Register new version in the changelog.

[youtubedl] / youtube_dl / extractor / flickr.py

diff --git a/youtube_dl/extractor/flickr.py b/youtube_dl/extractor/flickr.py
index e09982e88b913676a2f8c75946e79c82502bf650..2fe76d661bb432580cd2bd3f48c85035a4b6d7d9 100644 (file)
--- a/youtube_dl/extractor/flickr.py
+++ b/youtube_dl/extractor/flickr.py
@@ -3,9 +3,10 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

+from ..compat import compat_urllib_request

 from ..utils import (
     ExtractorError,
 from ..utils import (
     ExtractorError,

-    unescapeHTML,
+    find_xpath_attr,

 )
 
 
 )
 
 


@@ -17,8 +18,8 @@ class FlickrIE(InfoExtractor):
         'info_dict': {
             'id': '5645318632',
             'ext': 'mp4',
         '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", 
+            "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"
         }
     }
             "title": "Dark Hollow Waterfalls"
         }
     }


@@ -29,25 +30,31 @@ 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
         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 = compat_urllib_request.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, '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_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, '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_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)
 
 
         self.report_extraction(video_id)
 

-        mobj = re.search(r'<STREAM APP="(.+?)" FULLPATH="(.+?)"', second_xml)
-        if mobj is None:
+        stream = second_xml.find('.//STREAM')
+        if stream is None:

             raise ExtractorError('Unable to extract video url')
             raise ExtractorError('Unable to extract video url')

-        video_url = mobj.group(1) + unescapeHTML(mobj.group(2))
+        video_url = stream.attrib['APP'] + stream.attrib['FULLPATH']

 
         return {
             'id': video_id,
 
         return {
             'id': video_id,