From d4893e764bbda206b6194884e67b0acaf231d0d6 Mon Sep 17 00:00:00 2001 From: =?utf8?q?Sergey=20M=E2=80=A4?= Date: Thu, 22 Jun 2017 00:40:15 +0700 Subject: [PATCH] [youporn] Improve formats extraction --- youtube_dl/extractor/youporn.py | 32 ++++++++++++++++++++++++-------- 1 file changed, 24 insertions(+), 8 deletions(-) diff --git a/youtube_dl/extractor/youporn.py b/youtube_dl/extractor/youporn.py index 7bc2cefc7..64e3c3c94 100644 --- a/youtube_dl/extractor/youporn.py +++ b/youtube_dl/extractor/youporn.py @@ -3,6 +3,7 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import compat_str from ..utils import ( int_or_none, sanitized_Request, @@ -68,7 +69,7 @@ class YouPornIE(InfoExtractor): webpage = self._download_webpage(request, display_id) title = self._search_regex( - [r'(?:video_titles|videoTitle|title)\s*[:=]\s*(["\'])(?P(?:(?!\1).)+)\1', + [r'(?:video_titles|videoTitle)\s*[:=]\s*(["\'])(?P<title>(?:(?!\1).)+)\1', r'<h1[^>]+class=["\']heading\d?["\'][^>]*>(?P<title>[^<]+)<'], webpage, 'title', group='title', default=None) or self._og_search_title( @@ -77,22 +78,37 @@ class YouPornIE(InfoExtractor): links = [] + # Main source + definitions = self._parse_json( + self._search_regex( + r'mediaDefinition\s*=\s*(\[.+?\]);', webpage, + 'media definitions', default='[]'), + video_id, fatal=False) + if definitions: + for definition in definitions: + if not isinstance(definition, dict): + continue + video_url = definition.get('videoUrl') + if isinstance(video_url, compat_str) and 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[^>]+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 #1 + # Fallback #3 (unavailable as at 22.06.2017) for _, link in re.findall( - r'(?:videoUrl|videoSrc|videoIpadUrl|html5PlayerSrc)\s*[:=]\s*(["\'])(http.+?)\1', webpage): - links.append(link) - - # Fallback #2, this also contains extra low quality 180p format - for _, link in re.findall(r'<a[^>]+href=(["\'])(http.+?)\1[^>]+title=["\']Download [Vv]ideo', webpage): + r'(?:videoSrc|videoIpadUrl|html5PlayerSrc)\s*[:=]\s*(["\'])(http.+?)\1', webpage): links.append(link) - # Fallback #3, encrypted links + # 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')) -- 2.22.2