X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/38bb9b1b0a044cabaf5691553815e334cd2e9213..e5a601f21d25b8806d04be9671c8772e331e3bda:/youtube_dl/extractor/xtube.py

diff --git a/youtube_dl/extractor/xtube.py b/youtube_dl/extractor/xtube.py
index 779e4f4..4075b8a 100644
--- a/youtube_dl/extractor/xtube.py
+++ b/youtube_dl/extractor/xtube.py
@@ -1,21 +1,23 @@
 from __future__ import unicode_literals
 
+import itertools
 import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_urllib_request,
-    compat_urllib_parse_unquote,
-)
+from ..compat import compat_urllib_parse_unquote
 from ..utils import (
-    parse_duration,
+    int_or_none,
+    orderedSet,
+    sanitized_Request,
     str_to_int,
 )
 
 
 class XTubeIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?(?P<url>xtube\.com/watch\.php\?v=(?P<id>[^/?&#]+))'
-    _TEST = {
+    _VALID_URL = r'(?:xtube:|https?://(?:www\.)?xtube\.com/(?:watch\.php\?.*\bv=|video-watch/(?P<display_id>[^/]+)-))(?P<id>[^/?&#]+)'
+
+    _TESTS = [{
+        # old URL schema
         'url': 'http://www.xtube.com/watch.php?v=kVTUy_G222_',
         'md5': '092fbdd3cbe292c920ef6fc6a8a9cdab',
         'info_dict': {
@@ -27,108 +29,104 @@ class XTubeIE(InfoExtractor):
             'duration': 450,
             'age_limit': 18,
         }
-    }
+    }, {
+        # new URL schema
+        'url': 'http://www.xtube.com/video-watch/strange-erotica-625837',
+        'only_matching': True,
+    }, {
+        'url': 'xtube:625837',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
+
+        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)
 
-        req = compat_urllib_request.Request(url)
-        req.add_header('Cookie', 'age_verified=1')
-        webpage = self._download_webpage(req, video_id)
+        flashvars = self._parse_json(
+            self._search_regex(
+                r'xt\.playerOps\s*=\s*({.+?});', webpage, 'player ops'),
+            video_id)['flashvars']
 
-        video_title = self._html_search_regex(
-            r'<p class="title">([^<]+)', webpage, 'title')
-        video_uploader = self._html_search_regex(
-            [r"var\s+contentOwnerId\s*=\s*'([^']+)",
-             r'By:\s*<a href="/community/profile\.php\?user=([^"]+)'],
+        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'))
+
+        uploader = self._search_regex(
+            r'<input[^>]+name="contentOwnerId"[^>]+value="([^"]+)"',
             webpage, 'uploader', fatal=False)
-        video_description = self._html_search_regex(
-            r'<p class="fieldsDesc">([^<]+)',
-            webpage, 'description', fatal=False)
-        duration = parse_duration(self._html_search_regex(
-            r'<span class="bold">Runtime:</span> ([^<]+)</p>',
-            webpage, 'duration', fatal=False))
-        view_count = str_to_int(self._html_search_regex(
-            r'<span class="bold">Views:</span> ([\d,\.]+)</p>',
+        description = self._search_regex(
+            r'</h1>\s*<p>([^<]+)', webpage, 'description', fatal=False)
+        view_count = str_to_int(self._search_regex(
+            r'<dt>Views:</dt>\s*<dd>([\d,\.]+)</dd>',
             webpage, 'view count', fatal=False))
         comment_count = str_to_int(self._html_search_regex(
-            r'<div id="commentBar">([\d,\.]+) Comments</div>',
+            r'>Comments? \(([\d,\.]+)\)<',
             webpage, 'comment count', fatal=False))
 
-        formats = []
-        for format_id, video_url in re.findall(
-                r'flashvars\.quality_(.+?)\s*=\s*"([^"]+)"', webpage):
-            fmt = {
-                'url': compat_urllib_parse_unquote(video_url),
-                'format_id': format_id,
-            }
-            m = re.search(r'^(?P<height>\d+)[pP]', format_id)
-            if m:
-                fmt['height'] = int(m.group('height'))
-            formats.append(fmt)
-
-        if not formats:
-            video_url = compat_urllib_parse_unquote(self._search_regex(
-                r'flashvars\.video_url\s*=\s*"([^"]+)"',
-                webpage, 'video URL'))
-            formats.append({'url': video_url})
-
-        self._sort_formats(formats)
-
         return {
             'id': video_id,
-            'title': video_title,
-            'uploader': video_uploader,
-            'description': video_description,
+            'display_id': display_id,
+            'url': video_url,
+            'title': title,
+            'description': description,
+            'uploader': uploader,
             'duration': duration,
             'view_count': view_count,
             'comment_count': comment_count,
-            'formats': formats,
             'age_limit': 18,
         }
 
 
 class XTubeUserIE(InfoExtractor):
     IE_DESC = 'XTube user profile'
-    _VALID_URL = r'https?://(?:www\.)?xtube\.com/community/profile\.php\?(.*?)user=(?P<username>[^&#]+)(?:$|[&#])'
+    _VALID_URL = r'https?://(?:www\.)?xtube\.com/profile/(?P<id>[^/]+-\d+)'
     _TEST = {
-        'url': 'http://www.xtube.com/community/profile.php?user=greenshowers',
+        'url': 'http://www.xtube.com/profile/greenshowers-4056496',
         'info_dict': {
-            'id': 'greenshowers',
+            'id': 'greenshowers-4056496',
             'age_limit': 18,
         },
         'playlist_mincount': 155,
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        username = mobj.group('username')
-
-        profile_page = self._download_webpage(
-            url, username, note='Retrieving profile page')
-
-        video_count = int(self._search_regex(
-            r'<strong>%s\'s Videos \(([0-9]+)\)</strong>' % username, profile_page,
-            'video count'))
-
-        PAGE_SIZE = 25
-        urls = []
-        page_count = (video_count + PAGE_SIZE + 1) // PAGE_SIZE
-        for n in range(1, page_count + 1):
-            lpage_url = 'http://www.xtube.com/user_videos.php?page=%d&u=%s' % (n, username)
-            lpage = self._download_webpage(
-                lpage_url, username,
-                note='Downloading page %d/%d' % (n, page_count))
-            urls.extend(
-                re.findall(r'addthis:url="([^"]+)"', lpage))
-
-        return {
-            '_type': 'playlist',
-            'id': username,
-            'age_limit': 18,
-            'entries': [{
-                '_type': 'url',
-                'url': eurl,
-                'ie_key': 'XTube',
-            } for eurl in urls]
-        }
+        user_id = self._match_id(url)
+
+        entries = []
+        for pagenum in itertools.count(1):
+            request = sanitized_Request(
+                'http://www.xtube.com/profile/%s/videos/%d' % (user_id, pagenum),
+                headers={
+                    'Cookie': 'popunder=4',
+                    'X-Requested-With': 'XMLHttpRequest',
+                    'Referer': url,
+                })
+
+            page = self._download_json(
+                request, user_id, 'Downloading videos JSON page %d' % pagenum)
+
+            html = page.get('html')
+            if not html:
+                break
+
+            for video_id in orderedSet([video_id for _, video_id in re.findall(
+                    r'data-plid=(["\'])(.+?)\1', html)]):
+                entries.append(self.url_result('xtube:%s' % video_id, XTubeIE.ie_key()))
+
+            page_count = int_or_none(page.get('pageCount'))
+            if not page_count or pagenum == page_count:
+                break
+
+        playlist = self.playlist_result(entries, user_id)
+        playlist['age_limit'] = 18
+        return playlist