X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/03342304420e5daeb428ffdcc7bbd2bbfecfa61a..HEAD:/youtube_dl/extractor/xtube.py

diff --git a/youtube_dl/extractor/xtube.py b/youtube_dl/extractor/xtube.py
index 4075b8a..01b253d 100644
--- a/youtube_dl/extractor/xtube.py
+++ b/youtube_dl/extractor/xtube.py
@@ -4,17 +4,24 @@ import itertools
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_unquote
 from ..utils import (
     int_or_none,
+    js_to_json,
     orderedSet,
+    parse_duration,
     sanitized_Request,
     str_to_int,
 )
 
 
 class XTubeIE(InfoExtractor):
-    _VALID_URL = r'(?:xtube:|https?://(?:www\.)?xtube\.com/(?:watch\.php\?.*\bv=|video-watch/(?P<display_id>[^/]+)-))(?P<id>[^/?&#]+)'
+    _VALID_URL = r'''(?x)
+                        (?:
+                            xtube:|
+                            https?://(?:www\.)?xtube\.com/(?:watch\.php\?.*\bv=|video-watch/(?:embedded/)?(?P<display_id>[^/]+)-)
+                        )
+                        (?P<id>[^/?&#]+)
+                    '''
 
     _TESTS = [{
         # old URL schema
@@ -27,8 +34,26 @@ class XTubeIE(InfoExtractor):
             'description': 'contains:an ET kind of thing',
             'uploader': 'greenshowers',
             'duration': 450,
+            'view_count': int,
+            'comment_count': int,
             'age_limit': 18,
         }
+    }, {
+        # FLV videos with duplicated formats
+        'url': 'http://www.xtube.com/video-watch/A-Super-Run-Part-1-YT-9299752',
+        'md5': 'a406963eb349dd43692ec54631efd88b',
+        'info_dict': {
+            'id': '9299752',
+            'display_id': 'A-Super-Run-Part-1-YT',
+            'ext': 'flv',
+            'title': 'A Super Run - Part 1 (YT)',
+            'description': 'md5:4cc3af1aa1b0413289babc88f0d4f616',
+            'uploader': 'tshirtguy59',
+            'duration': 579,
+            'view_count': int,
+            'comment_count': int,
+            'age_limit': 18,
+        },
     }, {
         # new URL schema
         'url': 'http://www.xtube.com/video-watch/strange-erotica-625837',
@@ -36,6 +61,12 @@ class XTubeIE(InfoExtractor):
     }, {
         'url': 'xtube:625837',
         'only_matching': True,
+    }, {
+        'url': 'xtube:kVTUy_G222_',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.xtube.com/video-watch/embedded/milf-tara-and-teen-shared-and-cum-covered-extreme-bukkake-32203482?embedsize=big',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -45,29 +76,65 @@ class XTubeIE(InfoExtractor):
 
         if not display_id:
             display_id = video_id
-            url = 'http://www.xtube.com/watch.php?v=%s' % video_id
-
-        req = sanitized_Request(url)
-        req.add_header('Cookie', 'age_verified=1; cookiesAccepted=1')
-        webpage = self._download_webpage(req, display_id)
-
-        flashvars = self._parse_json(
-            self._search_regex(
-                r'xt\.playerOps\s*=\s*({.+?});', webpage, 'player ops'),
-            video_id)['flashvars']
-
-        title = flashvars.get('title') or self._search_regex(
-            r'<h1>([^<]+)</h1>', webpage, 'title')
-        video_url = compat_urllib_parse_unquote(flashvars['video_url'])
-        duration = int_or_none(flashvars.get('video_duration'))
 
+        if video_id.isdigit() and len(video_id) < 11:
+            url_pattern = 'http://www.xtube.com/video-watch/-%s'
+        else:
+            url_pattern = 'http://www.xtube.com/watch.php?v=%s'
+
+        webpage = self._download_webpage(
+            url_pattern % video_id, display_id, headers={
+                'Cookie': 'age_verified=1; cookiesAccepted=1',
+            })
+
+        title, thumbnail, duration = [None] * 3
+
+        config = self._parse_json(self._search_regex(
+            r'playerConf\s*=\s*({.+?})\s*,\s*\n', webpage, 'config',
+            default='{}'), video_id, transform_source=js_to_json, fatal=False)
+        if config:
+            config = config.get('mainRoll')
+            if isinstance(config, dict):
+                title = config.get('title')
+                thumbnail = config.get('poster')
+                duration = int_or_none(config.get('duration'))
+                sources = config.get('sources') or config.get('format')
+
+        if not isinstance(sources, dict):
+            sources = self._parse_json(self._search_regex(
+                r'(["\'])?sources\1?\s*:\s*(?P<sources>{.+?}),',
+                webpage, 'sources', group='sources'), video_id,
+                transform_source=js_to_json)
+
+        formats = []
+        for format_id, format_url in sources.items():
+            formats.append({
+                'url': format_url,
+                'format_id': format_id,
+                'height': int_or_none(format_id),
+            })
+        self._remove_duplicate_formats(formats)
+        self._sort_formats(formats)
+
+        if not title:
+            title = self._search_regex(
+                (r'<h1>\s*(?P<title>[^<]+?)\s*</h1>', r'videoTitle\s*:\s*(["\'])(?P<title>.+?)\1'),
+                webpage, 'title', group='title')
+        description = self._og_search_description(
+            webpage, default=None) or self._html_search_meta(
+            'twitter:description', webpage, default=None) or self._search_regex(
+            r'</h1>\s*<p>([^<]+)', webpage, 'description', fatal=False)
         uploader = self._search_regex(
-            r'<input[^>]+name="contentOwnerId"[^>]+value="([^"]+)"',
+            (r'<input[^>]+name="contentOwnerId"[^>]+value="([^"]+)"',
+             r'<span[^>]+class="nickname"[^>]*>([^<]+)'),
             webpage, 'uploader', fatal=False)
-        description = self._search_regex(
-            r'</h1>\s*<p>([^<]+)', webpage, 'description', fatal=False)
+        if not duration:
+            duration = parse_duration(self._search_regex(
+                r'<dt>Runtime:?</dt>\s*<dd>([^<]+)</dd>',
+                webpage, 'duration', fatal=False))
         view_count = str_to_int(self._search_regex(
-            r'<dt>Views:</dt>\s*<dd>([\d,\.]+)</dd>',
+            (r'["\']viewsCount["\'][^>]*>(\d+)\s+views',
+             r'<dt>Views:?</dt>\s*<dd>([\d,\.]+)</dd>'),
             webpage, 'view count', fatal=False))
         comment_count = str_to_int(self._html_search_regex(
             r'>Comments? \(([\d,\.]+)\)<',
@@ -76,14 +143,15 @@ class XTubeIE(InfoExtractor):
         return {
             'id': video_id,
             'display_id': display_id,
-            'url': video_url,
             'title': title,
             'description': description,
+            'thumbnail': thumbnail,
             'uploader': uploader,
             'duration': duration,
             'view_count': view_count,
             'comment_count': comment_count,
             'age_limit': 18,
+            'formats': formats,
         }
 
 
@@ -96,7 +164,7 @@ class XTubeUserIE(InfoExtractor):
             'id': 'greenshowers-4056496',
             'age_limit': 18,
         },
-        'playlist_mincount': 155,
+        'playlist_mincount': 154,
     }
 
     def _real_extract(self, url):