X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fsohu.py;h=c04791997f3672cdb643870086c7ed7f52db54c1;hb=5b9aefef773463977504081faa1db88bd0e94300;hp=cd049b6f0dc5ecbaf26e36277702f91df55d96f5;hpb=b5a6d408181c118bf51382f486a2492643ed74ec;p=youtube-dl.git diff --git a/youtube_dl/extractor/sohu.py b/youtube_dl/extractor/sohu.py index cd049b6f0..c04791997 100644 --- a/youtube_dl/extractor/sohu.py +++ b/youtube_dl/extractor/sohu.py @@ -1,91 +1,109 @@ # encoding: utf-8 +from __future__ import unicode_literals import re -import json -import time -import logging -import urllib2 from .common import InfoExtractor -from ..utils import compat_urllib_request, clean_html +from .common import compat_str class SohuIE(InfoExtractor): - _VALID_URL = r'https?://tv\.sohu\.com/\d+?/n(?P\d+)\.shtml.*?' + _VALID_URL = r'https?://(?Pmy\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P\d+)\.shtml.*?' _TEST = { - u'url': u'http://tv.sohu.com/20130724/n382479172.shtml#super', - u'file': u'382479172.flv', - u'md5': u'cc84eed6b6fbf0f2f9a8d3cb9da1939b', - u'info_dict': { - u'title': u'The Illest - Far East Movement Riff Raff', + 'url': 'http://tv.sohu.com/20130724/n382479172.shtml#super', + 'md5': 'bde8d9a6ffd82c63a1eefaef4eeefec7', + 'info_dict': { + 'id': '382479172', + 'ext': 'mp4', + 'title': 'MV：Far East Movement《The Illest》', }, + 'skip': 'Only available from China', } - def _real_extract(self, url): + + def _fetch_data(vid_id, mytv=False): + if mytv: + base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid=' + else: + base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid=' + + return self._download_json( + base_data_url + vid_id, video_id, + 'Downloading JSON data for %s' % vid_id) + mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') + mytv = mobj.group('mytv') is not None + webpage = self._download_webpage(url, video_id) - pattern = r'(.+?)' - compiled = re.compile(pattern, re.DOTALL) - title = self._search_regex(compiled, webpage, u'video title') - title = clean_html(title).split('-')[0].strip() - pattern = re.compile(r'var vid="(\d+)"') - result = re.search(pattern, webpage) - if not result: - logging.info('[Sohu] could not get vid') - return None - vid = result.group(1) - logging.info('vid: %s' % vid) - base_url_1 = 'http://hot.vrs.sohu.com/vrs_flash.action?vid=' - url_1 = base_url_1 + vid - logging.info('json url: %s' % url_1) - webpage = self._download_webpage(url_1, vid) - json_1 = json.loads(webpage) - # get the highest definition video vid and json infomation. - vids = [] - qualities = ('oriVid', 'superVid', 'highVid', 'norVid') - for vid_name in qualities: - vids.append(json_1['data'][vid_name]) - clearest_vid = 0 - for i, v in enumerate(vids): - if v != 0: - clearest_vid = v - logging.info('quality definition: %s' % qualities[i][:-3]) - break - if not clearest_vid: - logging.warning('could not find valid clearest_vid') - return None - if vid != clearest_vid: - url_1 = '%s%d' % (base_url_1, clearest_vid) - logging.info('highest definition json url: %s' % url_1) - json_1 = json.loads(urllib2.urlopen(url_1).read()) - allot = json_1['allot'] - prot = json_1['prot'] - clipsURL = json_1['data']['clipsURL'] - su = json_1['data']['su'] - num_of_parts = json_1['data']['totalBlocks'] - logging.info('Total parts: %d' % num_of_parts) - base_url_3 = 'http://allot/?prot=prot&file=clipsURL[i]&new=su[i]' - files_info = [] - for i in range(num_of_parts): - middle_url = 'http://%s/?prot=%s&file=%s&new=%s' % (allot, prot, clipsURL[i], su[i]) - logging.info('middle url part %d: %s' % (i, middle_url)) - middle_info = urllib2.urlopen(middle_url).read().split('|') - middle_part_1 = middle_info[0] - download_url = '%s%s?key=%s' % (middle_info[0], su[i], middle_info[3]) + raw_title = self._html_search_regex( + r'(?s)(.+?)', + webpage, 'video title') + title = raw_title.partition('-')[0].strip() - info = { - 'id': '%s_part%02d' % (video_id, i + 1), + vid = self._html_search_regex( + r'var vid ?= ?["\'](\d+)["\']', + webpage, 'video path') + vid_data = _fetch_data(vid, mytv) + + formats_json = {} + for format_id in ('nor', 'high', 'super', 'ori', 'h2644k', 'h2654k'): + vid_id = vid_data['data'].get('%sVid' % format_id) + if not vid_id: + continue + vid_id = compat_str(vid_id) + formats_json[format_id] = vid_data if vid == vid_id else _fetch_data(vid_id, mytv) + + part_count = vid_data['data']['totalBlocks'] + + playlist = [] + for i in range(part_count): + formats = [] + for format_id, format_data in formats_json.items(): + allot = format_data['allot'] + prot = format_data['prot'] + + data = format_data['data'] + clips_url = data['clipsURL'] + su = data['su'] + + part_str = self._download_webpage( + 'http://%s/?prot=%s&file=%s&new=%s' % + (allot, prot, clips_url[i], su[i]), + video_id, + 'Downloading %s video URL part %d of %d' + % (format_id, i + 1, part_count)) + + part_info = part_str.split('|') + video_url = '%s%s?key=%s' % (part_info[0], su[i], part_info[3]) + + formats.append({ + 'url': video_url, + 'format_id': format_id, + 'filesize': data['clipsBytes'][i], + 'width': data['width'], + 'height': data['height'], + 'fps': data['fps'], + }) + self._sort_formats(formats) + + playlist.append({ + 'id': '%s_part%d' % (video_id, i + 1), 'title': title, - 'url': download_url, - 'ext': 'mp4', - } - files_info.append(info) - time.sleep(1) - if num_of_parts == 1: - info = files_info[0] + 'duration': vid_data['data']['clipsDuration'][i], + 'formats': formats, + }) + + if len(playlist) == 1: + info = playlist[0] info['id'] = video_id - return info - return files_info + else: + info = { + '_type': 'playlist', + 'entries': playlist, + 'id': video_id, + } + + return info