Update changelog with bugs being fixed.

[youtubedl] / youtube_dl / extractor / xtube.py

diff --git a/youtube_dl/extractor/xtube.py b/youtube_dl/extractor/xtube.py
index 779e4f46a1dd5315c6a9be3dad09e65c07a205b2..c6c0b3291c8320064fa0a7529be5b5d78f14461c 100644 (file)
--- a/youtube_dl/extractor/xtube.py
+++ b/youtube_dl/extractor/xtube.py
@@ -1,21 +1,30 @@
 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 

+import itertools

 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

-from ..compat import (
-    compat_urllib_request,
-    compat_urllib_parse_unquote,
-)

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

+    int_or_none,
+    js_to_json,
+    orderedSet,

     parse_duration,
     parse_duration,

+    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<id>[^/?&#]+))'
-    _TEST = {
+    _VALID_URL = r'''(?x)
+                        (?:
+                            xtube:|
+                            https?://(?:www\.)?xtube\.com/(?:watch\.php\?.*\bv=|video-watch/(?:embedded/)?(?P<display_id>[^/]+)-)
+                        )
+                        (?P<id>[^/?&#]+)
+                    '''
+
+    _TESTS = [{
+        # old URL schema

         'url': 'http://www.xtube.com/watch.php?v=kVTUy_G222_',
         'md5': '092fbdd3cbe292c920ef6fc6a8a9cdab',
         'info_dict': {
         'url': 'http://www.xtube.com/watch.php?v=kVTUy_G222_',
         'md5': '092fbdd3cbe292c920ef6fc6a8a9cdab',
         'info_dict': {


@@ -25,110 +34,147 @@ class XTubeIE(InfoExtractor):
             'description': 'contains:an ET kind of thing',
             'uploader': 'greenshowers',
             'duration': 450,
             'description': 'contains:an ET kind of thing',
             'uploader': 'greenshowers',
             'duration': 450,

+            'view_count': int,
+            'comment_count': int,

             'age_limit': 18,
         }
             '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:ca0d47afff4a9b2942e4b41aa970fd93',
+            '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',
+        'only_matching': True,
+    }, {
+        '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):
 
     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')

 
 

-        req = compat_urllib_request.Request(url)
-        req.add_header('Cookie', 'age_verified=1')
-        webpage = self._download_webpage(req, video_id)
+        if not display_id:
+            display_id = video_id
+
+        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',
+            })
+
+        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)

 
 

-        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 = self._search_regex(
+            (r'<h1>\s*(?P<title>[^<]+?)\s*</h1>', r'videoTitle\s*:\s*(["\'])(?P<title>.+?)\1'),
+            webpage, 'title', group='title')
+        description = self._search_regex(
+            r'</h1>\s*<p>([^<]+)', webpage, 'description', fatal=False)
+        uploader = self._search_regex(
+            (r'<input[^>]+name="contentOwnerId"[^>]+value="([^"]+)"',
+             r'<span[^>]+class="nickname"[^>]*>([^<]+)'),

             webpage, 'uploader', fatal=False)
             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>',
+        duration = parse_duration(self._search_regex(
+            r'<dt>Runtime:?</dt>\s*<dd>([^<]+)</dd>',

             webpage, 'duration', fatal=False))
             webpage, 'duration', fatal=False))

-        view_count = str_to_int(self._html_search_regex(
-            r'<span class="bold">Views:</span> ([\d,\.]+)</p>',
+        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(
             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))
 
             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,
         return {
             'id': video_id,

-            'title': video_title,
-            'uploader': video_uploader,
-            'description': video_description,
+            'display_id': display_id,
+            '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,

+            'formats': formats,

         }
 
 
 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 = {
     _TEST = {

-        'url': 'http://www.xtube.com/community/profile.php?user=greenshowers',
+        'url': 'http://www.xtube.com/profile/greenshowers-4056496',

         'info_dict': {
         'info_dict': {

-            'id': 'greenshowers',
+            'id': 'greenshowers-4056496',

             'age_limit': 18,
         },
         'playlist_mincount': 155,
     }
 
     def _real_extract(self, url):
             '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