diff options
author | Yen Chi Hsuan <[email protected]> | 2015-02-24 23:55:57 +0800 |
---|---|---|
committer | Yen Chi Hsuan <[email protected]> | 2015-02-24 23:58:21 +0800 |
commit | 7f09a662a0db2ad0a338645e4f2dbd43056a6fa6 (patch) | |
tree | 6b5352f85a972f9cd7b09cbd2b163ba4c99e0990 | |
parent | 4f3b21e1c738a7dacd514eb59242da43e81b5ae1 (diff) | |
download | youtube-dl-7f09a662a0db2ad0a338645e4f2dbd43056a6fa6.tar.gz youtube-dl-7f09a662a0db2ad0a338645e4f2dbd43056a6fa6.zip |
[Letv] Add new extractor. Single video only
-rw-r--r-- | youtube_dl/extractor/__init__.py | 1 | ||||
-rw-r--r-- | youtube_dl/extractor/letv.py | 120 |
2 files changed, 121 insertions, 0 deletions
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 40fc92cf7..829ab18a9 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -237,6 +237,7 @@ from .krasview import KrasViewIE from .ku6 import Ku6IE from .la7 import LA7IE from .laola1tv import Laola1TvIE +from .letv import LetvIE from .lifenews import LifeNewsIE from .liveleak import LiveLeakIE from .livestream import ( diff --git a/youtube_dl/extractor/letv.py b/youtube_dl/extractor/letv.py new file mode 100644 index 000000000..3eb20678e --- /dev/null +++ b/youtube_dl/extractor/letv.py @@ -0,0 +1,120 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import os.path +import time +import datetime + +from .common import InfoExtractor +from ..compat import (compat_urlparse, compat_urllib_parse) +from ..utils import (ExtractorError, parse_iso8601) + + +class LetvIE(InfoExtractor): + _VALID_URL = r'http://www.letv.com/ptv/vplay/(?P<id>\d+).html' + + _TESTS = [{ + 'url': 'http://www.letv.com/ptv/vplay/22005890.html', + 'md5': 'cab23bd68d5a8db9be31c9a222c1e8df', + 'info_dict': { + 'id': '22005890', + 'ext': 'mp4', + 'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家', + 'timestamp': 1424747397, + 'upload_date': '20150224', + } + }, { + 'url': 'http://www.letv.com/ptv/vplay/1118082.html', + 'info_dict': { + 'id': '1118082', + 'ext': 'mp4', + } + }] + + @staticmethod + def urshift(val, n): + return val >> n if val >= 0 else (val + 0x100000000) >> n + + # ror() and calcTimeKey() are reversed from a embedded swf file in KLetvPlayer.swf + def ror(self, param1, param2): + _loc3_ = 0 + while _loc3_ < param2: + param1 = self.urshift(param1, 1) + ((param1 & 1) << 31) + _loc3_ += 1 + return param1 + + def calcTimeKey(self, param1): + _loc2_ = 773625421 + _loc3_ = self.ror(param1, _loc2_ % 13) + _loc3_ = _loc3_ ^ _loc2_ + _loc3_ = self.ror(_loc3_, _loc2_ % 17) + return _loc3_ + + def _real_extract(self, url): + media_id = self._match_id(url) + page = self._download_webpage(url, media_id) + params = { + 'id': media_id, + 'platid': 1, + 'splatid': 101, + 'format': 1, + 'tkey': self.calcTimeKey(int(time.time())), + 'domain': 'www.letv.com' + } + play_json = self._download_json( + 'http://api.letv.com/mms/out/video/playJson?' + compat_urllib_parse.urlencode(params), + media_id, 'playJson data') + + # Check for errors + playstatus = play_json['playstatus'] + if playstatus['status'] == 0: + flag = playstatus['flag'] + if flag == 1: + msg = 'Country %s auth error' % playstatus['country'] + else: + msg = 'Generic error. flag = %d' % flag + raise ExtractorError(msg, expected=True) + + playurl = play_json['playurl'] + + formats = ['350', '1000', '1300', '720p', '1080p'] + dispatch = playurl['dispatch'] + + urls = [] + for format_id in formats: + if format_id in dispatch: + media_url = playurl['domain'][0] + dispatch[format_id][0] + + # Mimic what flvxz.com do + url_parts = list(compat_urlparse.urlparse(media_url)) + qs = dict(compat_urlparse.parse_qs(url_parts[4])) + qs.update({ + 'platid': '14', + 'splatid': '1401', + 'tss': 'no', + 'retry': 1 + }) + url_parts[4] = compat_urllib_parse.urlencode(qs) + media_url = compat_urlparse.urlunparse(url_parts) + + url_info_dict = { + 'url': media_url, + 'ext': os.path.splitext(dispatch[format_id][1])[1][1:] + } + + if format_id[-1:] == 'p': + url_info_dict['height'] = format_id[:-1] + + urls.append(url_info_dict) + + publish_time = parse_iso8601(self._html_search_regex( + r'发布时间 ([^<>]+) ', page, 'publish time', fatal=False), + delimiter=' ', timezone=datetime.timedelta(hours=8)) + + return { + 'id': media_id, + 'formats': urls, + 'title': playurl['title'], + 'thumbnail': playurl['pic'], + 'timestamp': publish_time, + } |