentication to be performed
if username is None:
if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
@@ -246,9 +247,9 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
return True
- def _download_webpage(self, *args, **kwargs):
+ def _download_webpage_handle(self, *args, **kwargs):
kwargs.setdefault('query', {})['disable_polymer'] = 'true'
- return super(YoutubeBaseInfoExtractor, self)._download_webpage(
+ return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
*args, **compat_kwargs(kwargs))
def _real_initialize(self):
@@ -509,6 +510,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
'license': 'Standard YouTube License',
'creator': 'Icona Pop',
+ 'track': 'I Love It (feat. Charli XCX)',
+ 'artist': 'Icona Pop',
}
},
{
@@ -527,6 +530,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
'license': 'Standard YouTube License',
'creator': 'Justin Timberlake',
+ 'track': 'Tunnel Vision',
+ 'artist': 'Justin Timberlake',
'age_limit': 18,
}
},
@@ -596,7 +601,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'id': 'IB3lcPjvWLA',
'ext': 'm4a',
'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
- 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
+ 'description': 'md5:1900ed86ee514927b9e00fbead6969a5',
'duration': 244,
'uploader': 'AfrojackVEVO',
'uploader_id': 'AfrojackVEVO',
@@ -637,7 +642,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'ext': 'mp4',
'duration': 219,
'upload_date': '20100909',
- 'uploader': 'The Amazing Atheist',
+ 'uploader': 'TJ Kirk',
'uploader_id': 'TheAmazingAtheist',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
'license': 'Standard YouTube License',
@@ -667,10 +672,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
'info_dict': {
'id': '6kLq3WMV1nU',
- 'ext': 'mp4',
+ 'ext': 'webm',
'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
- 'duration': 247,
+ 'duration': 246,
'uploader': 'LloydVEVO',
'uploader_id': 'LloydVEVO',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
@@ -732,7 +737,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'uploader_id': 'AllenMeow',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
'description': 'made by Wacom from Korea | åå¹&å 油添é by TY\'s Allen | æè¬heylisa00cavey1001åå¸ç±æ
æä¾æ¢åç¿»è¯',
- 'uploader': 'å«è¾å«',
+ 'uploader': 'å«áá
',
'license': 'Standard YouTube License',
'title': '[A-made] è®æ
å¦åå¹ç å¤ªå¦ æå°±æ¯é樣ç人',
},
@@ -759,7 +764,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
'info_dict': {
'id': 'FIl7x6_3R5Y',
- 'ext': 'mp4',
+ 'ext': 'webm',
'title': 'md5:7b81415841e02ecd4313668cde88737a',
'description': 'md5:116377fd2963b81ec4ce64b542173306',
'duration': 220,
@@ -768,8 +773,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
'uploader': 'dorappi2000',
'license': 'Standard YouTube License',
- 'formats': 'mincount:32',
+ 'formats': 'mincount:31',
},
+ 'skip': 'not actual anymore',
},
# DASH manifest with segment_list
{
@@ -884,7 +890,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'id': 'lsguqyKfVQg',
'ext': 'mp4',
'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
- 'alt_title': 'Dark Walk',
+ 'alt_title': 'Dark Walk - Position Music',
'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
'duration': 133,
'upload_date': '20151119',
@@ -892,7 +898,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
'uploader': 'IronSoulElf',
'license': 'Standard YouTube License',
- 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
+ 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
+ 'track': 'Dark Walk - Position Music',
+ 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
},
'params': {
'skip_download': True,
@@ -949,7 +957,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'description': 'md5:dda0d780d5a6e120758d1711d062a867',
'duration': 4060,
'upload_date': '20151119',
- 'uploader': 'Bernie 2016',
+ 'uploader': 'Bernie Sanders',
'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
'license': 'Creative Commons Attribution license (reuse allowed)',
@@ -984,6 +992,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'params': {
'skip_download': True,
},
+ 'skip': 'This video is not available.',
},
{
# YouTube Red video with episode data
@@ -992,7 +1001,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'id': 'iqKdEhx-dD4',
'ext': 'mp4',
'title': 'Isolation - Mind Field (Ep 1)',
- 'description': 'md5:8013b7ddea787342608f63a13ddc9492',
+ 'description': 'md5:25b78d2f64ae81719f5c96319889b736',
'duration': 2085,
'upload_date': '20170118',
'uploader': 'Vsauce',
@@ -1025,7 +1034,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
'license': 'Standard YouTube License',
- 'view_count': int,
},
'params': {
'skip_download': True,
@@ -1537,7 +1545,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
if ytplayer_config:
args = ytplayer_config['args']
- if args.get('url_encoded_fmt_stream_map'):
+ if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
# Convert to the same format returned by compat_parse_qs
video_info = dict((k, [v]) for k, v in args.items())
add_dash_mpd(video_info)
@@ -1693,125 +1701,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
- # Start extracting information
- self.report_information_extraction(video_id)
-
- # uploader
- if 'author' not in video_info:
- raise ExtractorError('Unable to extract uploader name')
- video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
-
- # uploader_id
- video_uploader_id = None
- video_uploader_url = None
- mobj = re.search(
- r'',
- video_webpage)
- if mobj is not None:
- video_uploader_id = mobj.group('uploader_id')
- video_uploader_url = mobj.group('uploader_url')
- else:
- self._downloader.report_warning('unable to extract uploader nickname')
-
- # thumbnail image
- # We try first to get a high quality image:
- m_thumb = re.search(r'',
- video_webpage, re.DOTALL)
- if m_thumb is not None:
- video_thumbnail = m_thumb.group(1)
- elif 'thumbnail_url' not in video_info:
- self._downloader.report_warning('unable to extract video thumbnail')
- video_thumbnail = None
- else: # don't panic if we can't find it
- video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
-
- # upload date
- upload_date = self._html_search_meta(
- 'datePublished', video_webpage, 'upload date', default=None)
- if not upload_date:
- upload_date = self._search_regex(
- [r'(?s)id="eow-date.*?>(.*?)',
- r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
- video_webpage, 'upload date', default=None)
- upload_date = unified_strdate(upload_date)
-
- video_license = self._html_search_regex(
- r'
]+class="title"[^>]*>\s*License\s*
\s*]*>\s*- (.+?)
]+class="title"[^>]*>\s*Music\s*\s*
- ]*>\s*
- - (?P.+?)
- by (?P.+?)
- (?:
- \(.+?\)|
- ]*
- (?:
- \bhref=["\']/red[^>]*>| # drop possible
- >\s*Listen ad-free with YouTube Red # YouTube Red ad
- )
- .*?
- )?
]+id="watch7-headline"[^>]*>\s*]*>.*?>(?P[^<]+)\s*S(?P\d+)\s*â¢\s*E(?P\d+)',
- video_webpage)
- if m_episode:
- series = m_episode.group('series')
- season_number = int(m_episode.group('season'))
- episode_number = int(m_episode.group('episode'))
- else:
- series = season_number = episode_number = None
-
- m_cat_container = self._search_regex(
- r'(?s)]*>\s*Category\s*
\s*',
- video_webpage, 'categories', default=None)
- if m_cat_container:
- category = self._html_search_regex(
- r'(?s)(.*?)', m_cat_container, 'category',
- default=None)
- video_categories = None if category is None else [category]
- else:
- video_categories = None
-
- video_tags = [
- unescapeHTML(m.group('content'))
- for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
-
- def _extract_count(count_name):
- return str_to_int(self._search_regex(
- r'-%s-button[^>]+>]+class="yt-uix-button-content"[^>]*>([\d,]+)'
- % re.escape(count_name),
- video_webpage, count_name, default=None))
-
- like_count = _extract_count('like')
- dislike_count = _extract_count('dislike')
-
- # subtitles
- video_subtitles = self.extract_subtitles(video_id, video_webpage)
- automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
-
- video_duration = try_get(
- video_info, lambda x: int_or_none(x['length_seconds'][0]))
- if not video_duration:
- video_duration = parse_duration(self._html_search_meta(
- 'duration', video_webpage, 'video duration'))
-
- # annotations
- video_annotations = None
- if self._downloader.params.get('writeannotations', False):
- video_annotations = self._extract_annotations(video_id)
-
- chapters = self._extract_chapters(description_original, video_duration)
+ def _extract_filesize(media_url):
+ return int_or_none(self._search_regex(
+ r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
self.report_rtmp_download()
@@ -1838,6 +1730,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'width': int_or_none(width_height[0]),
'height': int_or_none(width_height[1]),
}
+ q = qualities(['small', 'medium', 'hd720'])
formats = []
for url_data_str in encoded_url_map.split(','):
url_data = compat_parse_qs(url_data_str)
@@ -1917,13 +1810,19 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
mobj = re.search(r'^(?P\d+)[xX](?P\d+)$', url_data.get('size', [''])[0])
width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
+ filesize = int_or_none(url_data.get(
+ 'clen', [None])[0]) or _extract_filesize(url)
+
+ quality = url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0]
+
more_fields = {
- 'filesize': int_or_none(url_data.get('clen', [None])[0]),
+ 'filesize': filesize,
'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
'width': width,
'height': height,
'fps': int_or_none(url_data.get('fps', [None])[0]),
- 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
+ 'format_note': quality,
+ 'quality': q(quality),
}
for key, value in more_fields.items():
if value:
@@ -1969,11 +1868,140 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
formats.append(a_format)
else:
- unavailable_message = extract_unavailable_message()
- if unavailable_message:
- raise ExtractorError(unavailable_message, expected=True)
+ error_message = clean_html(video_info.get('reason', [None])[0])
+ if not error_message:
+ error_message = extract_unavailable_message()
+ if error_message:
+ raise ExtractorError(error_message, expected=True)
raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
+ # uploader
+ video_uploader = try_get(video_info, lambda x: x['author'][0], compat_str)
+ if video_uploader:
+ video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
+ else:
+ self._downloader.report_warning('unable to extract uploader name')
+
+ # uploader_id
+ video_uploader_id = None
+ video_uploader_url = None
+ mobj = re.search(
+ r'',
+ video_webpage)
+ if mobj is not None:
+ video_uploader_id = mobj.group('uploader_id')
+ video_uploader_url = mobj.group('uploader_url')
+ else:
+ self._downloader.report_warning('unable to extract uploader nickname')
+
+ # thumbnail image
+ # We try first to get a high quality image:
+ m_thumb = re.search(r'',
+ video_webpage, re.DOTALL)
+ if m_thumb is not None:
+ video_thumbnail = m_thumb.group(1)
+ elif 'thumbnail_url' not in video_info:
+ self._downloader.report_warning('unable to extract video thumbnail')
+ video_thumbnail = None
+ else: # don't panic if we can't find it
+ video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
+
+ # upload date
+ upload_date = self._html_search_meta(
+ 'datePublished', video_webpage, 'upload date', default=None)
+ if not upload_date:
+ upload_date = self._search_regex(
+ [r'(?s)id="eow-date.*?>(.*?)',
+ r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
+ video_webpage, 'upload date', default=None)
+ upload_date = unified_strdate(upload_date)
+
+ video_license = self._html_search_regex(
+ r']+class="title"[^>]*>\s*License\s*
\s*]*>\s*- (.+?)
]+class="title"[^>]*>\s*Music\s*\s*
+ ]*>\s*
+ - (?P.+?)
+ by (?P.+?)
+ (?:
+ \(.+?\)|
+ ]*
+ (?:
+ \bhref=["\']/red[^>]*>| # drop possible
+ >\s*Listen ad-free with YouTube Red # YouTube Red ad
+ )
+ .*?
+ )?
]+class="title"[^>]*>\s*%s\s*\s*]*>\s*- (.+?)
\s*' % field,
+ video_webpage, field, default=None)
+
+ track = extract_meta('Song')
+ artist = extract_meta('Artist')
+
+ m_episode = re.search(
+ r']+id="watch7-headline"[^>]*>\s*
]*>.*?>(?P[^<]+)\s*S(?P\d+)\s*â¢\s*E(?P\d+)',
+ video_webpage)
+ if m_episode:
+ series = m_episode.group('series')
+ season_number = int(m_episode.group('season'))
+ episode_number = int(m_episode.group('episode'))
+ else:
+ series = season_number = episode_number = None
+
+ m_cat_container = self._search_regex(
+ r'(?s)
]*>\s*Category\s*
\s*
',
+ video_webpage, 'categories', default=None)
+ if m_cat_container:
+ category = self._html_search_regex(
+ r'(?s)
(.*?)', m_cat_container, 'category',
+ default=None)
+ video_categories = None if category is None else [category]
+ else:
+ video_categories = None
+
+ video_tags = [
+ unescapeHTML(m.group('content'))
+ for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
+
+ def _extract_count(count_name):
+ return str_to_int(self._search_regex(
+ r'-%s-button[^>]+>
]+class="yt-uix-button-content"[^>]*>([\d,]+)'
+ % re.escape(count_name),
+ video_webpage, count_name, default=None))
+
+ like_count = _extract_count('like')
+ dislike_count = _extract_count('dislike')
+
+ # subtitles
+ video_subtitles = self.extract_subtitles(video_id, video_webpage)
+ automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
+
+ video_duration = try_get(
+ video_info, lambda x: int_or_none(x['length_seconds'][0]))
+ if not video_duration:
+ video_duration = parse_duration(self._html_search_meta(
+ 'duration', video_webpage, 'video duration'))
+
+ # annotations
+ video_annotations = None
+ if self._downloader.params.get('writeannotations', False):
+ video_annotations = self._extract_annotations(video_id)
+
+ chapters = self._extract_chapters(description_original, video_duration)
+
# Look for the DASH manifest
if self._downloader.params.get('youtube_include_dash_manifest', True):
dash_mpd_fatal = True
@@ -1990,6 +2018,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
for df in self._extract_mpd_formats(
mpd_url, video_id, fatal=dash_mpd_fatal,
formats_dict=self._formats):
+ if not df.get('filesize'):
+ df['filesize'] = _extract_filesize(df['url'])
# Do not overwrite DASH format found in some previous DASH manifest
if df['format_id'] not in dash_formats:
dash_formats[df['format_id']] = df
@@ -2037,9 +2067,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'uploader_url': video_uploader_url,
'upload_date': upload_date,
'license': video_license,
- 'creator': video_creator,
+ 'creator': video_creator or artist,
'title': video_title,
- 'alt_title': video_alt_title,
+ 'alt_title': video_alt_title or track,
'thumbnail': video_thumbnail,
'description': video_description,
'categories': video_categories,
@@ -2062,6 +2092,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'series': series,
'season_number': season_number,
'episode_number': episode_number,
+ 'track': track,
+ 'artist': artist,
}