1
0
mirror of https://codeberg.org/polarisfm/youtube-dl synced 2024-12-05 06:37:54 +01:00
youtube-dl/youtube_dl/extractor/gamespot.py

140 lines
5.8 KiB
Python
Raw Normal View History

2014-01-17 03:13:40 +01:00
from __future__ import unicode_literals
import re
2016-06-21 14:36:56 +02:00
from .once import OnceIE
from ..compat import (
compat_urllib_parse_unquote,
)
from ..utils import (
unescapeHTML,
2016-06-21 14:36:56 +02:00
url_basename,
dict_get,
)
2016-06-21 14:36:56 +02:00
class GameSpotIE(OnceIE):
2018-12-02 20:01:36 +01:00
_VALID_URL = r'https?://(?:www\.)?gamespot\.com/(?:video|article|review)s/(?:[^/]+/\d+-|embed/)(?P<id>\d+)'
_TESTS = [{
'url': 'http://www.gamespot.com/videos/arma-3-community-guide-sitrep-i/2300-6410818/',
'md5': 'b2a30deaa8654fcccd43713a6b6a4825',
'info_dict': {
'id': 'gs-2300-6410818',
'ext': 'mp4',
'title': 'Arma 3 - Community Guide: SITREP I',
2014-01-17 03:13:40 +01:00
'description': 'Check out this video where some of the basics of Arma 3 is explained.',
},
}, {
'url': 'http://www.gamespot.com/videos/the-witcher-3-wild-hunt-xbox-one-now-playing/2300-6424837/',
'info_dict': {
'id': 'gs-2300-6424837',
'ext': 'mp4',
'title': 'Now Playing - The Witcher 3: Wild Hunt',
'description': 'Join us as we take a look at the early hours of The Witcher 3: Wild Hunt and more.',
},
'params': {
'skip_download': True, # m3u8 downloads
},
}, {
'url': 'https://www.gamespot.com/videos/embed/6439218/',
'only_matching': True,
2017-11-13 10:30:12 +01:00
}, {
'url': 'https://www.gamespot.com/articles/the-last-of-us-2-receives-new-ps4-trailer/1100-6454469/',
'only_matching': True,
2018-12-02 20:01:36 +01:00
}, {
'url': 'https://www.gamespot.com/reviews/gears-of-war-review/1900-6161188/',
'only_matching': True,
}]
def _real_extract(self, url):
2014-11-04 23:04:12 +01:00
page_id = self._match_id(url)
webpage = self._download_webpage(url, page_id)
2014-11-04 23:04:12 +01:00
data_video_json = self._search_regex(
r'data-video=["\'](.*?)["\']', webpage, 'data video')
2016-06-21 14:36:56 +02:00
data_video = self._parse_json(unescapeHTML(data_video_json), page_id)
streams = data_video['videoStreams']
2016-06-21 14:36:56 +02:00
manifest_url = None
formats = []
f4m_url = streams.get('f4m_stream')
2016-06-21 14:36:56 +02:00
if f4m_url:
manifest_url = f4m_url
formats.extend(self._extract_f4m_formats(
f4m_url + '?hdcore=3.7.0', page_id, f4m_id='hds', fatal=False))
m3u8_url = dict_get(streams, ('m3u8_stream', 'adaptive_stream'))
2016-06-21 14:36:56 +02:00
if m3u8_url:
manifest_url = m3u8_url
m3u8_formats = self._extract_m3u8_formats(
m3u8_url, page_id, 'mp4', 'm3u8_native',
m3u8_id='hls', fatal=False)
formats.extend(m3u8_formats)
progressive_url = dict_get(
streams, ('progressive_hd', 'progressive_high', 'progressive_low', 'other_lr'))
2016-06-21 14:36:56 +02:00
if progressive_url and manifest_url:
qualities_basename = self._search_regex(
r'/([^/]+)\.csmil/',
2016-06-21 14:36:56 +02:00
manifest_url, 'qualities basename', default=None)
if qualities_basename:
QUALITIES_RE = r'((,\d+)+,?)'
qualities = self._search_regex(
QUALITIES_RE, qualities_basename,
'qualities', default=None)
if qualities:
qualities = list(map(lambda q: int(q), qualities.strip(',').split(',')))
qualities.sort()
http_template = re.sub(QUALITIES_RE, r'%d', qualities_basename)
http_url_basename = url_basename(progressive_url)
if m3u8_formats:
self._sort_formats(m3u8_formats)
m3u8_formats = list(filter(
lambda f: f.get('vcodec') != 'none', m3u8_formats))
2016-06-21 14:36:56 +02:00
if len(qualities) == len(m3u8_formats):
for q, m3u8_format in zip(qualities, m3u8_formats):
f = m3u8_format.copy()
f.update({
'url': progressive_url.replace(
http_url_basename, http_template % q),
'format_id': f['format_id'].replace('hls', 'http'),
'protocol': 'http',
})
formats.append(f)
else:
for q in qualities:
formats.append({
'url': progressive_url.replace(
http_url_basename, http_template % q),
'ext': 'mp4',
'format_id': 'http-%d' % q,
'tbr': q,
})
onceux_json = self._search_regex(
r'data-onceux-options=["\'](.*?)["\']', webpage, 'data video', default=None)
if onceux_json:
onceux_url = self._parse_json(unescapeHTML(onceux_json), page_id).get('metadataUri')
if onceux_url:
formats.extend(self._extract_once_formats(re.sub(
r'https?://[^/]+', 'http://once.unicornmedia.com', onceux_url),
http_formats_preference=-1))
2016-06-21 14:36:56 +02:00
if not formats:
for quality in ['sd', 'hd']:
# It's actually a link to a flv file
flv_url = streams.get('f4m_{0}'.format(quality))
if flv_url is not None:
formats.append({
'url': flv_url,
'ext': 'flv',
'format_id': quality,
})
2016-06-21 14:36:56 +02:00
self._sort_formats(formats)
return {
'id': data_video['guid'],
2014-11-04 23:04:12 +01:00
'display_id': page_id,
'title': compat_urllib_parse_unquote(data_video['title']),
'formats': formats,
2014-11-04 23:04:12 +01:00
'description': self._html_search_meta('description', webpage),
'thumbnail': self._og_search_thumbnail(webpage),
}