From b540697a8a10bc742d6a241881196429eadd63ca Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Mon, 24 Mar 2014 20:53:03 +0700 Subject: [PATCH] [veoh] Improve extraction, fix youtube extraction (Closes #2616) --- youtube_dl/extractor/veoh.py | 117 +++++++++++++++++++++++++++-------- 1 file changed, 91 insertions(+), 26 deletions(-) diff --git a/youtube_dl/extractor/veoh.py b/youtube_dl/extractor/veoh.py index c90feefd2a..d16993daf0 100644 --- a/youtube_dl/extractor/veoh.py +++ b/youtube_dl/extractor/veoh.py @@ -4,26 +4,99 @@ import re import json from .common import InfoExtractor -from ..utils import compat_urllib_request +from ..utils import ( + compat_urllib_request, + int_or_none, +) class VeohIE(InfoExtractor): - _VALID_URL = r'http://(?:www\.)?veoh\.com/(?:watch|iphone/#_Watch)/v(?P\d*)' + _VALID_URL = r'http://(?:www\.)?veoh\.com/(?:watch|iphone/#_Watch)/(?P(?:v|yapi-)[\da-zA-Z]+)' - _TEST = { - 'url': 'http://www.veoh.com/watch/v56314296nk7Zdmz3', - 'file': '56314296.mp4', - 'md5': '620e68e6a3cff80086df3348426c9ca3', - 'info_dict': { - 'title': 'Straight Backs Are Stronger', - 'uploader': 'LUMOback', - 'description': 'At LUMOback, we believe straight backs are stronger. The LUMOback Posture & Movement Sensor: It gently vibrates when you slouch, inspiring improved posture and mobility. Use the app to track your data and improve your posture over time. ', + _TESTS = [ + { + 'url': 'http://www.veoh.com/watch/v56314296nk7Zdmz3', + 'md5': '620e68e6a3cff80086df3348426c9ca3', + 'info_dict': { + 'id': '56314296', + 'ext': 'mp4', + 'title': 'Straight Backs Are Stronger', + 'uploader': 'LUMOback', + 'description': 'At LUMOback, we believe straight backs are stronger. The LUMOback Posture & Movement Sensor: It gently vibrates when you slouch, inspiring improved posture and mobility. Use the app to track your data and improve your posture over time. ', + }, + }, + { + 'url': 'http://www.veoh.com/watch/v27701988pbTc4wzN?h1=Chile+workers+cover+up+to+avoid+skin+damage', + 'md5': '4a6ff84b87d536a6a71e6aa6c0ad07fa', + 'info_dict': { + 'id': '27701988', + 'ext': 'mp4', + 'title': 'Chile workers cover up to avoid skin damage', + 'description': 'md5:2bd151625a60a32822873efc246ba20d', + 'uploader': 'afp-news', + 'duration': 123, + }, + }, + { + 'url': 'http://www.veoh.com/watch/v69525809F6Nc4frX', + 'md5': '4fde7b9e33577bab2f2f8f260e30e979', + 'note': 'Embedded ooyala video', + 'info_dict': { + 'id': '69525809', + 'ext': 'mp4', + 'title': 'Doctors Alter Plan For Preteen\'s Weight Loss Surgery', + 'description': 'md5:f5a11c51f8fb51d2315bca0937526891', + 'uploader': 'newsy-videos', + }, + }, + ] + + def _extract_formats(self, source): + formats = [] + link = source.get('aowPermalink') + if link: + formats.append({ + 'url': link, + 'ext': 'mp4', + 'format_id': 'aow', + }) + link = source.get('fullPreviewHashLowPath') + if link: + formats.append({ + 'url': link, + 'format_id': 'low', + }) + link = source.get('fullPreviewHashHighPath') + if link: + formats.append({ + 'url': link, + 'format_id': 'high', + }) + return formats + + def _extract_video(self, source): + return { + 'id': source.get('videoId'), + 'title': source.get('title'), + 'description': source.get('description'), + 'thumbnail': source.get('highResImage') or source.get('medResImage'), + 'uploader': source.get('username'), + 'duration': int_or_none(source.get('length')), + 'view_count': int_or_none(source.get('views')), + 'age_limit': 18 if source.get('isMature') == 'true' or source.get('isSexy') == 'true' else 0, + 'formats': self._extract_formats(source), } - } def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') + + if video_id.startswith('v'): + rsp = self._download_xml( + r'http://www.veoh.com/api/findByPermalink?permalink=%s' % video_id, video_id, 'Downloading video XML') + if rsp.get('stat') == 'ok': + return self._extract_video(rsp.find('./videoList/video')) + webpage = self._download_webpage(url, video_id) age_limit = 0 if 'class="adultwarning-container"' in webpage: @@ -33,24 +106,16 @@ class VeohIE(InfoExtractor): request.add_header('Cookie', 'confirmedAdult=true') webpage = self._download_webpage(request, video_id) - m_youtube = re.search(r'http://www\.youtube\.com/v/(.*?)(\&|")', webpage) + m_youtube = re.search(r'http://www\.youtube\.com/v/(.*?)(\&|"|\?)', webpage) if m_youtube is not None: youtube_id = m_youtube.group(1) self.to_screen('%s: detected Youtube video.' % video_id) return self.url_result(youtube_id, 'Youtube') - self.report_extraction(video_id) - info = self._search_regex(r'videoDetailsJSON = \'({.*?})\';', webpage, 'info') - info = json.loads(info) - video_url = info.get('fullPreviewHashHighPath') or info.get('fullPreviewHashLowPath') + info = json.loads( + self._search_regex(r'videoDetailsJSON = \'({.*?})\';', webpage, 'info').replace('\\\'', '\'')) - return { - 'id': info['videoId'], - 'title': info['title'], - 'url': video_url, - 'uploader': info['username'], - 'thumbnail': info.get('highResImage') or info.get('medResImage'), - 'description': info['description'], - 'view_count': info['views'], - 'age_limit': age_limit, - } + video = self._extract_video(info) + video['age_limit'] = age_limit + + return video