X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/af478477605bdf3f5d57562035885cfee905f379..05a5c38aac7764a2b03015ee6f358cbb5a5e79a2:/youtube_dl/extractor/xtube.py?ds=sidebyside

diff --git a/youtube_dl/extractor/xtube.py b/youtube_dl/extractor/xtube.py
index 9826199..c6c0b32 100644
--- a/youtube_dl/extractor/xtube.py
+++ b/youtube_dl/extractor/xtube.py
@@ -1,56 +1,180 @@
 from __future__ import unicode_literals
 
-import os
+import itertools
 import re
 
 from .common import InfoExtractor
 from ..utils import (
-    compat_urllib_parse_urlparse,
-    compat_urllib_request,
+    int_or_none,
+    js_to_json,
+    orderedSet,
+    parse_duration,
+    sanitized_Request,
+    str_to_int,
 )
 
+
 class XTubeIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>xtube\.com/watch\.php\?v=(?P<videoid>[^/?&]+))'
-    _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_',
-        'file': 'kVTUy_G222_.mp4',
         'md5': '092fbdd3cbe292c920ef6fc6a8a9cdab',
         'info_dict': {
-            "title": "strange erotica",
-            "description": "surreal gay themed erotica...almost an ET kind of thing",
-            "uploader": "greenshowers",
-            "age_limit": 18,
+            'id': 'kVTUy_G222_',
+            'ext': 'mp4',
+            'title': 'strange erotica',
+            '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: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):
         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'<div class="p_5px[^>]*>([^<]+)', 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="video_description">([^<]+)', webpage, 'description', fatal=False)
-        video_url= self._html_search_regex(r'var videoMp4 = "([^"]+)', webpage, 'video_url').replace('\\/', '/')
-        path = compat_urllib_parse_urlparse(video_url).path
-        extension = os.path.splitext(path)[1][1:]
-        format = path.split('/')[5].split('_')[:2]
-        format[0] += 'p'
-        format[1] += 'k'
-        format = "-".join(format)
+        video_id = mobj.group('id')
+        display_id = mobj.group('display_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)
+
+        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)
+        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>',
+            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,
-            'title': video_title,
-            'uploader': video_uploader,
-            'description': video_description,
-            'url': video_url,
-            'ext': extension,
-            'format': format,
-            'format_id': format,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+            'uploader': uploader,
+            'duration': duration,
+            'view_count': view_count,
+            'comment_count': comment_count,
             'age_limit': 18,
+            'formats': formats,
         }
+
+
+class XTubeUserIE(InfoExtractor):
+    IE_DESC = 'XTube user profile'
+    _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):
+        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