1
0
mirror of https://github.com/ytdl-org/youtube-dl.git synced 2024-12-05 12:38:05 +01:00
youtube-dl/youtube_dl/extractor/metacritic.py

56 lines
2.0 KiB
Python
Raw Normal View History

from __future__ import unicode_literals
2013-09-06 18:08:07 +02:00
import re
from .common import InfoExtractor
2013-12-10 21:03:53 +01:00
from ..utils import (
2014-01-20 22:11:34 +01:00
fix_xml_ampersands,
2013-12-10 21:03:53 +01:00
)
2013-09-06 18:08:07 +02:00
class MetacriticIE(InfoExtractor):
_VALID_URL = r'https?://www\.metacritic\.com/.+?/trailers/(?P<id>\d+)'
_TEST = {
'url': 'http://www.metacritic.com/game/playstation-4/infamous-second-son/trailers/3698222',
'info_dict': {
2014-03-29 14:46:05 +01:00
'id': '3698222',
'ext': 'mp4',
'title': 'inFamous: Second Son - inSide Sucker Punch: Smoke & Mirrors',
'description': 'Take a peak behind-the-scenes to see how Sucker Punch brings smoke into the universe of inFAMOUS Second Son on the PS4.',
'duration': 221,
2013-09-06 18:08:07 +02:00
},
}
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
webpage = self._download_webpage(url, video_id)
# The xml is not well formatted, there are raw '&'
2013-12-10 21:03:53 +01:00
info = self._download_xml('http://www.metacritic.com/video_data?video=' + video_id,
2014-11-23 21:39:15 +01:00
video_id, 'Downloading info xml', transform_source=fix_xml_ampersands)
2013-09-06 18:08:07 +02:00
clip = next(c for c in info.findall('playList/clip') if c.find('id').text == video_id)
formats = []
for videoFile in clip.findall('httpURI/videoFile'):
rate_str = videoFile.find('rate').text
video_url = videoFile.find('filePath').text
formats.append({
'url': video_url,
'ext': 'mp4',
'format_id': rate_str,
'tbr': int(rate_str),
2013-09-06 18:08:07 +02:00
})
self._sort_formats(formats)
2013-09-06 18:08:07 +02:00
description = self._html_search_regex(r'<b>Description:</b>(.*?)</p>',
2014-11-23 21:39:15 +01:00
webpage, 'description', flags=re.DOTALL)
2013-09-06 18:08:07 +02:00
return {
2013-09-06 18:08:07 +02:00
'id': video_id,
'title': clip.find('title').text,
'formats': formats,
'description': description,
'duration': int(clip.find('duration').text),
}