From: Remita Amine <remitamine@gmail.com>
Date: Sat, 8 Oct 2016 07:33:02 +0000 (+0100)
Subject: [lego] improve info extraction and bypass geo restriction(closes #10872)
X-Git-Url: https://git.cielonegro.org/gitweb.cgi?a=commitdiff_plain;h=1dd58e14d846a64a3c014531b1dc7a377648c73b;p=youtube-dl.git

[lego] improve info extraction and bypass geo restriction(closes #10872)
---

diff --git a/youtube_dl/extractor/lego.py b/youtube_dl/extractor/lego.py
index 5be7d622c..d3bca6435 100644
--- a/youtube_dl/extractor/lego.py
+++ b/youtube_dl/extractor/lego.py
@@ -1,45 +1,86 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..compat import compat_str
 from ..utils import (
     unescapeHTML,
-    int_or_none,
+    parse_duration,
+    get_element_by_class,
 )
 
 
 class LEGOIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?lego\.com/(?:[^/]+/)*videos/(?:[^/]+/)*[^/?#]+-(?P<id>[0-9a-f]+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?lego\.com/(?P<locale>[^/]+)/(?:[^/]+/)*videos/(?:[^/]+/)*[^/?#]+-(?P<id>[0-9a-f]+)'
+    _TESTS = [{
         'url': 'http://www.lego.com/en-us/videos/themes/club/blocumentary-kawaguchi-55492d823b1b4d5e985787fa8c2973b1',
         'md5': 'f34468f176cfd76488767fc162c405fa',
         'info_dict': {
             'id': '55492d823b1b4d5e985787fa8c2973b1',
             'ext': 'mp4',
             'title': 'Blocumentary Great Creations: Akiyuki Kawaguchi',
-        }
-    }
+            'description': 'Blocumentary Great Creations: Akiyuki Kawaguchi',
+        },
+    }, {
+        # 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',
+        'info_dict': {
+            'id': '13bdc2299ab24d9685701a915b3d71e7',
+            'ext': 'mp4',
+            'title': 'Aflevering 20 - Helden van het koninkrijk',
+            'description': 'md5:8ee499aac26d7fa8bcb0cedb7f9c3941',
+        },
+    }, {
+        # special characters in title
+        'url': 'http://www.lego.com/en-us/starwars/videos/lego-star-wars-force-surprise-9685ee9d12e84ff38e84b4e3d0db533d',
+        'info_dict': {
+            'id': '9685ee9d12e84ff38e84b4e3d0db533d',
+            'ext': 'mp4',
+            'title': 'Force Surprise – LEGO® Star Wars™ Microfighters',
+            'description': 'md5:9c673c96ce6f6271b88563fe9dc56de3',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]
     _BITRATES = [256, 512, 1024, 1536, 2560]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(
-            'http://www.lego.com/en-US/mediaplayer/video/' + video_id, video_id)
-        title = self._search_regex(r'<title>(.+?)</title>', webpage, 'title')
-        video_data = self._parse_json(unescapeHTML(self._search_regex(
-            r"video='([^']+)'", webpage, 'video data')), video_id)
-        progressive_base = self._search_regex(
-            r'data-video-progressive-url="([^"]+)"',
-            webpage, 'progressive base', default='https://lc-mediaplayerns-live-s.legocdn.com/')
-        streaming_base = self._search_regex(
-            r'data-video-streaming-url="([^"]+)"',
-            webpage, 'streaming base', default='http://legoprod-f.akamaihd.net/')
-        item_id = video_data['ItemId']
+        locale, video_id = re.match(self._VALID_URL, url).groups()
+        webpage = self._download_webpage(url, video_id)
+        title = get_element_by_class('video-header', webpage).strip()
+        progressive_base = 'https://lc-mediaplayerns-live-s.legocdn.com/'
+        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])
+            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(
@@ -80,7 +121,8 @@ class LEGOIE(InfoExtractor):
         return {
             'id': video_id,
             'title': title,
-            'thumbnail': video_data.get('CoverImageUrl'),
-            'duration': int_or_none(video_data.get('Length')),
+            'description': self._html_search_meta('description', webpage),
+            'thumbnail': self._html_search_meta('thumbnail', webpage),
+            'duration': parse_duration(self._html_search_meta('duration', webpage)),
             'formats': formats,
         }