[gamespot] Extract DASH and HTTP formats

youtube_dl/extractor/common.py

@@ -2614,20 +2614,20 @@ class InfoExtractor(object):
         hls_host = hosts.get('hls')
         if hls_host:
             m3u8_url = re.sub(r'(https?://)[^/]+', r'\1' + hls_host, m3u8_url)
-        formats.extend(self._extract_m3u8_formats(
+        m3u8_formats = self._extract_m3u8_formats(
             m3u8_url, video_id, 'mp4', 'm3u8_native',
-            m3u8_id='hls', fatal=False))
+            m3u8_id='hls', fatal=False)
+        formats.extend(m3u8_formats)
 
         http_host = hosts.get('http')
-        if http_host and 'hdnea=' not in manifest_url:
-            REPL_REGEX = r'https://[^/]+/i/([^,]+),([^/]+),([^/]+).csmil/.+'
+        if http_host and m3u8_formats and 'hdnea=' not in m3u8_url:
+            REPL_REGEX = r'https?://[^/]+/i/([^,]+),([^/]+),([^/]+)\.csmil/.+'
             qualities = re.match(REPL_REGEX, m3u8_url).group(2).split(',')
             qualities_length = len(qualities)
-            if len(formats) in (qualities_length, qualities_length + 1, qualities_length * 2, qualities_length * 2 + 1):
+            if len(m3u8_formats) in (qualities_length, qualities_length + 1):
                 i = 0
-                http_formats = []
-                for f in formats:
-                    if f['protocol'] == 'm3u8_native' and f['vcodec'] != 'none':
+                for f in m3u8_formats:
+                    if f['vcodec'] != 'none':
                         for protocol in ('http', 'https'):
                             http_f = f.copy()
                             del http_f['manifest_url']
@@ -2638,9 +2638,8 @@ class InfoExtractor(object):
                                 'url': http_url,
                                 'protocol': protocol,
                             })
-                            http_formats.append(http_f)
+                            formats.append(http_f)
                         i += 1
-                formats.extend(http_formats)
 
         return formats
 

youtube_dl/extractor/gamespot.py

@@ -1,16 +1,7 @@
 from __future__ import unicode_literals
 
-import re
-
 from .once import OnceIE
-from ..compat import (
-    compat_urllib_parse_unquote,
-)
-from ..utils import (
-    unescapeHTML,
-    url_basename,
-    dict_get,
-)
+from ..compat import compat_urllib_parse_unquote
 
 
 class GameSpotIE(OnceIE):
@@ -24,17 +15,16 @@ class GameSpotIE(OnceIE):
             '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': '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,
@@ -49,90 +39,40 @@ class GameSpotIE(OnceIE):
     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 = self._parse_json(unescapeHTML(data_video_json), page_id)
+        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']
-
-        manifest_url = None
         formats = []
-        f4m_url = streams.get('f4m_stream')
-        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'))
+
+        m3u8_url = streams.get('adaptive_stream')
         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'))
-        if progressive_url and manifest_url:
-            qualities_basename = self._search_regex(
-                r'/([^/]+)\.csmil/',
-                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))
-                    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',
-                            })
+            for f in m3u8_formats:
                 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,
+                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'),
                 })
+                formats.append(http_f)
 
-        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))
+        mpd_url = streams.get('adaptive_dash')
+        if mpd_url:
+            formats.extend(self._extract_mpd_formats(
+                mpd_url, page_id, mpd_id='dash', fatal=False))
 
-        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,
-                    })
         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),

youtube_dl/extractor/tver.py

@@ -43,7 +43,7 @@ class TVerIE(InfoExtractor):
         info = {
             '_type': 'url_transparent',
             'description': try_get(main, lambda x: x['note'][0]['text'], compat_str),
-            'episode': int_or_none(try_get(main, lambda x: x['ext']['episode_number'])),
+            'episode_number': int_or_none(try_get(main, lambda x: x['ext']['episode_number'])),
         }
 
         if service == 'cx':