github-com-ytdl-org/youtube-dl
1
0
Fork 0
mirror of https://github.com/ytdl-org/youtube-dl.git synced 2024-01-07 17:16:08 +00:00
Code Issues Releases Wiki Activity

[youtube] Fix playlist extraction (Closes #2423, #2424, #2425)

Browse source
This commit is contained in:
Sergey M. 2014-02-21 17:19:55 +07:00
parent eae16eb67b
commit dbb94fb044
1 changed files with 16 additions and 14 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

28
youtube_dl/extractor/youtube.py
View file

@ -1443,9 +1443,9 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
| |
((?:PL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,}) ((?:PL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
)""" )"""
_TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&page=%s' _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
_MORE_PAGES_INDICATOR = r'data-link-type="next"' _MORE_PAGES_INDICATOR = r'data-link-type="next"'
_VIDEO_RE = r'href="/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)' _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)'
IE_NAME = u'youtube:playlist' IE_NAME = u'youtube:playlist'
def _real_initialize(self): def _real_initialize(self):
@ -1493,29 +1493,31 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
raise ExtractorError(u'For downloading YouTube.com top lists, use ' raise ExtractorError(u'For downloading YouTube.com top lists, use '
u'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True) u'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True)
url = self._TEMPLATE_URL % playlist_id
page = self._download_webpage(url, playlist_id)
more_widget_html = content_html = page
# Extract the video ids from the playlist pages # Extract the video ids from the playlist pages
ids = [] ids = []
for page_num in itertools.count(1): for page_num in itertools.count(1):
url = self._TEMPLATE_URL % (playlist_id, page_num) matches = re.finditer(self._VIDEO_RE, content_html)
page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num)
matches = re.finditer(self._VIDEO_RE, page)
# We remove the duplicates and the link with index 0 # We remove the duplicates and the link with index 0
# (it's not the first video of the playlist) # (it's not the first video of the playlist)
new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0') new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
ids.extend(new_ids) ids.extend(new_ids)
if re.search(self._MORE_PAGES_INDICATOR, page) is None: mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
if not mobj:
break break
try: more = self._download_json(
playlist_title = self._og_search_title(page) 'https://youtube.com/%s' % mobj.group('more'), playlist_id, 'Downloading page #%s' % page_num)
except RegexNotFoundError: content_html = more['content_html']
self.report_warning( more_widget_html = more['load_more_widget_html']
u'Playlist page is missing OpenGraph title, falling back ...',
playlist_id)
playlist_title = self._html_search_regex( playlist_title = self._html_search_regex(
r'<h1 class="pl-header-title">(.*?)</h1>', page, u'title') r'<h1 class="pl-header-title">\s*(.*?)\s*</h1>', page, u'title')
url_results = self._ids_to_results(ids) url_results = self._ids_to_results(ids)
return self.playlist_result(url_results, playlist_id, playlist_title) return self.playlist_result(url_results, playlist_id, playlist_title)