X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/15b1d10671b48df598afd70e17ba21e9e64ac766..d1f304bf54caa539923ca447251db0fbda7093c8:/youtube_dl/extractor/teamcoco.py diff --git a/youtube_dl/extractor/teamcoco.py b/youtube_dl/extractor/teamcoco.py index 5fa67eb..7534639 100644 --- a/youtube_dl/extractor/teamcoco.py +++ b/youtube_dl/extractor/teamcoco.py @@ -1,76 +1,160 @@ +# coding: utf-8 from __future__ import unicode_literals +import base64 +import binascii import re +import json from .common import InfoExtractor +from ..utils import ( + ExtractorError, + qualities, + determine_ext, +) +from ..compat import compat_ord class TeamcocoIE(InfoExtractor): - _VALID_URL = r'http://teamcoco\.com/video/(?P[0-9]+)?/?(?P.*)' + _VALID_URL = r'https?://teamcoco\.com/video/(?P[0-9]+)?/?(?P.*)' _TESTS = [ { 'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant', - 'file': '80187.mp4', 'md5': '3f7746aa0dc86de18df7539903d399ea', 'info_dict': { + 'id': '80187', + 'ext': 'mp4', 'title': 'Conan Becomes A Mary Kay Beauty Consultant', - 'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.' + 'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.', + 'duration': 504, + 'age_limit': 0, } }, { 'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush', - 'file': '19705.mp4', 'md5': 'cde9ba0fa3506f5f017ce11ead928f9a', 'info_dict': { - "description": "Louis C.K. got starstruck by George W. Bush, so what? Part one.", - "title": "Louis C.K. Interview Pt. 1 11/3/11" + 'id': '19705', + 'ext': 'mp4', + 'description': 'Louis C.K. got starstruck by George W. Bush, so what? Part one.', + 'title': 'Louis C.K. Interview Pt. 1 11/3/11', + 'duration': 288, + 'age_limit': 0, + } + }, { + 'url': 'http://teamcoco.com/video/timothy-olyphant-drinking-whiskey', + 'info_dict': { + 'id': '88748', + 'ext': 'mp4', + 'title': 'Timothy Olyphant Raises A Toast To “Justified”', + 'description': 'md5:15501f23f020e793aeca761205e42c24', + }, + 'params': { + 'skip_download': True, # m3u8 downloads + } + }, { + 'url': 'http://teamcoco.com/video/full-episode-mon-6-1-joel-mchale-jake-tapper-and-musical-guest-courtney-barnett?playlist=x;eyJ0eXBlIjoidGFnIiwiaWQiOjl9', + 'info_dict': { + 'id': '89341', + 'ext': 'mp4', + 'title': 'Full Episode - Mon. 6/1 - Joel McHale, Jake Tapper, And Musical Guest Courtney Barnett', + 'description': 'Guests: Joel McHale, Jake Tapper, And Musical Guest Courtney Barnett', + }, + 'params': { + 'skip_download': True, # m3u8 downloads } } ] + _VIDEO_ID_REGEXES = ( + r'"eVar42"\s*:\s*(\d+)', + r'Ginger\.TeamCoco\.openInApp\("video",\s*"([^"]+)"', + r'"id_not"\s*:\s*(\d+)' + ) def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) display_id = mobj.group('display_id') - webpage = self._download_webpage(url, display_id) + webpage, urlh = self._download_webpage_handle(url, display_id) + if 'src=expired' in urlh.geturl(): + raise ExtractorError('This video is expired.', expected=True) - video_id = mobj.group("video_id") + video_id = mobj.group('video_id') if not video_id: video_id = self._html_search_regex( - r'data-node-id="(\d+?)"', - webpage, 'video id') + self._VIDEO_ID_REGEXES, webpage, 'video id') + + data = None + + preload_codes = self._html_search_regex( + r'(function.+)setTimeout\(function\(\)\{playlist', + webpage, 'preload codes') + base64_fragments = re.findall(r'"([a-zA-Z0-9+/=]+)"', preload_codes) + base64_fragments.remove('init') + + def _check_sequence(cur_fragments): + if not cur_fragments: + return + for i in range(len(cur_fragments)): + cur_sequence = (''.join(cur_fragments[i:] + cur_fragments[:i])).encode('ascii') + try: + raw_data = base64.b64decode(cur_sequence) + if compat_ord(raw_data[0]) == compat_ord('{'): + return json.loads(raw_data.decode('utf-8')) + except (TypeError, binascii.Error, UnicodeDecodeError, ValueError): + continue - data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id - data = self._download_xml( - data_url, display_id, 'Downloading data webpage') + def _check_data(): + for i in range(len(base64_fragments) + 1): + for j in range(i, len(base64_fragments) + 1): + data = _check_sequence(base64_fragments[:i] + base64_fragments[j:]) + if data: + return data + + self.to_screen('Try to compute possible data sequence. This may take some time.') + data = _check_data() + + if not data: + raise ExtractorError( + 'Preload information could not be extracted', expected=True) - qualities = ['500k', '480p', '1000k', '720p', '1080p'] formats = [] - for filed in data.findall('files/file'): - if filed.attrib.get('playmode') == 'all': - # it just duplicates one of the entries - break - file_url = filed.text - m_format = re.search(r'(\d+(k|p))\.mp4', file_url) - if m_format is not None: - format_id = m_format.group(1) + get_quality = qualities(['500k', '480p', '1000k', '720p', '1080p']) + for filed in data['files']: + if determine_ext(filed['url']) == 'm3u8': + # compat_urllib_parse.urljoin does not work here + if filed['url'].startswith('/'): + m3u8_url = 'http://ht.cdn.turner.com/tbs/big/teamcoco' + filed['url'] + else: + m3u8_url = filed['url'] + m3u8_formats = self._extract_m3u8_formats( + m3u8_url, video_id, ext='mp4') + for m3u8_format in m3u8_formats: + if m3u8_format not in formats: + formats.append(m3u8_format) + elif determine_ext(filed['url']) == 'f4m': + # TODO Correct f4m extraction + continue else: - format_id = filed.attrib['bitrate'] - tbr = ( - int(filed.attrib['bitrate']) - if filed.attrib['bitrate'].isdigit() - else None) - - try: - quality = qualities.index(format_id) - except ValueError: - quality = -1 - formats.append({ - 'url': file_url, - 'ext': 'mp4', - 'tbr': tbr, - 'format_id': format_id, - 'quality': quality, - }) + if filed['url'].startswith('/mp4:protected/'): + # TODO Correct extraction for these files + continue + m_format = re.search(r'(\d+(k|p))\.mp4', filed['url']) + if m_format is not None: + format_id = m_format.group(1) + else: + format_id = filed['bitrate'] + tbr = ( + int(filed['bitrate']) + if filed['bitrate'].isdigit() + else None) + + formats.append({ + 'url': filed['url'], + 'ext': 'mp4', + 'tbr': tbr, + 'format_id': format_id, + 'quality': get_quality(format_id), + }) self._sort_formats(formats) @@ -78,7 +162,9 @@ class TeamcocoIE(InfoExtractor): 'id': video_id, 'display_id': display_id, 'formats': formats, - 'title': self._og_search_title(webpage), - 'thumbnail': self._og_search_thumbnail(webpage), - 'description': self._og_search_description(webpage), + 'title': data['title'], + 'thumbnail': data.get('thumb', {}).get('href'), + 'description': data.get('teaser'), + 'duration': data.get('duration'), + 'age_limit': self._family_friendly_search(webpage), }