X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fletv.py;h=ce25bb4dbf12b1959f36ac62b20570c7a1bd82c5;hb=2eb25b256bf38db9f35696822cee399099eb4777;hp=be648000ea1bfec896c7228159170e16aa8338ee;hpb=9c117d345ff05da2d4b107881cc387657208b76e;p=youtube-dl diff --git a/youtube_dl/extractor/letv.py b/youtube_dl/extractor/letv.py index be648000e..ce25bb4db 100644 --- a/youtube_dl/extractor/letv.py +++ b/youtube_dl/extractor/letv.py @@ -1,31 +1,39 @@ # coding: utf-8 from __future__ import unicode_literals +import base64 import datetime +import hashlib import re import time from .common import InfoExtractor from ..compat import ( - compat_urllib_parse, compat_ord, + compat_str, + compat_urllib_parse, ) from ..utils import ( determine_ext, + encode_data_uri, ExtractorError, + int_or_none, + orderedSet, parse_iso8601, sanitized_Request, - int_or_none, - encode_data_uri, + str_or_none, + url_basename, ) class LetvIE(InfoExtractor): IE_DESC = '乐视网' - _VALID_URL = r'http://www\.letv\.com/ptv/vplay/(?P\d+).html' + _VALID_URL = r'http://www\.le\.com/ptv/vplay/(?P\d+)\.html' + + _URL_TEMPLATE = 'http://www.le.com/ptv/vplay/%s.html' _TESTS = [{ - 'url': 'http://www.letv.com/ptv/vplay/22005890.html', + 'url': 'http://www.le.com/ptv/vplay/22005890.html', 'md5': 'edadcfe5406976f42f9f266057ee5e40', 'info_dict': { 'id': '22005890', @@ -37,7 +45,7 @@ class LetvIE(InfoExtractor): 'hls_prefer_native': True, }, }, { - 'url': 'http://www.letv.com/ptv/vplay/1415246.html', + 'url': 'http://www.le.com/ptv/vplay/1415246.html', 'info_dict': { 'id': '1415246', 'ext': 'mp4', @@ -49,7 +57,7 @@ class LetvIE(InfoExtractor): }, }, { 'note': 'This video is available only in Mainland China, thus a proxy is needed', - 'url': 'http://www.letv.com/ptv/vplay/1118082.html', + 'url': 'http://www.le.com/ptv/vplay/1118082.html', 'md5': '2424c74948a62e5f31988438979c5ad1', 'info_dict': { 'id': '1118082', @@ -89,17 +97,16 @@ class LetvIE(InfoExtractor): return encrypted_data encrypted_data = encrypted_data[5:] - _loc4_ = bytearray() - while encrypted_data: - b = compat_ord(encrypted_data[0]) - _loc4_.extend([b // 16, b & 0x0f]) - encrypted_data = encrypted_data[1:] + _loc4_ = bytearray(2 * len(encrypted_data)) + for idx, val in enumerate(encrypted_data): + b = compat_ord(val) + _loc4_[2 * idx] = b // 16 + _loc4_[2 * idx + 1] = b % 16 idx = len(_loc4_) - 11 _loc4_ = _loc4_[idx:] + _loc4_[:idx] - _loc7_ = bytearray() - while _loc4_: - _loc7_.append(_loc4_[0] * 16 + _loc4_[1]) - _loc4_ = _loc4_[2:] + _loc7_ = bytearray(len(encrypted_data)) + for i in range(len(encrypted_data)): + _loc7_[i] = _loc4_[2 * i] * 16 + _loc4_[2 * i + 1] return bytes(_loc7_) @@ -112,10 +119,10 @@ class LetvIE(InfoExtractor): 'splatid': 101, 'format': 1, 'tkey': self.calc_time_key(int(time.time())), - 'domain': 'www.letv.com' + 'domain': 'www.le.com' } play_json_req = sanitized_Request( - 'http://api.letv.com/mms/out/video/playJson?' + compat_urllib_parse.urlencode(params) + 'http://api.le.com/mms/out/video/playJson?' + compat_urllib_parse.urlencode(params) ) cn_verification_proxy = self._downloader.params.get('cn_verification_proxy') if cn_verification_proxy: @@ -188,26 +195,57 @@ class LetvIE(InfoExtractor): } -class LetvTvIE(InfoExtractor): - _VALID_URL = r'http://www.letv.com/tv/(?P\d+).html' +class LetvPlaylistIE(InfoExtractor): + _VALID_URL = r'http://[a-z]+\.le\.com/[a-z]+/(?P[a-z0-9_]+)' + _TESTS = [{ - 'url': 'http://www.letv.com/tv/46177.html', + 'url': 'http://www.le.com/tv/46177.html', 'info_dict': { 'id': '46177', 'title': '美人天下', 'description': 'md5:395666ff41b44080396e59570dbac01c' }, 'playlist_count': 35 + }, { + 'url': 'http://tv.le.com/izt/wuzetian/index.html', + 'info_dict': { + 'id': 'wuzetian', + 'title': '武媚娘传奇', + 'description': 'md5:e12499475ab3d50219e5bba00b3cb248' + }, + # This playlist contains some extra videos other than the drama itself + 'playlist_mincount': 96 + }, { + 'url': 'http://tv.le.com/pzt/lswjzzjc/index.shtml', + 'info_dict': { + 'id': 'lswjzzjc', + # The title should be "劲舞青春", but I can't find a simple way to + # determine the playlist title + 'title': '乐视午间自制剧场', + 'description': 'md5:b1eef244f45589a7b5b1af9ff25a4489' + }, + 'playlist_mincount': 7 + }, { + 'url': 'http://www.le.com/comic/92063.html', + 'only_matching': True, + }, { + 'url': 'http://list.le.com/listn/c1009_sc532002_d2_p1_o1.html', + 'only_matching': True, }] + @classmethod + def suitable(cls, url): + return False if LetvIE.suitable(url) else super(LetvPlaylistIE, cls).suitable(url) + def _real_extract(self, url): playlist_id = self._match_id(url) page = self._download_webpage(url, playlist_id) - media_urls = list(set(re.findall( - r'http://www.letv.com/ptv/vplay/\d+.html', page))) - entries = [self.url_result(media_url, ie='Letv') - for media_url in media_urls] + # Currently old domain names are still used in playlists + media_ids = orderedSet(re.findall( + r']+href="http://www\.letv\.com/ptv/vplay/(\d+)\.html', page)) + entries = [self.url_result(LetvIE._URL_TEMPLATE % media_id, ie='Letv') + for media_id in media_ids] title = self._html_search_meta('keywords', page, fatal=False).split('，')[0] @@ -217,25 +255,111 @@ class LetvTvIE(InfoExtractor): playlist_description=description) -class LetvPlaylistIE(LetvTvIE): - _VALID_URL = r'http://tv.letv.com/[a-z]+/(?P[a-z]+)/index.s?html' +class LetvCloudIE(InfoExtractor): + IE_DESC = '乐视云' + _VALID_URL = r'https?://yuntv\.letv\.com/bcloud.html\?.+' + _TESTS = [{ - 'url': 'http://tv.letv.com/izt/wuzetian/index.html', + 'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=467623dedf', + 'md5': '26450599afd64c513bc77030ad15db44', 'info_dict': { - 'id': 'wuzetian', - 'title': '武媚娘传奇', - 'description': 'md5:e12499475ab3d50219e5bba00b3cb248' + 'id': 'p7jnfw5hw9_467623dedf', + 'ext': 'mp4', + 'title': 'Video p7jnfw5hw9_467623dedf', }, - # This playlist contains some extra videos other than the drama itself - 'playlist_mincount': 96 }, { - 'url': 'http://tv.letv.com/pzt/lswjzzjc/index.shtml', + 'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=ec93197892&pu=2c7cd40209&auto_play=1&gpcflag=1&width=640&height=360', + 'md5': 'e03d9cc8d9c13191e1caf277e42dbd31', 'info_dict': { - 'id': 'lswjzzjc', - # The title should be "劲舞青春", but I can't find a simple way to - # determine the playlist title - 'title': '乐视午间自制剧场', - 'description': 'md5:b1eef244f45589a7b5b1af9ff25a4489' + 'id': 'p7jnfw5hw9_ec93197892', + 'ext': 'mp4', + 'title': 'Video p7jnfw5hw9_ec93197892', + }, + }, { + 'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=187060b6fd', + 'md5': 'cb988699a776b22d4a41b9d43acfb3ac', + 'info_dict': { + 'id': 'p7jnfw5hw9_187060b6fd', + 'ext': 'mp4', + 'title': 'Video p7jnfw5hw9_187060b6fd', }, - 'playlist_mincount': 7 }] + + @staticmethod + def sign_data(obj): + if obj['cf'] == 'flash': + salt = '2f9d6924b33a165a6d8b5d3d42f4f987' + items = ['cf', 'format', 'ran', 'uu', 'ver', 'vu'] + elif obj['cf'] == 'html5': + salt = 'fbeh5player12c43eccf2bec3300344' + items = ['cf', 'ran', 'uu', 'bver', 'vu'] + input_data = ''.join([item + obj[item] for item in items]) + salt + obj['sign'] = hashlib.md5(input_data.encode('utf-8')).hexdigest() + + def _get_formats(self, cf, uu, vu, media_id): + def get_play_json(cf, timestamp): + data = { + 'cf': cf, + 'ver': '2.2', + 'bver': 'firefox44.0', + 'format': 'json', + 'uu': uu, + 'vu': vu, + 'ran': compat_str(timestamp), + } + self.sign_data(data) + return self._download_json( + 'http://api.letvcloud.com/gpc.php?' + compat_urllib_parse.urlencode(data), + media_id, 'Downloading playJson data for type %s' % cf) + + play_json = get_play_json(cf, time.time()) + # The server time may be different from local time + if play_json.get('code') == 10071: + play_json = get_play_json(cf, play_json['timestamp']) + + if not play_json.get('data'): + if play_json.get('message'): + raise ExtractorError('Letv cloud said: %s' % play_json['message'], expected=True) + elif play_json.get('code'): + raise ExtractorError('Letv cloud returned error %d' % play_json['code'], expected=True) + else: + raise ExtractorError('Letv cloud returned an unknwon error') + + def b64decode(s): + return base64.b64decode(s.encode('utf-8')).decode('utf-8') + + formats = [] + for media in play_json['data']['video_info']['media'].values(): + play_url = media['play_url'] + url = b64decode(play_url['main_url']) + decoded_url = b64decode(url_basename(url)) + formats.append({ + 'url': url, + 'ext': determine_ext(decoded_url), + 'format_id': int_or_none(play_url.get('vtype')), + 'format_note': str_or_none(play_url.get('definition')), + 'width': int_or_none(play_url.get('vwidth')), + 'height': int_or_none(play_url.get('vheight')), + }) + + return formats + + def _real_extract(self, url): + uu_mobj = re.search('uu=([\w]+)', url) + vu_mobj = re.search('vu=([\w]+)', url) + + if not uu_mobj or not vu_mobj: + raise ExtractorError('Invalid URL: %s' % url, expected=True) + + uu = uu_mobj.group(1) + vu = vu_mobj.group(1) + media_id = uu + '_' + vu + + formats = self._get_formats('flash', uu, vu, media_id) + self._get_formats('html5', uu, vu, media_id) + self._sort_formats(formats) + + return { + 'id': media_id, + 'title': 'Video %s' % media_id, + 'formats': formats, + }