diff options
| author | Jaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com> | 2015-05-15 18:42:59 +0200 | 
|---|---|---|
| committer | Jaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com> | 2015-05-15 18:42:59 +0200 | 
| commit | 34fe5a94baf9e7ea437de68621a5fa73780c0f17 (patch) | |
| tree | cb9b105f9bae7c401ef2110b0132ee705636af3e | |
| parent | 62c95fd5fcb8dbea2faeb4edac4c5177cbac5912 (diff) | |
[gamespot] Add support for videos that don't use 'f4m_stream' (fixes #5707)
| -rw-r--r-- | youtube_dl/extractor/gamespot.py | 60 | 
1 files changed, 40 insertions, 20 deletions
diff --git a/youtube_dl/extractor/gamespot.py b/youtube_dl/extractor/gamespot.py index 47373e215..5927455f6 100644 --- a/youtube_dl/extractor/gamespot.py +++ b/youtube_dl/extractor/gamespot.py @@ -15,7 +15,7 @@ from ..utils import (  class GameSpotIE(InfoExtractor):      _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<id>\d+)/?' -    _TEST = { +    _TESTS = [{          'url': 'http://www.gamespot.com/videos/arma-3-community-guide-sitrep-i/2300-6410818/',          'md5': 'b2a30deaa8654fcccd43713a6b6a4825',          'info_dict': { @@ -23,8 +23,16 @@ class GameSpotIE(InfoExtractor):              'ext': 'mp4',              'title': 'Arma 3 - Community Guide: SITREP I',              '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': 'flv', +            'title': 'The Witcher 3: Wild Hunt [Xbox ONE]  - Now Playing', +            'description': 'Join us as we take a look at the early hours of The Witcher 3: Wild Hunt and more.', +        }, +    }]      def _real_extract(self, url):          page_id = self._match_id(url) @@ -32,25 +40,37 @@ class GameSpotIE(InfoExtractor):          data_video_json = self._search_regex(              r'data-video=["\'](.*?)["\']', webpage, 'data video')          data_video = json.loads(unescapeHTML(data_video_json)) +        streams = data_video['videoStreams'] -        # Transform the manifest url to a link to the mp4 files -        # they are used in mobile devices. -        f4m_url = data_video['videoStreams']['f4m_stream'] -        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)          formats = [] -        for q in qualities: -            formats.append({ -                'url': http_template % q, -                'ext': 'mp4', -                'format_id': q, -            }) +        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, +                }) +        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, +                    })          return {              'id': data_video['guid'],  | 
