Merge tag 'upstream/2016.06.25'

[youtubedl] / youtube_dl / extractor / xtube.py

diff --git a/youtube_dl/extractor/xtube.py b/youtube_dl/extractor/xtube.py
index b293e2665b81b9a486bff2ec91b410da4a6d9998..4075b8a4f8a705cf29aa1430656146350a8d07aa 100644 (file)
--- a/youtube_dl/extractor/xtube.py
+++ b/youtube_dl/extractor/xtube.py
@@ -1,114 +1,132 @@
 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 

+import itertools

 import re
 import re

-import json

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

+from ..compat import compat_urllib_parse_unquote

 from ..utils import (
 from ..utils import (

-    compat_urllib_request,
-    parse_duration,
+    int_or_none,
+    orderedSet,
+    sanitized_Request,

     str_to_int,
 )
 
 
 class XTubeIE(InfoExtractor):
     str_to_int,
 )
 
 
 class XTubeIE(InfoExtractor):

-    _VALID_URL = r'https?://(?:www\.)?(?P<url>xtube\.com/watch\.php\?v=(?P<videoid>[^/?&]+))'
-    _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': {
             'id': 'kVTUy_G222_',
             'ext': 'mp4',
             'title': 'strange erotica',
         'url': 'http://www.xtube.com/watch.php?v=kVTUy_G222_',
         'md5': '092fbdd3cbe292c920ef6fc6a8a9cdab',
         'info_dict': {
             'id': 'kVTUy_G222_',
             'ext': 'mp4',
             'title': 'strange erotica',

-            'description': 'surreal gay themed erotica...almost an ET kind of thing',
+            'description': 'contains:an ET kind of thing',

             'uploader': 'greenshowers',
             'duration': 450,
             'age_limit': 18,
         }
             'uploader': 'greenshowers',
             '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):
         mobj = re.match(self._VALID_URL, url)
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

-        video_id = mobj.group('videoid')
-        url = 'http://www.' + mobj.group('url')
-
-        req = compat_urllib_request.Request(url)
-        req.add_header('Cookie', 'age_verified=1')
-        webpage = self._download_webpage(req, video_id)
-
-        video_title = self._html_search_regex(r'<p class="title">([^<]+)', webpage, 'title')
-        video_uploader = self._html_search_regex(
-            r'so_s\.addVariable\("owner_u", "([^"]+)', 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 = self._html_search_regex(
-            r'<span class="bold">Views:</span> ([\d,\.]+)</p>', webpage, 'view count', fatal=False)
-        if view_count:
-            view_count = str_to_int(view_count)
-        comment_count = self._html_search_regex(
-            r'<div id="commentBar">([\d,\.]+) Comments</div>', webpage, 'comment count', fatal=False)
-        if comment_count:
-            comment_count = str_to_int(comment_count)
-
-        player_quality_option = json.loads(self._html_search_regex(
-            r'playerQualityOption = ({.+?});', webpage, 'player quality option'))
-
-        QUALITIES = ['3gp', 'mp4_normal', 'mp4_high', 'flv', 'mp4_ultra', 'mp4_720', 'mp4_1080']
-        formats = [
-            {
-                'url': furl,
-                'format_id': format_id,
-                'preference': QUALITIES.index(format_id) if format_id in QUALITIES else -1,
-            } for format_id, furl in player_quality_option.items()
-        ]
-        self._sort_formats(formats)
+        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)
+
+        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'))
+
+        uploader = self._search_regex(
+            r'<input[^>]+name="contentOwnerId"[^>]+value="([^"]+)"',
+            webpage, 'uploader', fatal=False)
+        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'>Comments? \(([\d,\.]+)\)<',
+            webpage, 'comment count', fatal=False))

 
         return {
             'id': video_id,
 
         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,
             'duration': duration,
             'view_count': view_count,
             'comment_count': comment_count,

-            'formats': formats,

             'age_limit': 18,
         }
 
             'age_limit': 18,
         }
 

+

 class XTubeUserIE(InfoExtractor):
     IE_DESC = 'XTube user profile'
 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/profile/greenshowers-4056496',
+        'info_dict': {
+            'id': 'greenshowers-4056496',
+            'age_limit': 18,
+        },
+        'playlist_mincount': 155,
+    }

 
     def _real_extract(self, url):
 
     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,
-            '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