diff options
| -rw-r--r-- | youtube_dl/extractor/__init__.py | 1 | ||||
| -rw-r--r-- | youtube_dl/extractor/gameone.py | 89 | ||||
| -rw-r--r-- | youtube_dl/utils.py | 6 |
3 files changed, 93 insertions, 3 deletions
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 5a273fc0f..fcc7d0b58 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -105,6 +105,7 @@ from .freesound import FreesoundIE from .freespeech import FreespeechIE from .funnyordie import FunnyOrDieIE from .gamekings import GamekingsIE +from .gameone import GameOneIE from .gamespot import GameSpotIE from .gametrailers import GametrailersIE from .gdcvault import GDCVaultIE diff --git a/youtube_dl/extractor/gameone.py b/youtube_dl/extractor/gameone.py new file mode 100644 index 000000000..2544ea521 --- /dev/null +++ b/youtube_dl/extractor/gameone.py @@ -0,0 +1,89 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + xpath_with_ns, + parse_iso8601 +) + +NAMESPACE_MAP = { + 'media': 'http://search.yahoo.com/mrss/', +} + +# URL prefix to download the mp4 files directly instead of streaming via rtmp +# Credits go to XBox-Maniac +# http://board.jdownloader.org/showpost.php?p=185835&postcount=31 +RAW_MP4_URL = 'http://cdn.riptide-mtvn.com/' + + +class GameOneIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?gameone\.de/tv/(?P<id>\d+)' + _TEST = { + 'url': 'http://www.gameone.de/tv/288', + 'md5': '136656b7fb4c9cb4a8e2d500651c499b', + 'info_dict': { + 'id': '288', + 'ext': 'mp4', + 'title': 'Game One - Folge 288', + 'duration': 1238, + 'thumbnail': 'http://s3.gameone.de/gameone/assets/video_metas/teaser_images/000/643/636/big/640x360.jpg', + 'description': 'FIFA-Pressepokal 2014, Star Citizen, Kingdom Come: Deliverance, Project Cars, Schöner Trants Nerdquiz Folge 2 Runde 1', + 'age_limit': 16, + 'upload_date': '20140513', + 'timestamp': 1399980122, + } + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + webpage = self._download_webpage(url, video_id) + og_video = self._og_search_video_url(webpage, secure=False) + description = self._html_search_meta('description', webpage) + age_limit = int( + self._search_regex( + r'age=(\d+)', + self._html_search_meta( + 'age-de-meta-label', + webpage), + 'age_limit', + '0')) + mrss_url = self._search_regex(r'mrss=([^&]+)', og_video, 'mrss') + + mrss = self._download_xml(mrss_url, video_id, 'Downloading mrss') + title = mrss.find('.//item/title').text + thumbnail = mrss.find('.//item/image').get('url') + timestamp = parse_iso8601(mrss.find('.//pubDate').text, delimiter=' ') + content = mrss.find(xpath_with_ns('.//media:content', NAMESPACE_MAP)) + content_url = content.get('url') + + content = self._download_xml( + content_url, + video_id, + 'Downloading media:content') + rendition_items = content.findall('.//rendition') + duration = int(rendition_items[0].get('duration')) + formats = [ + { + 'url': re.sub(r'.*/(r2)', RAW_MP4_URL + r'\1', r.find('./src').text), + 'width': int(r.get('width')), + 'height': int(r.get('height')), + 'tbr': int(r.get('bitrate')), + } + for r in rendition_items + ] + + return { + 'id': video_id, + 'title': title, + 'thumbnail': thumbnail, + 'duration': duration, + 'formats': formats, + 'description': description, + 'age_limit': age_limit, + 'timestamp': timestamp, + } diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 09312e81a..2cba2bfc1 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -775,7 +775,7 @@ class YoutubeDLHandler(compat_urllib_request.HTTPHandler): https_response = http_response -def parse_iso8601(date_str): +def parse_iso8601(date_str, delimiter='T'): """ Return a UNIX timestamp from the given date """ if date_str is None: @@ -795,8 +795,8 @@ def parse_iso8601(date_str): timezone = datetime.timedelta( hours=sign * int(m.group('hours')), minutes=sign * int(m.group('minutes'))) - - dt = datetime.datetime.strptime(date_str, '%Y-%m-%dT%H:%M:%S') - timezone + date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter) + dt = datetime.datetime.strptime(date_str, date_format) - timezone return calendar.timegm(dt.timetuple()) |