[Steam] Fix extractor (#2029)

Closes #1992
Authored by: u-spec-png

yt_dlp/extractor/steam.py

@@ -7,14 +7,13 @@ from ..utils import (
     extract_attributes,
     ExtractorError,
     get_element_by_class,
-    js_to_json,
 )
 
 
 class SteamIE(InfoExtractor):
     _VALID_URL = r"""(?x)
-        https?://store\.steampowered\.com/
+        https?://(?:store\.steampowered|steamcommunity)\.com/
-            (agecheck/)?
+            (?:agecheck/)?
             (?P<urltype>video|app)/ #If the page is only for videos or for a game
             (?P<gameID>\d+)/?
             (?P<videoID>\d*)(?P<extra>\??) # For urltype == video we sometimes get the videoID
@@ -26,22 +25,25 @@ class SteamIE(InfoExtractor):
     _TESTS = [{
         'url': 'http://store.steampowered.com/video/105600/',
         'playlist': [
+            {
+                'md5': '695242613303ffa2a4c44c9374ddc067',
+                'info_dict': {
+                    'id': '256785003',
+                    'ext': 'mp4',
+                    'title': 'Terraria video 256785003',
+                    'thumbnail': r're:^https://cdn\.[^\.]+\.steamstatic\.com',
+                    'n_entries': 2,
+                }
+            },
             {
                 'md5': '6a294ee0c4b1f47f5bb76a65e31e3592',
                 'info_dict': {
                     'id': '2040428',
                     'ext': 'mp4',
-                    'title': 'Terraria 1.3 Trailer',
+                    'title': 'Terraria video 2040428',
-                    'playlist_index': 1,
-                }
-            },
-            {
-                'md5': '911672b20064ca3263fa89650ba5a7aa',
-                'info_dict': {
-                    'id': '2029566',
-                    'ext': 'mp4',
-                    'title': 'Terraria 1.2 Trailer',
                     'playlist_index': 2,
+                    'thumbnail': r're:^https://cdn\.[^\.]+\.steamstatic\.com',
+                    'n_entries': 2,
                 }
             }
         ],
@@ -53,96 +55,76 @@ class SteamIE(InfoExtractor):
             'playlistend': 2,
         }
     }, {
-        'url': 'http://steamcommunity.com/sharedfiles/filedetails/?id=242472205',
+        'url': 'https://store.steampowered.com/app/271590/Grand_Theft_Auto_V/',
         'info_dict': {
-            'id': 'X8kpJBlzD2E',
+            'id': '256757115',
+            'title': 'Grand Theft Auto V video 256757115',
             'ext': 'mp4',
-            'upload_date': '20140617',
+            'thumbnail': r're:^https://cdn\.[^\.]+\.steamstatic\.com',
-            'title': 'FRONTIERS - Trapping',
+            'n_entries': 20,
-            'description': 'md5:bf6f7f773def614054089e5769c12a6e',
+        },
-            'uploader': 'AAD Productions',
-            'uploader_id': 'AtomicAgeDogGames',
-        }
     }]
 
     def _real_extract(self, url):
         m = self._match_valid_url(url)
         fileID = m.group('fileID')
         if fileID:
-            videourl = url
+            video_url = url
             playlist_id = fileID
         else:
             gameID = m.group('gameID')
             playlist_id = gameID
-            videourl = self._VIDEO_PAGE_TEMPLATE % playlist_id
+            video_url = self._VIDEO_PAGE_TEMPLATE % playlist_id
 
-        self._set_cookie('steampowered.com', 'mature_content', '1')
+        self._set_cookie('steampowered.com', 'wants_mature_content', '1')
+        self._set_cookie('steampowered.com', 'birthtime', '944006401')
+        self._set_cookie('steampowered.com', 'lastagecheckage', '1-0-2000')
 
-        webpage = self._download_webpage(videourl, playlist_id)
+        webpage = self._download_webpage(video_url, playlist_id)
 
-        if re.search('<h2>Please enter your birth date to continue:</h2>', webpage) is not None:
+        if re.search('<div[^>]+>Please enter your birth date to continue:</div>', webpage) is not None:
-            videourl = self._AGECHECK_TEMPLATE % playlist_id
+            video_url = self._AGECHECK_TEMPLATE % playlist_id
             self.report_age_confirmation()
-            webpage = self._download_webpage(videourl, playlist_id)
+            webpage = self._download_webpage(video_url, playlist_id)
 
-        flash_vars = self._parse_json(self._search_regex(
+        videos = re.findall(r'(<div[^>]+id=[\'"]highlight_movie_(\d+)[\'"][^>]+>)', webpage)
-            r'(?s)rgMovieFlashvars\s*=\s*({.+?});', webpage,
-            'flash vars'), playlist_id, js_to_json)
-
-        playlist_title = None
         entries = []
-        if fileID:
+        playlist_title = get_element_by_class('apphub_AppName', webpage)
-            playlist_title = get_element_by_class('workshopItemTitle', webpage)
+        for movie, movie_id in videos:
-            for movie in flash_vars.values():
+            if not movie:
-                if not movie:
+                continue
-                    continue
+            movie = extract_attributes(movie)
-                youtube_id = movie.get('YOUTUBE_VIDEO_ID')
+            if not movie_id:
-                if not youtube_id:
+                continue
-                    continue
+            entry = {
+                'id': movie_id,
+                'title': f'{playlist_title} video {movie_id}',
+            }
+            formats = []
+            if movie:
+                entry['thumbnail'] = movie.get('data-poster')
+                for quality in ('', '-hd'):
+                    for ext in ('webm', 'mp4'):
+                        video_url = movie.get('data-%s%s-source' % (ext, quality))
+                        if video_url:
+                            formats.append({
+                                'format_id': ext + quality,
+                                'url': video_url,
+                            })
+            self._sort_formats(formats)
+            entry['formats'] = formats
+            entries.append(entry)
+        embedded_videos = re.findall(r'(<iframe[^>]+>)', webpage)
+        for evideos in embedded_videos:
+            evideos = extract_attributes(evideos).get('src')
+            video_id = self._search_regex(r'youtube\.com/embed/([0-9A-Za-z_-]{11})', evideos, 'youtube_video_id', default=None)
+            if video_id:
                 entries.append({
-                    '_type': 'url',
+                    '_type': 'url_transparent',
-                    'url': youtube_id,
+                    'id': video_id,
+                    'url': video_id,
                     'ie_key': 'Youtube',
                 })
-        else:
-            playlist_title = get_element_by_class('apphub_AppName', webpage)
-            for movie_id, movie in flash_vars.items():
-                if not movie:
-                    continue
-                video_id = self._search_regex(r'movie_(\d+)', movie_id, 'video id', fatal=False)
-                title = movie.get('MOVIE_NAME')
-                if not title or not video_id:
-                    continue
-                entry = {
-                    'id': video_id,
-                    'title': title.replace('+', ' '),
-                }
-                formats = []
-                flv_url = movie.get('FILENAME')
-                if flv_url:
-                    formats.append({
-                        'format_id': 'flv',
-                        'url': flv_url,
-                    })
-                highlight_element = self._search_regex(
-                    r'(<div[^>]+id="highlight_movie_%s"[^>]+>)' % video_id,
-                    webpage, 'highlight element', fatal=False)
-                if highlight_element:
-                    highlight_attribs = extract_attributes(highlight_element)
-                    if highlight_attribs:
-                        entry['thumbnail'] = highlight_attribs.get('data-poster')
-                        for quality in ('', '-hd'):
-                            for ext in ('webm', 'mp4'):
-                                video_url = highlight_attribs.get('data-%s%s-source' % (ext, quality))
-                                if video_url:
-                                    formats.append({
-                                        'format_id': ext + quality,
-                                        'url': video_url,
-                                    })
-                if not formats and not self.get_param('ignore_no_formats'):
-                    continue
-                entry['formats'] = formats
-                entries.append(entry)
         if not entries:
             raise ExtractorError('Could not find any videos')