Fix extraction from youtube.

[youtubedl] / youtube_dl / extractor / teachertube.py

diff --git a/youtube_dl/extractor/teachertube.py b/youtube_dl/extractor/teachertube.py
index 4d9666c6b14980d6b04ebe1bd07cc3e79a9dd2bd..1272078c50b8703aa906e055fc57842078542e95 100644 (file)
--- a/youtube_dl/extractor/teachertube.py
+++ b/youtube_dl/extractor/teachertube.py
@@ -1,12 +1,13 @@
-# -*- coding: utf-8 -*-
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
 
 from .common import InfoExtractor
 from ..utils import (
-    qualities,
     determine_ext,
+    ExtractorError,
+    qualities,
 )
 
 
@@ -14,48 +15,59 @@ class TeacherTubeIE(InfoExtractor):
     IE_NAME = 'teachertube'
     IE_DESC = 'teachertube.com videos'
 
-    _VALID_URL = r'https?://(?:www\.)?teachertube\.com/(viewVideo\.php\?video_id=|music\.php\?music_id=)(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?teachertube\.com/(viewVideo\.php\?video_id=|music\.php\?music_id=|video/(?:[\da-z-]+-)?|audio/)(?P<id>\d+)'
 
     _TESTS = [{
+        # flowplayer
         'url': 'http://www.teachertube.com/viewVideo.php?video_id=339997',
         'md5': 'f9434ef992fd65936d72999951ee254c',
         'info_dict': {
             'id': '339997',
             'ext': 'mp4',
-            'title': 'Measures of dispersion from a frequency table_x264',
-            'description': 'md5:a3e9853487185e9fcd7181a07164650b',
-            'thumbnail': 're:http://.*\.jpg',
-        },
-    }, {
-        'url': 'http://www.teachertube.com/viewVideo.php?video_id=340064',
-        'md5': '0d625ec6bc9bf50f70170942ad580676',
-        'info_dict': {
-            'id': '340064',
-            'ext': 'mp4',
-            'title': 'How to Make Paper Dolls _ Paper Art Projects',
-            'description': 'md5:2ca52b20cd727773d1dc418b3d6bd07b',
-            'thumbnail': 're:http://.*\.jpg',
+            'title': 'Measures of dispersion from a frequency table',
+            'description': 'Measures of dispersion from a frequency table',
+            'thumbnail': r're:https?://.*\.(?:jpg|png)',
         },
     }, {
+        # jwplayer
         'url': 'http://www.teachertube.com/music.php?music_id=8805',
         'md5': '01e8352006c65757caf7b961f6050e21',
         'info_dict': {
             'id': '8805',
             'ext': 'mp3',
             'title': 'PER ASPERA AD ASTRA',
-            'description': 'RADIJSKA EMISIJA ZRAKOPLOVNE TEHNIČKE ŠKOLE PER ASPERA AD ASTRA',
+            'description': 'RADIJSKA EMISIJA ZRAKOPLOVNE TEHNI?KE ?KOLE P',
         },
+    }, {
+        # unavailable video
+        'url': 'http://www.teachertube.com/video/intro-video-schleicher-297790',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
+        error = self._search_regex(
+            r'<div\b[^>]+\bclass=["\']msgBox error[^>]+>([^<]+)', webpage,
+            'error', default=None)
+        if error:
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
+
+        title = self._html_search_meta('title', webpage, 'title', fatal=True)
+        TITLE_SUFFIX = ' - TeacherTube'
+        if title.endswith(TITLE_SUFFIX):
+            title = title[:-len(TITLE_SUFFIX)].strip()
+
+        description = self._html_search_meta('description', webpage, 'description')
+        if description:
+            description = description.strip()
+
         quality = qualities(['mp3', 'flv', 'mp4'])
 
-        _, media_urls = zip(*re.findall(r'([\'"])file\1\s*:\s*"([^"]+)"', webpage))
+        media_urls = re.findall(r'data-contenturl="([^"]+)"', webpage)
+        media_urls.extend(re.findall(r'var\s+filePath\s*=\s*"([^"]+)"', webpage))
+        media_urls.extend(re.findall(r'\'file\'\s*:\s*["\']([^"\']+)["\'],', webpage))
 
         formats = [
             {
@@ -66,30 +78,52 @@ class TeacherTubeIE(InfoExtractor):
 
         self._sort_formats(formats)
 
+        thumbnail = self._og_search_thumbnail(
+            webpage, default=None) or self._html_search_meta(
+            'thumbnail', webpage)
+
         return {
             'id': video_id,
-            'title': self._og_search_title(webpage),
-            'thumbnail': self._og_search_thumbnail(webpage),
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
             'formats': formats,
-            'description': self._og_search_description(webpage),
         }
 
 
-class TeacherTubeClassroomIE(InfoExtractor):
-    IE_NAME = 'teachertube:classroom'
-    IE_DESC = 'teachertube.com online classrooms'
+class TeacherTubeUserIE(InfoExtractor):
+    IE_NAME = 'teachertube:user:collection'
+    IE_DESC = 'teachertube.com user and collection videos'
+
+    _VALID_URL = r'https?://(?:www\.)?teachertube\.com/(user/profile|collection)/(?P<user>[0-9a-zA-Z]+)/?'
 
-    _VALID_URL = r'https?://(?:www\.)?teachertube\.com/view_classroom\.php\?user=(?P<user>[0-9a-zA-Z]+)'
+    _MEDIA_RE = r'''(?sx)
+        class="?sidebar_thumb_time"?>[0-9:]+</div>
+        \s*
+        <a\s+href="(https?://(?:www\.)?teachertube\.com/(?:video|audio)/[^"]+)"
+    '''
+    _TEST = {
+        'url': 'http://www.teachertube.com/user/profile/rbhagwati2',
+        'info_dict': {
+            'id': 'rbhagwati2'
+        },
+        'playlist_mincount': 179,
+    }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         user_id = mobj.group('user')
 
-        rss = self._download_xml('http://www.teachertube.com/rssclassroom.php?mode=user&username=%s' % user_id,
-                                      user_id, 'Downloading classroom RSS')
+        urls = []
+        webpage = self._download_webpage(url, user_id)
+        urls.extend(re.findall(self._MEDIA_RE, webpage))
 
-        entries = []
-        for url in rss.findall('.//{http://search.yahoo.com/mrss/}player'):
-            entries.append(self.url_result(url.attrib['url'], 'TeacherTube'))
+        pages = re.findall(r'/ajax-user/user-videos/%s\?page=([0-9]+)' % user_id, webpage)[:-1]
+        for p in pages:
+            more = 'http://www.teachertube.com/ajax-user/user-videos/%s?page=%s' % (user_id, p)
+            webpage = self._download_webpage(more, user_id, 'Downloading page %s/%s' % (p, len(pages)))
+            video_urls = re.findall(self._MEDIA_RE, webpage)
+            urls.extend(video_urls)
 
+        entries = [self.url_result(vurl, 'TeacherTube') for vurl in urls]
         return self.playlist_result(entries, user_id)