From 7f09a662a0db2ad0a338645e4f2dbd43056a6fa6 Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Tue, 24 Feb 2015 23:55:57 +0800 Subject: [PATCH 1/3] [Letv] Add new extractor. Single video only --- youtube_dl/extractor/__init__.py | 1 + youtube_dl/extractor/letv.py | 120 +++++++++++++++++++++++++++++++ 2 files changed, 121 insertions(+) create mode 100644 youtube_dl/extractor/letv.py 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\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, + } From 570311610e33eb67d5a65a86705d54daaee17361 Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Wed, 25 Feb 2015 00:45:11 +0800 Subject: [PATCH 2/3] [Letv] Add playlist support --- youtube_dl/extractor/__init__.py | 6 +++- youtube_dl/extractor/letv.py | 54 +++++++++++++++++++++++++++++++- 2 files changed, 58 insertions(+), 2 deletions(-) diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 829ab18a9..7b7d41adf 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -237,7 +237,11 @@ from .krasview import KrasViewIE from .ku6 import Ku6IE from .la7 import LA7IE from .laola1tv import Laola1TvIE -from .letv import LetvIE +from .letv import ( + LetvIE, + LetvTvIE, + LetvPlaylistIE +) 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 index 3eb20678e..d7d315634 100644 --- a/youtube_dl/extractor/letv.py +++ b/youtube_dl/extractor/letv.py @@ -2,6 +2,7 @@ from __future__ import unicode_literals import os.path +import re import time import datetime @@ -11,7 +12,7 @@ from ..utils import (ExtractorError, parse_iso8601) class LetvIE(InfoExtractor): - _VALID_URL = r'http://www.letv.com/ptv/vplay/(?P\d+).html' + _VALID_URL = r'http://www\.letv\.com/ptv/vplay/(?P\d+).html' _TESTS = [{ 'url': 'http://www.letv.com/ptv/vplay/22005890.html', @@ -118,3 +119,54 @@ class LetvIE(InfoExtractor): 'thumbnail': playurl['pic'], 'timestamp': publish_time, } + + +class LetvTvIE(InfoExtractor): + _VALID_URL = r'http://www.letv.com/tv/(?P\d+).html' + _TESTS = [{ + 'url': 'http://www.letv.com/tv/46177.html', + 'info_dict': { + 'id': '46177', + 'title': '美人天下', + 'description': 'md5:395666ff41b44080396e59570dbac01c' + }, + 'playlist_count': 35 + }] + + 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] + + title = self._html_search_meta('keywords', page, fatal=False).split('，')[0] + description = self._html_search_meta('description', page, fatal=False) + + return self.playlist_result(entries, playlist_id, playlist_title=title, + playlist_description=description) + + +class LetvPlaylistIE(LetvTvIE): + _VALID_URL = r'http://tv.letv.com/[a-z]+/(?P[a-z]+)/index.s?html' + _TESTS = [{ + 'url': 'http://tv.letv.com/izt/wuzetian/index.html', + 'info_dict': { + 'id': 'wuzetian', + 'title': '武媚娘传奇', + 'description': 'md5:e12499475ab3d50219e5bba00b3cb248' + }, + 'playlist_count': 96 + }, { + 'url': 'http://tv.letv.com/pzt/lswjzzjc/index.shtml', + 'info_dict': { + 'id': 'lswjzzjc', + # 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 + }] From 677063594e28d11ada9d2bfd3a30397d4de24360 Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Wed, 25 Feb 2015 02:10:55 +0800 Subject: [PATCH 3/3] [Letv] Update testcases --- youtube_dl/extractor/letv.py | 26 +++++++++++++++++++------- 1 file changed, 19 insertions(+), 7 deletions(-) diff --git a/youtube_dl/extractor/letv.py b/youtube_dl/extractor/letv.py index d7d315634..d5839263c 100644 --- a/youtube_dl/extractor/letv.py +++ b/youtube_dl/extractor/letv.py @@ -23,14 +23,22 @@ class LetvIE(InfoExtractor): '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', + 'description': 'md5:a9cb175fd753e2962176b7beca21a47c', } }, { - 'url': 'http://www.letv.com/ptv/vplay/1118082.html', + 'url': 'http://www.letv.com/ptv/vplay/1415246.html', 'info_dict': { - 'id': '1118082', + 'id': '1415246', 'ext': 'mp4', - } + 'title': '美人天下01', + 'description': 'md5:f88573d9d7225ada1359eaf0dbf8bcda', + }, + 'expected_warnings': [ + 'publish time' + ] }] + # http://www.letv.com/ptv/vplay/1118082.html + # This video is available only in Mainland China @staticmethod def urshift(val, n): @@ -111,12 +119,14 @@ class LetvIE(InfoExtractor): publish_time = parse_iso8601(self._html_search_regex( r'发布时间 ([^<>]+) ', page, 'publish time', fatal=False), delimiter=' ', timezone=datetime.timedelta(hours=8)) + description = self._html_search_meta('description', page, fatal=False) return { 'id': media_id, 'formats': urls, 'title': playurl['title'], 'thumbnail': playurl['pic'], + 'description': description, 'timestamp': publish_time, } @@ -142,7 +152,8 @@ class LetvTvIE(InfoExtractor): entries = [self.url_result(media_url, ie='Letv') for media_url in media_urls] - title = self._html_search_meta('keywords', page, fatal=False).split('，')[0] + title = self._html_search_meta('keywords', page, + fatal=False).split('，')[0] description = self._html_search_meta('description', page, fatal=False) return self.playlist_result(entries, playlist_id, playlist_title=title, @@ -158,13 +169,14 @@ class LetvPlaylistIE(LetvTvIE): 'title': '武媚娘传奇', 'description': 'md5:e12499475ab3d50219e5bba00b3cb248' }, - 'playlist_count': 96 + # This playlist contains some extra videos other than the drama itself + 'playlist_mincount': 96 }, { 'url': 'http://tv.letv.com/pzt/lswjzzjc/index.shtml', 'info_dict': { 'id': 'lswjzzjc', - # should be "劲舞青春", but I can't find a simple way to determine - # the playlist title + # 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' },