[lego] fix extraction and extract subtitle(closes #23687)

youtube_dl/extractor/lego.py

@@ -2,23 +2,24 @@
 from __future__ import unicode_literals
 
 import re
+import uuid
 
 from .common import InfoExtractor
-from ..compat import compat_str
+from ..compat import compat_HTTPError
 from ..utils import (
-    unescapeHTML,
+    ExtractorError,
-    parse_duration,
+    int_or_none,
-    get_element_by_class,
+    qualities,
 )
 
 
 class LEGOIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?lego\.com/(?P<locale>[^/]+)/(?:[^/]+/)*videos/(?:[^/]+/)*[^/?#]+-(?P<id>[0-9a-f]+)'
+    _VALID_URL = r'https?://(?:www\.)?lego\.com/(?P<locale>[a-z]{2}-[a-z]{2})/(?:[^/]+/)*videos/(?:[^/]+/)*[^/?#]+-(?P<id>[0-9a-f]{32})'
     _TESTS = [{
         'url': 'http://www.lego.com/en-us/videos/themes/club/blocumentary-kawaguchi-55492d823b1b4d5e985787fa8c2973b1',
         'md5': 'f34468f176cfd76488767fc162c405fa',
         'info_dict': {
-            'id': '55492d823b1b4d5e985787fa8c2973b1',
+            'id': '55492d82-3b1b-4d5e-9857-87fa8c2973b1_en-US',
             'ext': 'mp4',
             'title': 'Blocumentary Great Creations: Akiyuki Kawaguchi',
             'description': 'Blocumentary Great Creations: Akiyuki Kawaguchi',
@@ -26,103 +27,123 @@ class LEGOIE(InfoExtractor):
     }, {
         # geo-restricted but the contentUrl contain a valid url
         'url': 'http://www.lego.com/nl-nl/videos/themes/nexoknights/episode-20-kingdom-of-heroes-13bdc2299ab24d9685701a915b3d71e7##sp=399',
-        'md5': '4c3fec48a12e40c6e5995abc3d36cc2e',
+        'md5': 'c7420221f7ffd03ff056f9db7f8d807c',
         'info_dict': {
-            'id': '13bdc2299ab24d9685701a915b3d71e7',
+            'id': '13bdc229-9ab2-4d96-8570-1a915b3d71e7_nl-NL',
             'ext': 'mp4',
-            'title': 'Aflevering 20 - Helden van het koninkrijk',
+            'title': 'Aflevering 20:  Helden van het koninkrijk',
             'description': 'md5:8ee499aac26d7fa8bcb0cedb7f9c3941',
+            'age_limit': 5,
         },
     }, {
-        # special characters in title
+        # with subtitle
-        'url': 'http://www.lego.com/en-us/starwars/videos/lego-star-wars-force-surprise-9685ee9d12e84ff38e84b4e3d0db533d',
+        'url': 'https://www.lego.com/nl-nl/kids/videos/classic/creative-storytelling-the-little-puppy-aa24f27c7d5242bc86102ebdc0f24cba',
         'info_dict': {
-            'id': '9685ee9d12e84ff38e84b4e3d0db533d',
+            'id': 'aa24f27c-7d52-42bc-8610-2ebdc0f24cba_nl-NL',
             'ext': 'mp4',
-            'title': 'Force Surprise – LEGO® Star Wars™ Microfighters',
+            'title': 'De kleine puppy',
-            'description': 'md5:9c673c96ce6f6271b88563fe9dc56de3',
+            'description': 'md5:5b725471f849348ac73f2e12cfb4be06',
+            'age_limit': 1,
+            'subtitles': {
+                'nl': [{
+                    'ext': 'srt',
+                    'url': r're:^https://.+\.srt$',
+                }],
+            },
         },
         'params': {
             'skip_download': True,
         },
     }]
-    _BITRATES = [256, 512, 1024, 1536, 2560]
+    _QUALITIES = {
+        'Lowest': (64, 180, 320),
+        'Low': (64, 270, 480),
+        'Medium': (96, 360, 640),
+        'High': (128, 540, 960),
+        'Highest': (128, 720, 1280),
+    }
 
     def _real_extract(self, url):
         locale, video_id = re.match(self._VALID_URL, url).groups()
-        webpage = self._download_webpage(url, video_id)
+        countries = [locale.split('-')[1].upper()]
-        title = get_element_by_class('video-header', webpage).strip()
+        self._initialize_geo_bypass({
-        progressive_base = 'https://lc-mediaplayerns-live-s.legocdn.com/'
+            'countries': countries,
-        streaming_base = 'http://legoprod-f.akamaihd.net/'
-        content_url = self._html_search_meta('contentUrl', webpage)
-        path = self._search_regex(
-            r'(?:https?:)?//[^/]+/(?:[iz]/s/)?public/(.+)_[0-9,]+\.(?:mp4|webm)',
-            content_url, 'video path', default=None)
-        if not path:
-            player_url = self._proto_relative_url(self._search_regex(
-                r'<iframe[^>]+src="((?:https?)?//(?:www\.)?lego\.com/[^/]+/mediaplayer/video/[^"]+)',
-                webpage, 'player url', default=None))
-            if not player_url:
-                base_url = self._proto_relative_url(self._search_regex(
-                    r'data-baseurl="([^"]+)"', webpage, 'base url',
-                    default='http://www.lego.com/%s/mediaplayer/video/' % locale))
-                player_url = base_url + video_id
-            player_webpage = self._download_webpage(player_url, video_id)
-            video_data = self._parse_json(unescapeHTML(self._search_regex(
-                r"video='([^']+)'", player_webpage, 'video data')), video_id)
-            progressive_base = self._search_regex(
-                r'data-video-progressive-url="([^"]+)"',
-                player_webpage, 'progressive base', default='https://lc-mediaplayerns-live-s.legocdn.com/')
-            streaming_base = self._search_regex(
-                r'data-video-streaming-url="([^"]+)"',
-                player_webpage, 'streaming base', default='http://legoprod-f.akamaihd.net/')
-            item_id = video_data['ItemId']
-
-            net_storage_path = video_data.get('NetStoragePath') or '/'.join([item_id[:2], item_id[2:4]])
-            base_path = '_'.join([item_id, video_data['VideoId'], video_data['Locale'], compat_str(video_data['VideoVersion'])])
-            path = '/'.join([net_storage_path, base_path])
-        streaming_path = ','.join(map(lambda bitrate: compat_str(bitrate), self._BITRATES))
-
-        formats = self._extract_akamai_formats(
-            '%si/s/public/%s_,%s,.mp4.csmil/master.m3u8' % (streaming_base, path, streaming_path), video_id)
-        m3u8_formats = list(filter(
-            lambda f: f.get('protocol') == 'm3u8_native' and f.get('vcodec') != 'none',
-            formats))
-        if len(m3u8_formats) == len(self._BITRATES):
-            self._sort_formats(m3u8_formats)
-            for bitrate, m3u8_format in zip(self._BITRATES, m3u8_formats):
-                progressive_base_url = '%spublic/%s_%d.' % (progressive_base, path, bitrate)
-                mp4_f = m3u8_format.copy()
-                mp4_f.update({
-                    'url': progressive_base_url + 'mp4',
-                    'format_id': m3u8_format['format_id'].replace('hls', 'mp4'),
-                    'protocol': 'http',
         })
-                web_f = {
+
-                    'url': progressive_base_url + 'webm',
+        try:
-                    'format_id': m3u8_format['format_id'].replace('hls', 'webm'),
+            item = self._download_json(
-                    'width': m3u8_format['width'],
+                # https://contentfeed.services.lego.com/api/v2/item/[VIDEO_ID]?culture=[LOCALE]&contentType=Video
-                    'height': m3u8_format['height'],
+                'https://services.slingshot.lego.com/mediaplayer/v2',
-                    'tbr': m3u8_format.get('tbr'),
+                video_id, query={
-                    'ext': 'webm',
+                    'videoId': '%s_%s' % (uuid.UUID(video_id), locale),
-                }
+                }, headers=self.geo_verification_headers())
-                formats.extend([web_f, mp4_f])
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 451:
+                self.raise_geo_restricted(countries=countries)
+            raise
+
+        video = item['Video']
+        video_id = video['Id']
+        title = video['Title']
+
+        q = qualities(['Lowest', 'Low', 'Medium', 'High', 'Highest'])
+        formats = []
+        for video_source in item.get('VideoFormats', []):
+            video_source_url = video_source.get('Url')
+            if not video_source_url:
+                continue
+            video_source_format = video_source.get('Format')
+            if video_source_format == 'F4M':
+                formats.extend(self._extract_f4m_formats(
+                    video_source_url, video_id,
+                    f4m_id=video_source_format, fatal=False))
+            elif video_source_format == 'M3U8':
+                formats.extend(self._extract_m3u8_formats(
+                    video_source_url, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id=video_source_format, fatal=False))
             else:
-            for bitrate in self._BITRATES:
+                video_source_quality = video_source.get('Quality')
-                for ext in ('web', 'mp4'):
+                format_id = []
-                    formats.append({
+                for v in (video_source_format, video_source_quality):
-                        'format_id': '%s-%s' % (ext, bitrate),
+                    if v:
-                        'url': '%spublic/%s_%d.%s' % (progressive_base, path, bitrate, ext),
+                        format_id.append(v)
-                        'tbr': bitrate,
+                f = {
-                        'ext': ext,
+                    'format_id': '-'.join(format_id),
-                    })
+                    'quality': q(video_source_quality),
+                    'url': video_source_url,
+                }
+                quality = self._QUALITIES.get(video_source_quality)
+                if quality:
+                    f.update({
+                        'abr': quality[0],
+                        'height': quality[1],
+                        'width': quality[2],
+                    }),
+                formats.append(f)
         self._sort_formats(formats)
 
+        subtitles = {}
+        sub_file_id = video.get('SubFileId')
+        if sub_file_id and sub_file_id != '00000000-0000-0000-0000-000000000000':
+            net_storage_path = video.get('NetstoragePath')
+            invariant_id = video.get('InvariantId')
+            video_file_id = video.get('VideoFileId')
+            video_version = video.get('VideoVersion')
+            if net_storage_path and invariant_id and video_file_id and video_version:
+                subtitles.setdefault(locale[:2], []).append({
+                    'url': 'https://lc-mediaplayerns-live-s.legocdn.com/public/%s/%s_%s_%s_%s_sub.srt' % (net_storage_path, invariant_id, video_file_id, locale, video_version),
+                })
+
         return {
             'id': video_id,
             'title': title,
-            'description': self._html_search_meta('description', webpage),
+            'description': video.get('Description'),
-            'thumbnail': self._html_search_meta('thumbnail', webpage),
+            'thumbnail': video.get('GeneratedCoverImage') or video.get('GeneratedThumbnail'),
-            'duration': parse_duration(self._html_search_meta('duration', webpage)),
+            'duration': int_or_none(video.get('Length')),
             'formats': formats,
+            'subtitles': subtitles,
+            'age_limit': int_or_none(video.get('AgeFrom')),
+            'season': video.get('SeasonTitle'),
+            'season_number': int_or_none(video.get('Season')) or None,
+            'episode_number': int_or_none(video.get('Episode')) or None,
         }