-from __future__ import unicode_literals
-
import re
from .common import InfoExtractor
from ..utils import (
+ extract_attributes,
int_or_none,
+ merge_dicts,
str_to_int,
- unescapeHTML,
+ traverse_obj,
unified_strdate,
url_or_none,
)
-from ..aes import aes_decrypt_text
class YouPornIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?youporn\.com/(?:watch|embed)/(?P<id>\d+)(?:/(?P<display_id>[^/?#&]+))?'
+ _EMBED_REGEX = [r'<iframe[^>]+\bsrc=["\'](?P<url>(?:https?:)?//(?:www\.)?youporn\.com/embed/\d+)']
_TESTS = [{
'url': 'http://www.youporn.com/watch/505835/sex-ed-is-it-safe-to-masturbate-daily/',
'md5': '3744d24c50438cf5b6f6d59feb5055c2',
'tags': list,
'age_limit': 18,
},
+ 'skip': 'This video has been disabled',
}, {
# Unknown uploader
'url': 'http://www.youporn.com/watch/561726/big-tits-awesome-brunette-on-amazing-webcam-show/?from=related3&al=2&from_id=561726&pos=4',
}, {
'url': 'https://www.youporn.com/watch/13922959/femdom-principal/',
'only_matching': True,
+ }, {
+ 'url': 'https://www.youporn.com/watch/16290308/tinderspecial-trailer1/',
+ 'info_dict': {
+ 'id': '16290308',
+ 'age_limit': 18,
+ 'categories': [],
+ 'description': 'md5:00ea70f642f431c379763c17c2f396bc',
+ 'display_id': 'tinderspecial-trailer1',
+ 'duration': 298.0,
+ 'ext': 'mp4',
+ 'upload_date': '20201123',
+ 'uploader': 'Ersties',
+ 'tags': [],
+ 'thumbnail': 'https://fi1.ypncdn.com/202011/23/16290308/original/8/tinderspecial-trailer1-8(m=eaAaaEPbaaaa).jpg',
+ 'timestamp': 1606089600,
+ 'title': 'Tinder In Real Life',
+ 'view_count': int,
+ }
}]
- @staticmethod
- def _extract_urls(webpage):
- return re.findall(
- r'<iframe[^>]+\bsrc=["\']((?:https?:)?//(?:www\.)?youporn\.com/embed/\d+)',
- webpage)
-
def _real_extract(self, url):
- mobj = re.match(self._VALID_URL, url)
- video_id = mobj.group('id')
- display_id = mobj.group('display_id') or video_id
-
- webpage = self._download_webpage(
- 'http://www.youporn.com/watch/%s' % video_id, display_id,
- headers={'Cookie': 'age_verified=1'})
-
- title = self._html_search_regex(
- r'(?s)<div[^>]+class=["\']watchVideoTitle[^>]+>(.+?)</div>',
- webpage, 'title', default=None) or self._og_search_title(
- webpage, default=None) or self._html_search_meta(
- 'title', webpage, fatal=True)
+ video_id, display_id = self._match_valid_url(url).group('id', 'display_id')
+ definitions = self._download_json(
+ f'https://www.youporn.com/api/video/media_definitions/{video_id}/', display_id or video_id)
- links = []
-
- # Main source
- definitions = self._parse_json(
- self._search_regex(
- r'mediaDefinition\s*[=:]\s*(\[.+?\])\s*[;,]', webpage,
- 'media definitions', default='[]'),
- video_id, fatal=False)
- if definitions:
- for definition in definitions:
- if not isinstance(definition, dict):
- continue
- video_url = url_or_none(definition.get('videoUrl'))
- if video_url:
- links.append(video_url)
-
- # Fallback #1, this also contains extra low quality 180p format
- for _, link in re.findall(r'<a[^>]+href=(["\'])(http(?:(?!\1).)+\.mp4(?:(?!\1).)*)\1[^>]+title=["\']Download [Vv]ideo', webpage):
- links.append(link)
-
- # Fallback #2 (unavailable as at 22.06.2017)
- sources = self._search_regex(
- r'(?s)sources\s*:\s*({.+?})', webpage, 'sources', default=None)
- if sources:
- for _, link in re.findall(r'[^:]+\s*:\s*(["\'])(http.+?)\1', sources):
- links.append(link)
-
- # Fallback #3 (unavailable as at 22.06.2017)
- for _, link in re.findall(
- r'(?:videoSrc|videoIpadUrl|html5PlayerSrc)\s*[:=]\s*(["\'])(http.+?)\1', webpage):
- links.append(link)
-
- # Fallback #4, encrypted links (unavailable as at 22.06.2017)
- for _, encrypted_link in re.findall(
- r'encryptedQuality\d{3,4}URL\s*=\s*(["\'])([\da-zA-Z+/=]+)\1', webpage):
- links.append(aes_decrypt_text(encrypted_link, title, 32).decode('utf-8'))
+ def get_format_data(data, f):
+ return traverse_obj(data, lambda _, v: v['format'] == f and url_or_none(v['videoUrl']))
formats = []
- for video_url in set(unescapeHTML(link) for link in links):
- f = {
- 'url': video_url,
- }
+ # Try to extract only the actual master m3u8 first, avoiding the duplicate single resolution "master" m3u8s
+ for hls_url in traverse_obj(get_format_data(definitions, 'hls'), (
+ lambda _, v: not isinstance(v['defaultQuality'], bool), 'videoUrl'), (..., 'videoUrl')):
+ formats.extend(self._extract_m3u8_formats(hls_url, video_id, 'mp4', fatal=False, m3u8_id='hls'))
+
+ for definition in get_format_data(definitions, 'mp4'):
+ f = traverse_obj(definition, {
+ 'url': 'videoUrl',
+ 'filesize': ('videoSize', {int_or_none})
+ })
+ height = int_or_none(definition.get('quality'))
# Video URL's path looks like this:
# /201012/17/505835/720p_1500k_505835/YouPorn%20-%20Sex%20Ed%20Is%20It%20Safe%20To%20Masturbate%20Daily.mp4
# /201012/17/505835/vl_240p_240k_505835/YouPorn%20-%20Sex%20Ed%20Is%20It%20Safe%20To%20Masturbate%20Daily.mp4
# /videos/201703/11/109285532/1080P_4000K_109285532.mp4
# We will benefit from it by extracting some metadata
- mobj = re.search(r'(?P<height>\d{3,4})[pP]_(?P<bitrate>\d+)[kK]_\d+', video_url)
+ mobj = re.search(r'(?P<height>\d{3,4})[pP]_(?P<bitrate>\d+)[kK]_\d+', definition['videoUrl'])
if mobj:
- height = int(mobj.group('height'))
+ if not height:
+ height = int(mobj.group('height'))
bitrate = int(mobj.group('bitrate'))
f.update({
'format_id': '%dp-%dk' % (height, bitrate),
- 'height': height,
'tbr': bitrate,
})
+ f['height'] = height
formats.append(f)
- self._sort_formats(formats)
+
+ webpage = self._download_webpage(
+ 'http://www.youporn.com/watch/%s' % video_id, display_id,
+ headers={'Cookie': 'age_verified=1'})
+
+ title = self._html_search_regex(
+ r'(?s)<div[^>]+class=["\']watchVideoTitle[^>]+>(.+?)</div>',
+ webpage, 'title', default=None) or self._og_search_title(
+ webpage, default=None) or self._html_search_meta(
+ 'title', webpage, fatal=True)
description = self._html_search_regex(
r'(?s)<div[^>]+\bid=["\']description["\'][^>]*>(.+?)</div>',
r'(?s)<div[^>]+class=["\']submitByLink["\'][^>]*>(.+?)</div>',
webpage, 'uploader', fatal=False)
upload_date = unified_strdate(self._html_search_regex(
- [r'UPLOADED:\s*<span>([^<]+)',
+ (r'UPLOADED:\s*<span>([^<]+)',
r'Date\s+[Aa]dded:\s*<span>([^<]+)',
- r'(?s)<div[^>]+class=["\']videoInfo(?:Date|Time)["\'][^>]*>(.+?)</div>'],
+ r'''(?s)<div[^>]+class=["']videoInfo(?:Date|Time)\b[^>]*>(.+?)</div>''',
+ r'(?s)<label\b[^>]*>Uploaded[^<]*</label>\s*<span\b[^>]*>(.+?)</span>'),
webpage, 'upload date', fatal=False))
age_limit = self._rta_search(webpage)
- average_rating = int_or_none(self._search_regex(
- r'<div[^>]+class=["\']videoRatingPercentage["\'][^>]*>(\d+)%</div>',
- webpage, 'average rating', fatal=False))
-
- view_count = str_to_int(self._search_regex(
- r'(?s)<div[^>]+class=(["\']).*?\bvideoInfoViews\b.*?\1[^>]*>.*?(?P<count>[\d,.]+)<',
- webpage, 'view count', fatal=False, group='count'))
+ view_count = None
+ views = self._search_regex(
+ r'(<div[^>]+\bclass=["\']js_videoInfoViews["\']>)', webpage,
+ 'views', default=None)
+ if views:
+ view_count = str_to_int(extract_attributes(views).get('data-value'))
comment_count = str_to_int(self._search_regex(
r'>All [Cc]omments? \(([\d,.]+)\)',
webpage, 'comment count', default=None))
r'(?s)Tags:.*?</div>\s*<div[^>]+class=["\']tagBoxContent["\'][^>]*>(.+?)</div>',
'tags')
- return {
+ data = self._search_json_ld(webpage, video_id, expected_type='VideoObject', fatal=False)
+ data.pop('url', None)
+ return merge_dicts(data, {
'id': video_id,
'display_id': display_id,
'title': title,
'duration': duration,
'uploader': uploader,
'upload_date': upload_date,
- 'average_rating': average_rating,
'view_count': view_count,
'comment_count': comment_count,
'categories': categories,
'tags': tags,
'age_limit': age_limit,
'formats': formats,
- }
+ })
projects / yt-dlp.git / blobdiff