From b1b01841afac9b65b706c3436a5717b603458491 Mon Sep 17 00:00:00 2001 From: Charles Chen Date: Mon, 14 Jul 2014 11:00:55 -0700 Subject: [PATCH] [MLB] Add new extractor --- youtube_dl/extractor/__init__.py | 1 + youtube_dl/extractor/mlb.py | 67 ++++++++++++++++++++++++++++++++ 2 files changed, 68 insertions(+) create mode 100644 youtube_dl/extractor/mlb.py diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 15d2f0e2a..f75939a05 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -169,6 +169,7 @@ from .metacafe import MetacafeIE from .metacritic import MetacriticIE from .mit import TechTVMITIE, MITIE, OCWMITIE from .mixcloud import MixcloudIE +from .mlb import MlbIE from .mpora import MporaIE from .mofosex import MofosexIE from .mooshare import MooshareIE diff --git a/youtube_dl/extractor/mlb.py b/youtube_dl/extractor/mlb.py new file mode 100644 index 000000000..2b500bdff --- /dev/null +++ b/youtube_dl/extractor/mlb.py @@ -0,0 +1,67 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor + + +class MlbIE(InfoExtractor): + _VALID_URL = r'http?://m\.mlb\.com/video/topic/[0-9]+/v(?Pn?\d+)/.*$' + _TEST = { + 'url': 'http://m.mlb.com/video/topic/81536970/v34496663/mianym-stanton-practices-for-the-home-run-derby', + 'md5': u'd9c022c10d21f849f49c05ae12a8a7e9', + 'info_dict': { + 'id': '34496663', + 'ext': 'mp4', + 'format': 'mp4', + 'description': "7/11/14: Giancarlo Stanton practices for the Home Run Derby prior to the game against the Mets", + 'title': "Stanton prepares for Derby", + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + webpage = self._download_webpage(url, video_id) + + title = self._og_search_title(webpage, default=video_id) + description = self._html_search_regex(r'', webpage, 'description', fatal=False) + thumbnail = self._html_search_regex(r'', webpage, 'image', fatal=False) + + # use the thumbnail URL to find the folder that contains the videos + _image_url = r'http://mediadownloads.mlb.com/mlbam/(?P<_date>n?.+)/images/.*$' + bobj = re.match(_image_url, thumbnail) + datestr = bobj.group('_date') + base_url = 'http://mediadownloads.mlb.com/mlbam/' + datestr + filespage = self._download_webpage(base_url, video_id) + + # Try 1800K, 1500K, 1200K, 600K, then 300K videos + video = self._html_search_regex(r'