Merge remote-tracking branch 'rzhxeo/crunchyroll'

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

diff --git a/youtube_dl/extractor/internetvideoarchive.py b/youtube_dl/extractor/internetvideoarchive.py
index 52e3f9eecb9dbc7192e3298f0a5f766d84d62394..16a6f73c87e27d47d401c444aa02b80d5a6313b3 100644 (file)
--- a/youtube_dl/extractor/internetvideoarchive.py
+++ b/youtube_dl/extractor/internetvideoarchive.py
@@ -1,9 +1,9 @@
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
     compat_urlparse,
+    compat_urllib_parse,
     xpath_with_ns,
     determine_ext,
 )
@@ -18,7 +18,7 @@ class InternetVideoArchiveIE(InfoExtractor):
         u'info_dict': {
             u'title': u'SKYFALL',
             u'description': u'In SKYFALL, Bond\'s loyalty to M is tested as her past comes back to haunt her. As MI6 comes under attack, 007 must track down and destroy the threat, no matter how personal the cost.',
-            u'duration': 156,
+            u'duration': 153,
         },
     }
 
@@ -26,20 +26,33 @@ class InternetVideoArchiveIE(InfoExtractor):
     def _build_url(query):
         return 'http://video.internetvideoarchive.net/flash/players/flashconfiguration.aspx?' + query
 
+    @staticmethod
+    def _clean_query(query):
+        NEEDED_ARGS = ['publishedid', 'customerid']
+        query_dic = compat_urlparse.parse_qs(query)
+        cleaned_dic = dict((k,v[0]) for (k,v) in query_dic.items() if k in NEEDED_ARGS)
+        # Other player ids return m3u8 urls
+        cleaned_dic['playerid'] = '247'
+        cleaned_dic['videokbrate'] = '100000'
+        return compat_urllib_parse.urlencode(cleaned_dic)
+
     def _real_extract(self, url):
         query = compat_urlparse.urlparse(url).query
         query_dic = compat_urlparse.parse_qs(query)
         video_id = query_dic['publishedid'][0]
         url = self._build_url(query)
 
-        flashconfiguration_xml = self._download_webpage(url, video_id,
+        flashconfiguration = self._download_xml(url, video_id,
             u'Downloading flash configuration')
-        flashconfiguration = xml.etree.ElementTree.fromstring(flashconfiguration_xml.encode('utf-8'))
         file_url = flashconfiguration.find('file').text
         file_url = file_url.replace('/playlist.aspx', '/mrssplaylist.aspx')
-        info_xml = self._download_webpage(file_url, video_id,
+        # Replace some of the parameters in the query to get the best quality
+        # and http links (no m3u8 manifests)
+        file_url = re.sub(r'(?<=\?)(.+)$',
+            lambda m: self._clean_query(m.group()),
+            file_url)
+        info = self._download_xml(file_url, video_id,
             u'Downloading video info')
-        info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
         item = info.find('channel/item')
 
         def _bp(p):
@@ -58,7 +71,7 @@ class InternetVideoArchiveIE(InfoExtractor):
             })
         formats = sorted(formats, key=lambda f: f['bitrate'])
 
-        info = {
+        return {
             'id': video_id,
             'title': item.find('title').text,
             'formats': formats,
@@ -66,6 +79,3 @@ class InternetVideoArchiveIE(InfoExtractor):
             'description': item.find('description').text,
             'duration': int(attr['duration']),
         }
-        # TODO: Remove when #980 has been merged
-        info.update(formats[-1])
-        return info