Start new release and annotate with bugs being fixed.

[youtubedl] / youtube_dl / extractor / beeg.py

diff --git a/youtube_dl/extractor/beeg.py b/youtube_dl/extractor/beeg.py
index b38057f2f500f520829ff9c5d7324e66558eb356..bf22a41b745db2eef277c6fc41cf716cef6d6366 100644 (file)
--- a/youtube_dl/extractor/beeg.py
+++ b/youtube_dl/extractor/beeg.py
@@ -1,65 +1,137 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
+from ..compat import (
+    compat_chr,
+    compat_ord,
+    compat_urllib_parse_unquote,
+)
+from ..utils import (
+    int_or_none,
+    parse_iso8601,
+    urljoin,
+)
 
 
 class BeegIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?beeg\.com/(?P<id>\d+)'
     _TEST = {
         'url': 'http://beeg.com/5416503',
-        'md5': '1bff67111adb785c51d1b42959ec10e5',
+        'md5': 'a1a1b1a8bc70a89e49ccfd113aed0820',
         'info_dict': {
             'id': '5416503',
             'ext': 'mp4',
             'title': 'Sultry Striptease',
-            'description': 'md5:6db3c6177972822aaba18652ff59c773',
-            'categories': list,  # NSFW
-            'thumbnail': 're:https?://.*\.jpg$',
+            'description': 'md5:d22219c09da287c14bed3d6c37ce4bc2',
+            'timestamp': 1391813355,
+            'upload_date': '20140207',
+            'duration': 383,
+            'tags': list,
             'age_limit': 18,
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
 
-        quality_arr = self._search_regex(
-            r'(?s)var\s+qualityArr\s*=\s*{\s*(.+?)\s*}', webpage, 'quality formats')
+        cpl_url = self._search_regex(
+            r'<script[^>]+src=(["\'])(?P<url>(?:/static|(?:https?:)?//static\.beeg\.com)/cpl/\d+\.js.*?)\1',
+            webpage, 'cpl', default=None, group='url')
+
+        cpl_url = urljoin(url, cpl_url)
+
+        beeg_version, beeg_salt = [None] * 2
+
+        if cpl_url:
+            cpl = self._download_webpage(
+                self._proto_relative_url(cpl_url), video_id,
+                'Downloading cpl JS', fatal=False)
+            if cpl:
+                beeg_version = int_or_none(self._search_regex(
+                    r'beeg_version\s*=\s*([^\b]+)', cpl,
+                    'beeg version', default=None)) or self._search_regex(
+                    r'/(\d+)\.js', cpl_url, 'beeg version', default=None)
+                beeg_salt = self._search_regex(
+                    r'beeg_salt\s*=\s*(["\'])(?P<beeg_salt>.+?)\1', cpl, 'beeg salt',
+                    default=None, group='beeg_salt')
+
+        beeg_version = beeg_version or '2185'
+        beeg_salt = beeg_salt or 'pmweAkq8lAYKdfWcFCUj0yoVgoPlinamH5UE1CB3H'
 
-        formats = [{
-            'url': fmt[1],
-            'format_id': fmt[0],
-            'height': int(fmt[0][:-1]),
-        } for fmt in re.findall(r"'([^']+)'\s*:\s*'([^']+)'", quality_arr)]
+        for api_path in ('', 'api.'):
+            video = self._download_json(
+                'https://%sbeeg.com/api/v6/%s/video/%s'
+                % (api_path, beeg_version, video_id), video_id,
+                fatal=api_path == 'api.')
+            if video:
+                break
 
+        def split(o, e):
+            def cut(s, x):
+                n.append(s[:x])
+                return s[x:]
+            n = []
+            r = len(o) % e
+            if r > 0:
+                o = cut(o, r)
+            while len(o) > e:
+                o = cut(o, e)
+            n.append(o)
+            return n
+
+        def decrypt_key(key):
+            # Reverse engineered from http://static.beeg.com/cpl/1738.js
+            a = beeg_salt
+            e = compat_urllib_parse_unquote(key)
+            o = ''.join([
+                compat_chr(compat_ord(e[n]) - compat_ord(a[n % len(a)]) % 21)
+                for n in range(len(e))])
+            return ''.join(split(o, 3)[::-1])
+
+        def decrypt_url(encrypted_url):
+            encrypted_url = self._proto_relative_url(
+                encrypted_url.replace('{DATA_MARKERS}', ''), 'https:')
+            key = self._search_regex(
+                r'/key=(.*?)%2Cend=', encrypted_url, 'key', default=None)
+            if not key:
+                return encrypted_url
+            return encrypted_url.replace(key, decrypt_key(key))
+
+        formats = []
+        for format_id, video_url in video.items():
+            if not video_url:
+                continue
+            height = self._search_regex(
+                r'^(\d+)[pP]$', format_id, 'height', default=None)
+            if not height:
+                continue
+            formats.append({
+                'url': decrypt_url(video_url),
+                'format_id': format_id,
+                'height': int(height),
+            })
         self._sort_formats(formats)
 
-        title = self._html_search_regex(
-            r'<title>([^<]+)\s*-\s*beeg\.?</title>', webpage, 'title')
+        title = video['title']
+        video_id = video.get('id') or video_id
+        display_id = video.get('code')
+        description = video.get('desc')
 
-        description = self._html_search_regex(
-            r'<meta name="description" content="([^"]*)"',
-            webpage, 'description', fatal=False)
-        thumbnail = self._html_search_regex(
-            r'\'previewer.url\'\s*:\s*"([^"]*)"',
-            webpage, 'thumbnail', fatal=False)
+        timestamp = parse_iso8601(video.get('date'), ' ')
+        duration = int_or_none(video.get('duration'))
 
-        categories_str = self._html_search_regex(
-            r'<meta name="keywords" content="([^"]+)"', webpage, 'categories', fatal=False)
-        categories = (
-            None if categories_str is None
-            else categories_str.split(','))
+        tags = [tag.strip() for tag in video['tags'].split(',')] if video.get('tags') else None
 
         return {
             'id': video_id,
+            'display_id': display_id,
             'title': title,
             'description': description,
-            'thumbnail': thumbnail,
-            'categories': categories,
+            'timestamp': timestamp,
+            'duration': duration,
+            'tags': tags,
             'formats': formats,
-            'age_limit': 18,
+            'age_limit': self._rta_search(webpage),
         }