diff options
Diffstat (limited to 'youtube_dl/extractor/gamespot.py')
| -rw-r--r-- | youtube_dl/extractor/gamespot.py | 93 |
1 files changed, 45 insertions, 48 deletions
diff --git a/youtube_dl/extractor/gamespot.py b/youtube_dl/extractor/gamespot.py index 4ffdd7515..7a1beae3c 100644 --- a/youtube_dl/extractor/gamespot.py +++ b/youtube_dl/extractor/gamespot.py @@ -1,20 +1,11 @@ from __future__ import unicode_literals -import re -import json +from .once import OnceIE +from ..compat import compat_urllib_parse_unquote -from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse_unquote, - compat_urlparse, -) -from ..utils import ( - unescapeHTML, -) - -class GameSpotIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?gamespot\.com/.*-(?P<id>\d+)/?' +class GameSpotIE(OnceIE): + _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', @@ -24,58 +15,64 @@ class GameSpotIE(InfoExtractor): 'title': 'Arma 3 - Community Guide: SITREP I', 'description': 'Check out this video where some of the basics of Arma 3 is explained.', }, + 'skip': 'manifest URL give HTTP Error 404: Not Found', }, { 'url': 'http://www.gamespot.com/videos/the-witcher-3-wild-hunt-xbox-one-now-playing/2300-6424837/', + 'md5': '173ea87ad762cf5d3bf6163dceb255a6', 'info_dict': { 'id': 'gs-2300-6424837', - 'ext': 'flv', - 'title': 'The Witcher 3: Wild Hunt [Xbox ONE] - Now Playing', + '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.', }, + }, { + 'url': 'https://www.gamespot.com/videos/embed/6439218/', + 'only_matching': True, + }, { + 'url': 'https://www.gamespot.com/articles/the-last-of-us-2-receives-new-ps4-trailer/1100-6454469/', + 'only_matching': True, + }, { + 'url': 'https://www.gamespot.com/reviews/gears-of-war-review/1900-6161188/', + 'only_matching': True, }] def _real_extract(self, url): page_id = self._match_id(url) webpage = self._download_webpage(url, page_id) - data_video_json = self._search_regex( - r'data-video=["\'](.*?)["\']', webpage, 'data video') - data_video = json.loads(unescapeHTML(data_video_json)) + data_video = self._parse_json(self._html_search_regex( + r'data-video=(["\'])({.*?})\1', webpage, + 'video data', group=2), page_id) + title = compat_urllib_parse_unquote(data_video['title']) streams = data_video['videoStreams'] - formats = [] - f4m_url = streams.get('f4m_stream') - if f4m_url is not None: - # Transform the manifest url to a link to the mp4 files - # they are used in mobile devices. - f4m_path = compat_urlparse.urlparse(f4m_url).path - QUALITIES_RE = r'((,\d+)+,?)' - qualities = self._search_regex(QUALITIES_RE, f4m_path, 'qualities').strip(',').split(',') - http_path = f4m_path[1:].split('/', 1)[1] - http_template = re.sub(QUALITIES_RE, r'%s', http_path) - http_template = http_template.replace('.csmil/manifest.f4m', '') - http_template = compat_urlparse.urljoin( - 'http://video.gamespotcdn.com/', http_template) - for q in qualities: - formats.append({ - 'url': http_template % q, - 'ext': 'mp4', - 'format_id': q, + + m3u8_url = streams.get('adaptive_stream') + if m3u8_url: + m3u8_formats = self._extract_m3u8_formats( + m3u8_url, page_id, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False) + for f in m3u8_formats: + formats.append(f) + http_f = f.copy() + del http_f['manifest_url'] + http_f.update({ + 'format_id': f['format_id'].replace('hls-', 'http-'), + 'protocol': 'http', + 'url': f['url'].replace('.m3u8', '.mp4'), }) - else: - 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, - }) + formats.append(http_f) + + mpd_url = streams.get('adaptive_dash') + if mpd_url: + formats.extend(self._extract_mpd_formats( + mpd_url, page_id, mpd_id='dash', fatal=False)) + + self._sort_formats(formats) return { - 'id': data_video['guid'], + 'id': data_video.get('guid') or page_id, 'display_id': page_id, - 'title': compat_urllib_parse_unquote(data_video['title']), + 'title': title, 'formats': formats, 'description': self._html_search_meta('description', webpage), 'thumbnail': self._og_search_thumbnail(webpage), |