X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;ds=sidebyside;f=youtube_dl%2Fextractor%2Fudemy.py;h=9d31a3f5bbbf5528b2b541d46a720d3070b277b0;hb=3b35c3425e2c4d9836a3efe5cf73e35a60f674d1;hp=0e4d386a8ba32387f6f9025e633efb4c1ee59700;hpb=61c0663c1e40a6bfabb2a7ac933ce468f2989808;p=youtube-dl

diff --git a/youtube_dl/extractor/udemy.py b/youtube_dl/extractor/udemy.py
index 0e4d386a8..9d31a3f5b 100644
--- a/youtube_dl/extractor/udemy.py
+++ b/youtube_dl/extractor/udemy.py
@@ -3,17 +3,22 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import (
+from ..compat import (
     compat_urllib_parse,
     compat_urllib_request,
+)
+from ..utils import (
     ExtractorError,
+    int_or_none,
+    sanitized_Request,
 )
 
 
 class UdemyIE(InfoExtractor):
     IE_NAME = 'udemy'
     _VALID_URL = r'https?://www\.udemy\.com/(?:[^#]+#/lecture/|lecture/view/?\?lectureId=)(?P<id>\d+)'
-    _LOGIN_URL = 'https://www.udemy.com/join/login-submit/'
+    _LOGIN_URL = 'https://www.udemy.com/join/login-popup/?displayType=ajax&showSkipButton=1'
+    _ORIGIN_URL = 'https://www.udemy.com'
     _NETRC_MACHINE = 'udemy'
 
     _TESTS = [{
@@ -55,7 +60,7 @@ class UdemyIE(InfoExtractor):
             for header, value in headers.items():
                 url_or_request.add_header(header, value)
         else:
-            url_or_request = compat_urllib_request.Request(url_or_request, headers=headers)
+            url_or_request = sanitized_Request(url_or_request, headers=headers)
 
         response = super(UdemyIE, self)._download_json(url_or_request, video_id, note)
         self._handle_error(response)
@@ -67,41 +72,43 @@ class UdemyIE(InfoExtractor):
     def _login(self):
         (username, password) = self._get_login_info()
         if username is None:
-            raise ExtractorError(
-                'Udemy account is required, use --username and --password options to provide account credentials.',
-                expected=True)
+            self.raise_login_required('Udemy account is required')
 
         login_popup = self._download_webpage(
-            'https://www.udemy.com/join/login-popup?displayType=ajax&showSkipButton=1', None,
-            'Downloading login popup')
+            self._LOGIN_URL, None, 'Downloading login popup')
 
-        if login_popup == '<div class="run-command close-popup redirect" data-url="https://www.udemy.com/"></div>':
+        def is_logged(webpage):
+            return any(p in webpage for p in ['href="https://www.udemy.com/user/logout/', '>Logout<'])
+
+        # already logged in
+        if is_logged(login_popup):
             return
 
-        csrf = self._html_search_regex(
-            r'<input type="hidden" name="csrf" value="(.+?)"',
-            login_popup, 'csrf token')
+        login_form = self._form_hidden_inputs('login-form', login_popup)
 
-        login_form = {
-            'email': username,
-            'password': password,
-            'csrf': csrf,
-            'displayType': 'json',
-            'isSubmitted': '1',
-        }
-        request = compat_urllib_request.Request(
+        login_form.update({
+            'email': username.encode('utf-8'),
+            'password': password.encode('utf-8'),
+        })
+
+        request = sanitized_Request(
             self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
-        response = self._download_json(
+        request.add_header('Referer', self._ORIGIN_URL)
+        request.add_header('Origin', self._ORIGIN_URL)
+
+        response = self._download_webpage(
             request, None, 'Logging in as %s' % username)
 
-        if 'returnUrl' not in response:
+        if not is_logged(response):
+            error = self._html_search_regex(
+                r'(?s)<div[^>]+class="form-errors[^"]*">(.+?)</div>',
+                response, 'error message', default=None)
+            if error:
+                raise ExtractorError('Unable to login: %s' % error, expected=True)
             raise ExtractorError('Unable to log in')
 
-
-
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        lecture_id = mobj.group('id')
+        lecture_id = self._match_id(url)
 
         lecture = self._download_json(
             'https://www.udemy.com/api-1.1/lectures/%s' % lecture_id,
@@ -121,26 +128,47 @@ class UdemyIE(InfoExtractor):
 
         video_id = asset['id']
         thumbnail = asset.get('thumbnailUrl') or asset.get('thumbnail_url')
-        duration = asset['data']['duration']
+        duration = int_or_none(asset.get('data', {}).get('duration'))
 
         download_url = asset.get('downloadUrl') or asset.get('download_url')
 
         video = download_url.get('Video') or download_url.get('video')
         video_480p = download_url.get('Video480p') or download_url.get('video_480p')
 
-        formats = [
-            {
-                'url': video_480p[0],
-                'format_id': '360p',
-            },
-            {
-                'url': video[0],
-                'format_id': '720p',
-            },
-        ]
+        formats = [{
+            'url': video_480p[0],
+            'format_id': 'download-360p',
+        }, {
+            'url': video[0],
+            'format_id': 'download-720p',
+        }]
+
+        # Some videos also contain formats in asset['data']['outputs'] (e.g.
+        # https://www.udemy.com/ios9-swift/learn/#/lecture/3383208)
+        outputs = asset.get('data', {}).get('outputs')
+        if isinstance(outputs, dict):
+            for format_id, f in outputs.items():
+                video_url = f.get('url')
+                if video_url:
+                    formats.append({
+                        'url': video_url,
+                        'format_id': '%sp' % (f.get('labe1l') or format_id),
+                        'width': int_or_none(f.get('width')),
+                        'height': int_or_none(f.get('height')),
+                        'vbr': int_or_none(f.get('video_bitrate_in_kbps')),
+                        'vcodec': f.get('video_codec'),
+                        'fps': int_or_none(f.get('frame_rate')),
+                        'abr': int_or_none(f.get('audio_bitrate_in_kbps')),
+                        'acodec': f.get('audio_codec'),
+                        'asr': int_or_none(f.get('audio_sample_rate')),
+                        'tbr': int_or_none(f.get('total_bitrate_in_kbps')),
+                        'filesize': int_or_none(f.get('file_size_in_bytes')),
+                    })
+
+        self._sort_formats(formats)
 
         title = lecture['title']
-        description = lecture['description']
+        description = lecture.get('description')
 
         return {
             'id': video_id,