Imported Debian patch 2015.11.27.1-1.1

[youtubedl] / youtube_dl / extractor / flickr.py

diff --git a/youtube_dl/extractor/flickr.py b/youtube_dl/extractor/flickr.py
index 791d5b61dd27efcaf0aa94d3bce2e1744aa72678..91cd46e76cbacaf2dac242d14987b2ae777d6995 100644 (file)
--- a/youtube_dl/extractor/flickr.py
+++ b/youtube_dl/extractor/flickr.py
@@ -1,15 +1,28 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
-    unescapeHTML,
+    find_xpath_attr,
+    sanitized_Request,
 )
 
 
 class FlickrIE(InfoExtractor):
-    """Information Extractor for Flickr videos"""
-    _VALID_URL = r'(?:https?://)?(?:www\.)?flickr\.com/photos/(?P<uploader_id>[\w\-_@]+)/(?P<id>\d+).*'
+    _VALID_URL = r'https?://(?:www\.|secure\.)?flickr\.com/photos/(?P<uploader_id>[\w\-_@]+)/(?P<id>\d+).*'
+    _TEST = {
+        'url': 'http://www.flickr.com/photos/forestwander-nature-pictures/5645318632/in/photostream/',
+        'md5': '6fdc01adbc89d72fc9c4f15b4a4ba87b',
+        'info_dict': {
+            'id': '5645318632',
+            'ext': 'mp4',
+            "description": "Waterfalls in the Springtime at Dark Hollow Waterfalls. These are located just off of Skyline Drive in Virginia. They are only about 6/10 of a mile hike but it is a pretty steep hill and a good climb back up.",
+            "uploader_id": "forestwander-nature-pictures",
+            "title": "Dark Hollow Waterfalls"
+        }
+    }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -17,41 +30,38 @@ class FlickrIE(InfoExtractor):
         video_id = mobj.group('id')
         video_uploader_id = mobj.group('uploader_id')
         webpage_url = 'http://www.flickr.com/photos/' + video_uploader_id + '/' + video_id
-        webpage = self._download_webpage(webpage_url, video_id)
+        req = sanitized_Request(webpage_url)
+        req.add_header(
+            'User-Agent',
+            # it needs a more recent version
+            'Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20150101 Firefox/38.0 (Chrome)')
+        webpage = self._download_webpage(req, video_id)
 
-        secret = self._search_regex(r"photo_secret: '(\w+)'", webpage, u'secret')
+        secret = self._search_regex(r'secret"\s*:\s*"(\w+)"', webpage, 'secret')
 
         first_url = 'https://secure.flickr.com/apps/video/video_mtl_xml.gne?v=x&photo_id=' + video_id + '&secret=' + secret + '&bitrate=700&target=_self'
-        first_xml = self._download_webpage(first_url, video_id, 'Downloading first data webpage')
+        first_xml = self._download_xml(first_url, video_id, 'Downloading first data webpage')
 
-        node_id = self._html_search_regex(r'<Item id="id">(\d+-\d+)</Item>',
-            first_xml, u'node_id')
+        node_id = find_xpath_attr(
+            first_xml, './/{http://video.yahoo.com/YEP/1.0/}Item', 'id',
+            'id').text
 
         second_url = 'https://secure.flickr.com/video_playlist.gne?node_id=' + node_id + '&tech=flash&mode=playlist&bitrate=700&secret=' + secret + '&rd=video.yahoo.com&noad=1'
-        second_xml = self._download_webpage(second_url, video_id, 'Downloading second data webpage')
+        second_xml = self._download_xml(second_url, video_id, 'Downloading second data webpage')
 
         self.report_extraction(video_id)
 
-        mobj = re.search(r'<STREAM APP="(.+?)" FULLPATH="(.+?)"', second_xml)
-        if mobj is None:
-            raise ExtractorError(u'Unable to extract video url')
-        video_url = mobj.group(1) + unescapeHTML(mobj.group(2))
-
-        video_title = self._html_search_regex(r'<meta property="og:title" content=(?:"([^"]+)"|\'([^\']+)\')',
-            webpage, u'video title')
-
-        video_description = self._html_search_regex(r'<meta property="og:description" content=(?:"([^"]+)"|\'([^\']+)\')',
-            webpage, u'description', fatal=False)
-
-        thumbnail = self._html_search_regex(r'<meta property="og:image" content=(?:"([^"]+)"|\'([^\']+)\')',
-            webpage, u'thumbnail', fatal=False)
+        stream = second_xml.find('.//STREAM')
+        if stream is None:
+            raise ExtractorError('Unable to extract video url')
+        video_url = stream.attrib['APP'] + stream.attrib['FULLPATH']
 
-        return [{
-            'id':          video_id,
-            'url':         video_url,
-            'ext':         'mp4',
-            'title':       video_title,
-            'description': video_description,
-            'thumbnail':   thumbnail,
+        return {
+            'id': video_id,
+            'url': video_url,
+            'ext': 'mp4',
+            'title': self._og_search_title(webpage),
+            'description': self._og_search_description(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
             'uploader_id': video_uploader_id,
-        }]
+        }