MarcoBuster/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-24 04:05:53 +01:00
Code Issues Projects Releases Wiki Activity

[glomex] Simplify embed detection (#2600)

Browse Source 
Closes #2512
This commit is contained in:
pukkandan 2022-02-03 13:16:47 +05:30 committed by GitHub
parent 54bb39065c
commit 5e51f4a8ad
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 30 additions and 40 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

70
yt_dlp/extractor/glomex.py
View File

@ -7,6 +7,7 @@
from .common import InfoExtractor
from ..utils import (
determine_ext,
extract_attributes,
ExtractorError,
int_or_none,
parse_qs,
@ -177,49 +178,38 @@ def build_player_url(cls, video_id, integration, origin_url=None):
@classmethod
def _extract_urls(cls, webpage, origin_url):
VALID_SRC = rf'(?:https?:)?{cls._BASE_PLAYER_URL_RE}\?(?:(?!(?P=_q1)).)+'
# https://docs.glomex.com/publisher/video-player-integration/javascript-api/
EMBED_RE = r'''(?x)(?:
<iframe[^>]+?src=(?P<_q1>%(quot_re)s)(?P<url>%(url_re)s)(?P=_q1)|
<(?P<html_tag>glomex-player|div)(?:
data-integration-id=(?P<_q2>%(quot_re)s)(?P<integration_html>(?:(?!(?P=_q2)).)+)(?P=_q2)|
data-playlist-id=(?P<_q3>%(quot_re)s)(?P<id_html>(?:(?!(?P=_q3)).)+)(?P=_q3)|
data-glomex-player=(?P<_q4>%(quot_re)s)(?P<glomex_player>true)(?P=_q4)|
[^>]*?
)+>|
# naive parsing of inline scripts for hard-coded integration parameters
<(?P<script_tag>script)[^<]*?>(?:
(?P<_stjs1>dataset\.)?integrationId\s*(?(_stjs1)=|:)\s*
(?P<_q5>%(quot_re)s)(?P<integration_js>(?:(?!(?P=_q5)).)+)(?P=_q5)\s*(?(_stjs1);|,)?|
(?P<_stjs2>dataset\.)?playlistId\s*(?(_stjs2)=|:)\s*
(?P<_q6>%(quot_re)s)(?P<id_js>(?:(?!(?P=_q6)).)+)(?P=_q6)\s*(?(_stjs2);|,)?|
(?:\s|.)*?
)+</script>
)''' % {'quot_re': r'["\']', 'url_re': VALID_SRC}
quot_re = r'["\']'
for mtup in re.findall(EMBED_RE, webpage):
# re.finditer causes a memory spike. See https://github.com/yt-dlp/yt-dlp/issues/2512
mdict = dict(zip((
'url', '_',
'html_tag', '_', 'integration_html', '_', 'id_html', '_', 'glomex_player',
'script_tag', '_', '_', 'integration_js', '_', 'id_js',
), mtup))
if mdict.get('url'):
url = unescapeHTML(mdict['url'])
if not cls.suitable(url):
continue
regex = fr'''(?x)
<iframe[^>]+?src=(?P<q>{quot_re})(?P<url>
(?:https?:)?{cls._BASE_PLAYER_URL_RE}\?(?:(?!(?P=q)).)+
)(?P=q)'''
for mobj in re.finditer(regex, webpage):
url = unescapeHTML(mobj.group('url'))
if cls.suitable(url):
yield cls._smuggle_origin_url(url, origin_url)
elif mdict.get('html_tag'):
if mdict['html_tag'] == 'div' and not mdict.get('glomex_player'):
continue
if not mdict.get('video_id_html') or not mdict.get('integration_html'):
continue
yield cls.build_player_url(mdict['video_id_html'], mdict['integration_html'], origin_url)
elif mdict.get('script_tag'):
if not mdict.get('video_id_js') or not mdict.get('integration_js'):
continue
yield cls.build_player_url(mdict['video_id_js'], mdict['integration_js'], origin_url)
regex = fr'''(?x)
<glomex-player [^>]+?>|
<div[^>]* data-glomex-player=(?P<q>{quot_re})true(?P=q)[^>]*>'''
for mobj in re.finditer(regex, webpage):
attrs = extract_attributes(mobj.group(0))
if attrs.get('data-integration-id') and attrs.get('data-playlist-id'):
yield cls.build_player_url(attrs['data-playlist-id'], attrs['data-integration-id'], origin_url)
# naive parsing of inline scripts for hard-coded integration parameters
regex = fr'''(?x)
(?P<is_js>dataset\.)?%s\s*(?(is_js)=|:)\s*
(?P<q>{quot_re})(?P<id>(?:(?!(?P=q)).)+)(?P=q)\s'''
for mobj in re.finditer(r'(?x)<script[^<]*>.+?</script>', webpage):
script = mobj.group(0)
integration_id = re.search(regex % 'integrationId', script)
if not integration_id:
continue
playlist_id = re.search(regex % 'playlistId', script)
if playlist_id:
yield cls.build_player_url(playlist_id, integration_id, origin_url)
def _real_extract(self, url):
url, origin_url = self._unsmuggle_origin_url(url)