Work around buggy HTML Parser in Python < 2.7.3 (Closes #662)

[youtube-dl.git] / youtube_dl / InfoExtractors.py

diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index a708cc75068c5f627775920fee6359ddea2fd18c..0860937eefaf023877dbfb2978e52a641b0ad4ed 100755 (executable)
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -5,6 +5,7 @@ from __future__ import absolute_import
 
 import base64
 import datetime
+import itertools
 import netrc
 import os
 import re
@@ -2868,8 +2869,7 @@ class StanfordOpenClassroomIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         if mobj is None:
-            self._downloader.trouble(u'ERROR: invalid URL: %s' % url)
-            return
+            raise ExtractorError(u'Invalid URL: %s' % url)
 
         if mobj.group('course') and mobj.group('video'): # A specific video
             course = mobj.group('course')
@@ -2906,12 +2906,9 @@ class StanfordOpenClassroomIE(InfoExtractor):
                 'upload_date': None,
             }
 
-            self.report_download_webpage(info['id'])
-            try:
-                coursepage = compat_urllib_request.urlopen(url).read()
-            except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-                self._downloader.trouble(u'ERROR: unable to download course info page: ' + compat_str(err))
-                return
+            coursepage = self._download_webpage(url, info['id'],
+                                        note='Downloading course info page',
+                                        errnote='Unable to download course info page')
 
             m = re.search('<h1>([^<]+)</h1>', coursepage)
             if m:
@@ -2935,7 +2932,6 @@ class StanfordOpenClassroomIE(InfoExtractor):
                 assert entry['type'] == 'reference'
                 results += self.extract(entry['url'])
             return results
-
         else: # Root page
             info = {
                 'id': 'Stanford OpenClassroom',
@@ -3812,8 +3808,6 @@ class PornotubeIE(InfoExtractor):
 
         return [info]
 
-
-
 class YouJizzIE(InfoExtractor):
     """Information extractor for youjizz.com."""
     _VALID_URL = r'^(?:https?://)?(?:\w+\.)?youjizz\.com/videos/(?P<videoid>[^.]+).html$'
@@ -3860,6 +3854,49 @@ class YouJizzIE(InfoExtractor):
 
         return [info]
 
+class EightTracksIE(InfoExtractor):
+    IE_NAME = '8tracks'
+    _VALID_URL = r'https?://8tracks.com/(?P<user>[^/]+)/(?P<id>[^/#]+)(?:#.*)?$'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        if mobj is None:
+            raise ExtractorError(u'Invalid URL: %s' % url)
+        playlist_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, playlist_id)
+
+        m = re.search(r"new TRAX.Mix\((.*?)\);\n*\s*TRAX.initSearchAutocomplete\('#search'\);", webpage, flags=re.DOTALL)
+        if not m:
+            raise ExtractorError(u'Cannot find trax information')
+        json_like = m.group(1)
+        data = json.loads(json_like)
+
+        session = str(random.randint(0, 1000000000))
+        mix_id = data['id']
+        track_count = data['tracks_count']
+        first_url = 'http://8tracks.com/sets/%s/play?player=sm&mix_id=%s&format=jsonh' % (session, mix_id)
+        next_url = first_url
+        res = []
+        for i in itertools.count():
+            api_json = self._download_webpage(next_url, playlist_id,
+                note=u'Downloading song information %s/%s' % (str(i+1), track_count),
+                errnote=u'Failed to download song information')
+            api_data = json.loads(api_json)
+            track_data = api_data[u'set']['track']
+            info = {
+                'id': track_data['id'],
+                'url': track_data['track_file_stream_url'],
+                'title': track_data['performer'] + u' - ' + track_data['name'],
+                'raw_title': track_data['name'],
+                'uploader_id': data['user']['login'],
+                'ext': 'm4a',
+            }
+            res.append(info)
+            if api_data['set']['at_last_track']:
+                break
+            next_url = 'http://8tracks.com/sets/%s/next?player=sm&mix_id=%s&format=jsonh&track_id=%s' % (session, mix_id, track_data['id'])
+        return res
 
 def gen_extractors():
     """ Return a list of an instance of every supported extractor.
@@ -3906,6 +3943,7 @@ def gen_extractors():
         SteamIE(),
         UstreamIE(),
         RBMARadioIE(),
+        EightTracksIE(),
         GenericIE()
     ]