add support for Flickr videos - closes #261

diff --git a/test/tests.json b/test/tests.json
index 1ff9ad7ee00a267076cd939dd1de4cfd59cc6be7..eed0628138e22ceb4ff0483ec744f13c183f8bb2 100644 (file)
--- a/test/tests.json
+++ b/test/tests.json
@@ -461,5 +461,16 @@
       "title": "Chicken.",
       "uploader": "Jack Dorsey"
     }
+  },
+  {
+    "name": "Flickr",
+    "url": "http://www.flickr.com/photos/forestwander-nature-pictures/5645318632/in/photostream/",
+    "file": "5645318632.mp4",
+    "md5": "6fdc01adbc89d72fc9c4f15b4a4ba87b",
+    "info_dict":{
+      "title": "Dark Hollow Waterfalls",
+      "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."
+    }
   }
 ]

diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index c38592e794ab8253e9707009138e0ad87fd4705b..fe47e5254c5755ac9a800edfada209754bb27946 100755 (executable)
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -4279,6 +4279,69 @@ class VineIE(InfoExtractor):
             'uploader':  uploader,
         }]
 
+class FlickrIE(InfoExtractor):
+    """Information Extractor for Flickr videos"""
+    _VALID_URL = r'(?:https?://)?(?:www\.)?flickr\.com/photos/(?P<uploader_id>[\w\-]+)/(?P<id>\d+).*'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+
+        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)
+
+        self.report_extraction(video_id)
+        
+        mobj = re.search(r"photo_secret: '(\w+)'", webpage)
+        if mobj is None:
+            raise ExtractorError(u'Unable to extract video secret')
+        secret = mobj.group(1)
+
+        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)
+
+        mobj = re.search(r'<Item id="id">(\d+-\d+)</Item>', first_xml)
+        if mobj is None:
+            raise ExtractorError(u'Unable to extract node_id')
+        node_id = mobj.group(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)
+
+        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))
+
+        mobj = re.search(r'<meta property="og:title" content=(?:"([^"]+)"|\'([^\']+)\')', webpage)
+        if mobj is None:
+            raise ExtractorError(u'Unable to extract title')
+        video_title = mobj.group(1) or mobj.group(2)
+
+        mobj = re.search(r'<meta property="og:description" content=(?:"([^"]+)"|\'([^\']+)\')', webpage)
+        if mobj is None:
+            self._downloader.report_warning(u'unable to extract description')
+            video_description = None
+        else:
+            video_description = mobj.group(1) or mobj.group(2)
+
+        mobj = re.search(r'<meta property="og:image" content=(?:"([^"]+)"|\'([^\']+)\')', webpage)
+        if mobj is None:
+            raise ExtractorError(u'Unable to extract thumbnail')
+        thumbnail = mobj.group(1) or mobj.group(2)
+
+        return [{
+            'id':       video_id,
+            'url':      video_url,
+            'ext':      'mp4',
+            'title':    video_title,
+            'description': video_description,
+            'thumbnail': thumbnail,
+            'uploader_id': video_uploader_id,
+        }]
+
+
 def gen_extractors():
     """ Return a list of an instance of every supported extractor.
     The order does matter; the first extractor matched is the one handling the URL.
@@ -4338,6 +4401,7 @@ def gen_extractors():
         InaIE(),
         HowcastIE(),
         VineIE(),
+        FlickrIE(),
         GenericIE()
     ]