d/control: Add build-dependency on dh-python as per warning from dh_python2.

[youtubedl] / youtube_dl / extractor / xtube.py

diff --git a/youtube_dl/extractor/xtube.py b/youtube_dl/extractor/xtube.py
index 03ad88bededd60a7d8462bb85e5a8cb23db381b2..1644f53c876329f053406be3d3dc1aa463cddc1b 100644 (file)
--- a/youtube_dl/extractor/xtube.py
+++ b/youtube_dl/extractor/xtube.py
@@ -1,55 +1,134 @@
-import os
+from __future__ import unicode_literals
+

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

-from ..utils import (
-    compat_urllib_parse_urlparse,
+from ..compat import (

     compat_urllib_request,
     compat_urllib_parse,
 )
     compat_urllib_request,
     compat_urllib_parse,
 )

+from ..utils import (
+    parse_duration,
+    str_to_int,
+)
+

 
 class XTubeIE(InfoExtractor):
 
 class XTubeIE(InfoExtractor):

-    _VALID_URL = r'^(?:https?://)?(?:www\.)?(?P<url>xtube\.com/watch\.php\?v=(?P<videoid>[^/?&]+))'
+    _VALID_URL = r'https?://(?:www\.)?(?P<url>xtube\.com/watch\.php\?v=(?P<id>[^/?&#]+))'

     _TEST = {
     _TEST = {

-        u'url': u'http://www.xtube.com/watch.php?v=kVTUy_G222_',
-        u'file': u'kVTUy_G222_.mp4',
-        u'md5': u'092fbdd3cbe292c920ef6fc6a8a9cdab',
-        u'info_dict': {
-            u"title": u"strange erotica",
-            u"description": u"surreal gay themed erotica...almost an ET kind of thing",
-            u"uploader": u"greenshowers",
-            u"age_limit": 18,
+        'url': 'http://www.xtube.com/watch.php?v=kVTUy_G222_',
+        'md5': '092fbdd3cbe292c920ef6fc6a8a9cdab',
+        'info_dict': {
+            'id': 'kVTUy_G222_',
+            'ext': 'mp4',
+            'title': 'strange erotica',
+            'description': 'contains:an ET kind of thing',
+            'uploader': 'greenshowers',
+            'duration': 450,
+            'age_limit': 18,

         }
     }
 
     def _real_extract(self, url):
         }
     }
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('videoid')
-        url = 'http://www.' + mobj.group('url')
+        video_id = self._match_id(url)

 
         req = compat_urllib_request.Request(url)
         req.add_header('Cookie', 'age_verified=1')
         webpage = self._download_webpage(req, video_id)
 
 
         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, u'title')
-        video_uploader = self._html_search_regex(r'so_s\.addVariable\("owner_u", "([^"]+)', webpage, u'uploader', fatal=False)
-        video_description = self._html_search_regex(r'<p class="video_description">([^<]+)', webpage, u'description', default=None)
-        video_url= self._html_search_regex(r'var videoMp4 = "([^"]+)', webpage, u'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_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=([^"]+)'],
+            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>',
+            webpage, 'view count', fatal=False))
+        comment_count = str_to_int(self._html_search_regex(
+            r'<div id="commentBar">([\d,\.]+) Comments</div>',
+            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,
 
         return {
             'id': video_id,
             'title': video_title,
             'uploader': video_uploader,
             'description': video_description,

-            'url': video_url,
-            'ext': extension,
-            'format': format,
-            'format_id': format,
+            '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>[^&#]+)(?:$|[&#])'
+    _TEST = {
+        'url': 'http://www.xtube.com/community/profile.php?user=greenshowers',
+        'info_dict': {
+            'id': 'greenshowers',
+            '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,
             'age_limit': 18,

+            'entries': [{
+                '_type': 'url',
+                'url': eurl,
+                'ie_key': 'XTube',
+            } for eurl in urls]

         }
         }