[youtube] Separate methods for embeds extraction

This commit is contained in:
Sergey M․ 2017-09-06 00:48:37 +07:00
parent c5c9bf0c12
commit 66c9fa36c1
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D
2 changed files with 41 additions and 29 deletions

View File

@ -2243,36 +2243,11 @@ class GenericIE(InfoExtractor):
if vid_me_embed_url is not None: if vid_me_embed_url is not None:
return self.url_result(vid_me_embed_url, 'Vidme') return self.url_result(vid_me_embed_url, 'Vidme')
# Look for embedded YouTube player # Look for YouTube embeds
matches = re.findall(r'''(?x) youtube_urls = YoutubeIE._extract_urls(webpage)
(?: if youtube_urls:
<iframe[^>]+?src=|
data-video-url=|
<embed[^>]+?src=|
embedSWF\(?:\s*|
<object[^>]+data=|
new\s+SWFObject\(
)
(["\'])
(?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
(?:embed|v|p)/.+?)
\1''', webpage)
if matches:
return self.playlist_from_matches( return self.playlist_from_matches(
matches, video_id, video_title, lambda m: unescapeHTML(m[1])) youtube_urls, video_id, video_title, ie=YoutubeIE.ie_key())
# Look for lazyYT YouTube embed
matches = re.findall(
r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)
if matches:
return self.playlist_from_matches(matches, video_id, video_title, lambda m: unescapeHTML(m))
# Look for Wordpress "YouTube Video Importer" plugin
matches = re.findall(r'''(?x)<div[^>]+
class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
if matches:
return self.playlist_from_matches(matches, video_id, video_title, lambda m: m[-1])
matches = DailymotionIE._extract_urls(webpage) matches = DailymotionIE._extract_urls(webpage)
if matches: if matches:

View File

@ -1374,6 +1374,43 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
playback_url, video_id, 'Marking watched', playback_url, video_id, 'Marking watched',
'Unable to mark watched', fatal=False) 'Unable to mark watched', fatal=False)
@staticmethod
def _extract_urls(webpage):
# Embedded YouTube player
entries = [
unescapeHTML(mobj.group('url'))
for mobj in re.finditer(r'''(?x)
(?:
<iframe[^>]+?src=|
data-video-url=|
<embed[^>]+?src=|
embedSWF\(?:\s*|
<object[^>]+data=|
new\s+SWFObject\(
)
(["\'])
(?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
(?:embed|v|p)/.+?)
\1''', webpage)]
# lazyYT YouTube embed
entries.extend(list(map(
unescapeHTML,
re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
# Wordpress "YouTube Video Importer" plugin
matches = re.findall(r'''(?x)<div[^>]+
class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
entries.extend(m[-1] for m in matches)
return entries
@staticmethod
def _extract_url(webpage):
urls = YoutubeIE._extract_urls(webpage)
return urls[0] if urls else None
@classmethod @classmethod
def extract_id(cls, url): def extract_id(cls, url):
mobj = re.match(cls._VALID_URL, url, re.VERBOSE) mobj = re.match(cls._VALID_URL, url, re.VERBOSE)