Switch codebase to use sanitized_Request instead of

[youtube-dl] / youtube_dl / extractor / flickr.py

diff --git a/youtube_dl/extractor/flickr.py b/youtube_dl/extractor/flickr.py
index 5b0bc9d219dab5a997c165068732d88308910882..91cd46e76cbacaf2dac242d14987b2ae777d6995 100644 (file)
--- a/youtube_dl/extractor/flickr.py
+++ b/youtube_dl/extractor/flickr.py
@@ -5,7 +5,8 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
-    unescapeHTML,
+    find_xpath_attr,
+    sanitized_Request,
 )
 
 
@@ -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
-        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, '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, '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:
+        stream = second_xml.find('.//STREAM')
+        if stream is None:
             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,