Prepare to upload.

[youtubedl] / youtube_dl / extractor / gamespot.py

diff --git a/youtube_dl/extractor/gamespot.py b/youtube_dl/extractor/gamespot.py
index 7585b70618d1e4f92e8297fbf4d1397359a5224b..4236a5ed8a9bd638c31cfe33f040c555268c937f 100644 (file)
--- a/youtube_dl/extractor/gamespot.py
+++ b/youtube_dl/extractor/gamespot.py
@@ -1,55 +1,139 @@
+from __future__ import unicode_literals
+

 import re
 import re

-import xml.etree.ElementTree

 
 

-from .common import InfoExtractor
+from .once import OnceIE
+from ..compat import (
+    compat_urllib_parse_unquote,
+)

 from ..utils import (
 from ..utils import (

-    unified_strdate,
-    compat_urllib_parse,
+    unescapeHTML,
+    url_basename,
+    dict_get,

 )
 
 )
 

-class GameSpotIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<page_id>\d+)/?'
-    _TEST = {
-        u"url": u"http://www.gamespot.com/arma-iii/videos/arma-iii-community-guide-sitrep-i-6410818/",
-        u"file": u"6410818.mp4",
-        u"md5": u"b2a30deaa8654fcccd43713a6b6a4825",
-        u"info_dict": {
-            u"title": u"Arma III - Community Guide: SITREP I",
-            u"upload_date": u"20130627", 
-        }
-    }

 
 

+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',
+        'info_dict': {
+            'id': 'gs-2300-6410818',
+            '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': '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,
+    }, {
+        '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):
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        page_id = mobj.group('page_id')
+        page_id = self._match_id(url)

         webpage = self._download_webpage(url, page_id)
         webpage = self._download_webpage(url, page_id)

-        video_id = self._html_search_regex([r'"og:video" content=".*?\?id=(\d+)"',
-                                            r'http://www\.gamespot\.com/videoembed/(\d+)'],
-                                           webpage, 'video id')
-        data = compat_urllib_parse.urlencode({'id': video_id, 'newplayer': '1'})
-        info_url = 'http://www.gamespot.com/pages/video_player/xml.php?' + data
-        info_xml = self._download_webpage(info_url, video_id)
-        doc = xml.etree.ElementTree.fromstring(info_xml)
-        clip_el = doc.find('./playList/clip')
-
-        http_urls = [{'url': node.find('filePath').text,
-                      'rate': int(node.find('rate').text)}
-            for node in clip_el.find('./httpURI')]
-        best_quality = sorted(http_urls, key=lambda f: f['rate'])[-1]
-        video_url = best_quality['url']
-        title = clip_el.find('./title').text
-        ext = video_url.rpartition('.')[2]
-        thumbnail_url = clip_el.find('./screenGrabURI').text
-        view_count = int(clip_el.find('./views').text)
-        upload_date = unified_strdate(clip_el.find('./postDate').text)
-
-        return [{
-            'id'          : video_id,
-            'url'         : video_url,
-            'ext'         : ext,
-            'title'       : title,
-            'thumbnail'   : thumbnail_url,
-            'upload_date' : upload_date,
-            'view_count'  : view_count,
-        }]
+        data_video_json = self._search_regex(
+            r'data-video=["\'](.*?)["\']', webpage, 'data video')
+        data_video = self._parse_json(unescapeHTML(data_video_json), page_id)
+        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'))
+        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',
+                            })
+                            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))
+
+        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'],
+            'display_id': page_id,
+            'title': compat_urllib_parse_unquote(data_video['title']),
+            'formats': formats,
+            'description': self._html_search_meta('description', webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+        }