Merge tag 'upstream/2015.11.27.1'

[youtubedl] / youtube_dl / extractor / novamov.py

diff --git a/youtube_dl/extractor/novamov.py b/youtube_dl/extractor/novamov.py
index 48ee00da3dbc0c75a5a5b09c8d7b2faf7e0a6ceb..6163e885558859bb00abb4be7d40da9852132c08 100644 (file)
--- a/youtube_dl/extractor/novamov.py
+++ b/youtube_dl/extractor/novamov.py
@@ -3,54 +3,86 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_urlparse
 from ..utils import (
     ExtractorError,
-    compat_urlparse
+    NO_DEFAULT,
+    encode_dict,
+    sanitized_Request,
+    urlencode_postdata,
 )
 
 
-class NovamovIE(InfoExtractor):
-    _VALID_URL = r'http://(?:(?:www\.)?novamov\.com/video/|(?:(?:embed|www)\.)novamov\.com/embed\.php\?v=)(?P<videoid>[a-z\d]{13})'
+class NovaMovIE(InfoExtractor):
+    IE_NAME = 'novamov'
+    IE_DESC = 'NovaMov'
+
+    _VALID_URL_TEMPLATE = r'http://(?:(?:www\.)?%(host)s/(?:file|video)/|(?:(?:embed|www)\.)%(host)s/embed\.php\?(?:.*?&)?v=)(?P<id>[a-z\d]{13})'
+    _VALID_URL = _VALID_URL_TEMPLATE % {'host': 'novamov\.com'}
+
+    _HOST = 'www.novamov.com'
+
+    _FILE_DELETED_REGEX = r'This file no longer exists on our servers!</h2>'
+    _FILEKEY_REGEX = r'flashvars\.filekey="(?P<filekey>[^"]+)";'
+    _TITLE_REGEX = r'(?s)<div class="v_tab blockborder rounded5" id="v_tab1">\s*<h3>([^<]+)</h3>'
+    _DESCRIPTION_REGEX = r'(?s)<div class="v_tab blockborder rounded5" id="v_tab1">\s*<h3>[^<]+</h3><p>([^<]+)</p>'
 
     _TEST = {
         'url': 'http://www.novamov.com/video/4rurhn9x446jj',
-        'file': '4rurhn9x446jj.flv',
         'md5': '7205f346a52bbeba427603ba10d4b935',
         'info_dict': {
+            'id': '4rurhn9x446jj',
+            'ext': 'flv',
             'title': 'search engine optimization',
             'description': 'search engine optimization is used to rank the web page in the google search engine'
-        }
+        },
+        'skip': '"Invalid token" errors abound (in web interface as well as youtube-dl, there is nothing we can do about it.)'
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('videoid')
+        video_id = self._match_id(url)
+
+        url = 'http://%s/video/%s' % (self._HOST, video_id)
+
+        webpage = self._download_webpage(
+            url, video_id, 'Downloading video page')
+
+        if re.search(self._FILE_DELETED_REGEX, webpage) is not None:
+            raise ExtractorError('Video %s does not exist' % video_id, expected=True)
 
-        page = self._download_webpage('http://www.novamov.com/video/%s' % video_id,
-                                      video_id, 'Downloading video page')
+        def extract_filekey(default=NO_DEFAULT):
+            return self._search_regex(
+                self._FILEKEY_REGEX, webpage, 'filekey', default=default)
 
-        if re.search(r'This file no longer exists on our servers!</h2>', page) is not None:
-            raise ExtractorError(u'Video %s does not exist' % video_id, expected=True)
+        filekey = extract_filekey(default=None)
 
-        filekey = self._search_regex(
-            r'flashvars\.filekey="(?P<filekey>[^"]+)";', page, 'filekey')
+        if not filekey:
+            fields = self._hidden_inputs(webpage)
+            post_url = self._search_regex(
+                r'<form[^>]+action=(["\'])(?P<url>.+?)\1', webpage,
+                'post url', default=url, group='url')
+            if not post_url.startswith('http'):
+                post_url = compat_urlparse.urljoin(url, post_url)
+            request = sanitized_Request(
+                post_url, urlencode_postdata(encode_dict(fields)))
+            request.add_header('Content-Type', 'application/x-www-form-urlencoded')
+            request.add_header('Referer', post_url)
+            webpage = self._download_webpage(
+                request, video_id, 'Downloading continue to the video page')
 
-        title = self._html_search_regex(
-            r'(?s)<div class="v_tab blockborder rounded5" id="v_tab1">\s*<h3>([^<]+)</h3>',
-            page, 'title', fatal=False)
+        filekey = extract_filekey()
 
-        description = self._html_search_regex(
-            r'(?s)<div class="v_tab blockborder rounded5" id="v_tab1">\s*<h3>[^<]+</h3><p>([^<]+)</p>',
-            page, 'description', fatal=False)
+        title = self._html_search_regex(self._TITLE_REGEX, webpage, 'title', fatal=False)
+        description = self._html_search_regex(self._DESCRIPTION_REGEX, webpage, 'description', default='', fatal=False)
 
         api_response = self._download_webpage(
-            'http://www.novamov.com/api/player.api.php?key=%s&file=%s' % (filekey, video_id),
-            video_id, 'Downloading video api response')
+            'http://%s/api/player.api.php?key=%s&file=%s' % (self._HOST, filekey, video_id), video_id,
+            'Downloading video api response')
 
         response = compat_urlparse.parse_qs(api_response)
 
         if 'error_msg' in response:
-            raise ExtractorError('novamov returned error: %s' % response['error_msg'][0], expected=True)
+            raise ExtractorError('%s returned error: %s' % (self.IE_NAME, response['error_msg'][0]), expected=True)
 
         video_url = response['url'][0]