[generic] Detect more MLB videos (fixes #5443)

[youtube-dl] / youtube_dl / extractor / escapist.py

diff --git a/youtube_dl/extractor/escapist.py b/youtube_dl/extractor/escapist.py
index 476fc22b93424b13255d5eec3578eb985dbfbdfd..8facf11858fefe48490f4b40e1129c0c45d36b06 100644 (file)
--- a/youtube_dl/extractor/escapist.py
+++ b/youtube_dl/extractor/escapist.py
@@ -1,82 +1,105 @@
 from __future__ import unicode_literals
 
-import re
+import json
 
 from .common import InfoExtractor
-from ..utils import (
-    compat_urllib_parse,
+from ..compat import compat_urllib_request
 
-    ExtractorError,
+from ..utils import (
+    determine_ext,
+    clean_html,
+    qualities,
 )
 
 
+def _decrypt_config(key, string):
+    a = ''
+    i = ''
+    r = ''
+
+    while len(a) < (len(string) / 2):
+        a += key
+
+    a = a[0:int(len(string) / 2)]
+
+    t = 0
+    while t < len(string):
+        i += chr(int(string[t] + string[t + 1], 16))
+        t += 2
+
+    icko = [s for s in i]
+
+    for t, c in enumerate(a):
+        r += chr(ord(c) ^ ord(icko[t]))
+
+    return r
+
+
 class EscapistIE(InfoExtractor):
-    _VALID_URL = r'^https?://?(www\.)?escapistmagazine\.com/videos/view/(?P<showname>[^/]+)/(?P<id>[0-9]+)-'
-    _TEST = {
+    _VALID_URL = r'https?://?(www\.)?escapistmagazine\.com/videos/view/[^/?#]+/(?P<id>[0-9]+)-[^/?#]*(?:$|[?#])'
+    _TESTS = [{
         'url': 'http://www.escapistmagazine.com/videos/view/the-escapist-presents/6618-Breaking-Down-Baldurs-Gate',
-        'md5': 'ab3a706c681efca53f0a35f1415cf0d1',
+        'md5': 'c6793dbda81388f4264c1ba18684a74d',
         'info_dict': {
             'id': '6618',
             'ext': 'mp4',
             'description': "Baldur's Gate: Original, Modded or Enhanced Edition? I'll break down what you can expect from the new Baldur's Gate: Enhanced Edition.",
-            'uploader': 'the-escapist-presents',
             'title': "Breaking Down Baldur's Gate",
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 264,
+        }
+    }, {
+        'url': 'http://www.escapistmagazine.com/videos/view/zero-punctuation/10044-Evolve-One-vs-Multiplayer',
+        'md5': 'cf8842a8a46444d241f9a9980d7874f2',
+        'info_dict': {
+            'id': '10044',
+            'ext': 'mp4',
+            'description': 'This week, Zero Punctuation reviews Evolve.',
+            'title': 'Evolve - One vs Multiplayer',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 304,
         }
-    }
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        showName = mobj.group('showname')
-        video_id = mobj.group('id')
-
-        self.report_extraction(video_id)
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        videoDesc = self._html_search_regex(
-            r'<meta name="description" content="([^"]*)"',
-            webpage, 'description', fatal=False)
+        imsVideo = self._parse_json(
+            self._search_regex(
+                r'imsVideo\.play\(({.+?})\);', webpage, 'imsVideo'),
+            video_id)
+        video_id = imsVideo['videoID']
+        key = imsVideo['hash']
+
+        quality = qualities(['lq', 'hq', 'hd'])
 
-        playerUrl = self._og_search_video_url(webpage, name='player URL')
+        formats = []
+        for q in ['lq', 'hq', 'hd']:
+            config_req = compat_urllib_request.Request(
+                'http://www.escapistmagazine.com/videos/'
+                'vidconfig.php?videoID=%s&hash=%s&quality=%s' % (video_id, key, 'mp4_' + q))
+            config_req.add_header('Referer', url)
+            config = self._download_webpage(config_req, video_id, 'Downloading video config ' + q.upper())
 
-        title = self._html_search_regex(
-            r'<meta name="title" content="([^"]*)"',
-            webpage, 'title').split(' : ')[-1]
+            data = json.loads(_decrypt_config(key, config))
 
-        configUrl = self._search_regex('config=(.*)$', playerUrl, 'config URL')
-        configUrl = compat_urllib_parse.unquote(configUrl)
+            title = clean_html(data['videoData']['title'])
+            duration = data['videoData']['duration'] / 1000
 
-        formats = []
+            for i, v in enumerate(data['files']['videos']):
 
-        def _add_format(name, cfgurl, quality):
-            config = self._download_json(
-                cfgurl, video_id,
-                'Downloading ' + name + ' configuration',
-                'Unable to download ' + name + ' configuration',
-                transform_source=lambda s: s.replace("'", '"'))
-
-            playlist = config['playlist']
-            formats.append({
-                'url': playlist[1]['url'],
-                'format_id': name,
-                'quality': quality,
-            })
-
-        _add_format('normal', configUrl, quality=0)
-        hq_url = (configUrl +
-                  ('&hq=1' if '?' in configUrl else configUrl + '?hq=1'))
-        try:
-            _add_format('hq', hq_url, quality=1)
-        except ExtractorError:
-            pass  # That's fine, we'll just use normal quality
-
-        self._sort_formats(formats)
+                formats.append({
+                    'url': v,
+                    'format_id': determine_ext(v) + '_' + q + str(i),
+                    'quality': quality(q),
+                })
 
         return {
             'id': video_id,
             'formats': formats,
-            'uploader': showName,
             'title': title,
             'thumbnail': self._og_search_thumbnail(webpage),
-            'description': videoDesc,
-            'player_url': playerUrl,
+            'description': self._og_search_description(webpage),
+            'duration': duration,
         }