Merge tag 'upstream/2016.08.17'

[youtubedl] / youtube_dl / extractor / pluralsight.py

diff --git a/youtube_dl/extractor/pluralsight.py b/youtube_dl/extractor/pluralsight.py
index fd32836ccaf61b171f9dcafbbfce24c7764b18c1..9aab7764559523e3bbc4a82f5b6ab9b71056be59 100644 (file)
--- a/youtube_dl/extractor/pluralsight.py
+++ b/youtube_dl/extractor/pluralsight.py
@@ -2,28 +2,36 @@ from __future__ import unicode_literals
 
 import re
 import json
 
 import re
 import json

+import random
+import collections

 
 from .common import InfoExtractor
 from ..compat import (
     compat_str,
 
 from .common import InfoExtractor
 from ..compat import (
     compat_str,

-    compat_urllib_parse,
-    compat_urllib_request,

     compat_urlparse,
 )
 from ..utils import (
     ExtractorError,
     int_or_none,
     parse_duration,
     compat_urlparse,
 )
 from ..utils import (
     ExtractorError,
     int_or_none,
     parse_duration,

+    qualities,
+    sanitized_Request,
+    urlencode_postdata,

 )
 
 
 )
 
 

-class PluralsightIE(InfoExtractor):
+class PluralsightBaseIE(InfoExtractor):
+    _API_BASE = 'http://app.pluralsight.com'
+
+
+class PluralsightIE(PluralsightBaseIE):

     IE_NAME = 'pluralsight'
     IE_NAME = 'pluralsight'

-    _VALID_URL = r'https?://(?:www\.)?pluralsight\.com/training/player\?author=(?P<author>[^&]+)&name=(?P<name>[^&]+)(?:&mode=live)?&clip=(?P<clip>\d+)&course=(?P<course>[^&]+)'
-    _LOGIN_URL = 'https://www.pluralsight.com/id/'
+    _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/training/player\?'
+    _LOGIN_URL = 'https://app.pluralsight.com/id/'
+

     _NETRC_MACHINE = 'pluralsight'
 
     _NETRC_MACHINE = 'pluralsight'
 

-    _TEST = {
+    _TESTS = [{

         'url': 'http://www.pluralsight.com/training/player?author=mike-mckeown&name=hosting-sql-server-windows-azure-iaas-m7-mgmt&mode=live&clip=3&course=hosting-sql-server-windows-azure-iaas',
         'md5': '4d458cf5cf4c593788672419a8dd4cf8',
         'info_dict': {
         'url': 'http://www.pluralsight.com/training/player?author=mike-mckeown&name=hosting-sql-server-windows-azure-iaas-m7-mgmt&mode=live&clip=3&course=hosting-sql-server-windows-azure-iaas',
         'md5': '4d458cf5cf4c593788672419a8dd4cf8',
         'info_dict': {


@@ -33,7 +41,14 @@ class PluralsightIE(InfoExtractor):
             'duration': 338,
         },
         'skip': 'Requires pluralsight account credentials',
             'duration': 338,
         },
         'skip': 'Requires pluralsight account credentials',

-    }
+    }, {
+        'url': 'https://app.pluralsight.com/training/player?course=angularjs-get-started&author=scott-allen&name=angularjs-get-started-m1-introduction&clip=0&mode=live',
+        'only_matching': True,
+    }, {
+        # available without pluralsight account
+        'url': 'http://app.pluralsight.com/training/player?author=scott-allen&name=angularjs-get-started-m1-introduction&mode=live&clip=0&course=angularjs-get-started',
+        'only_matching': True,
+    }]

 
     def _real_initialize(self):
         self._login()
 
     def _real_initialize(self):
         self._login()


@@ -41,7 +56,7 @@ class PluralsightIE(InfoExtractor):
     def _login(self):
         (username, password) = self._get_login_info()
         if username is None:
     def _login(self):
         (username, password) = self._get_login_info()
         if username is None:

-            self.raise_login_required('Pluralsight account is required')
+            return

 
         login_page = self._download_webpage(
             self._LOGIN_URL, None, 'Downloading login page')
 
         login_page = self._download_webpage(
             self._LOGIN_URL, None, 'Downloading login page')


@@ -49,8 +64,8 @@ class PluralsightIE(InfoExtractor):
         login_form = self._hidden_inputs(login_page)
 
         login_form.update({
         login_form = self._hidden_inputs(login_page)
 
         login_form.update({

-            'Username': username.encode('utf-8'),
-            'Password': password.encode('utf-8'),
+            'Username': username,
+            'Password': password,

         })
 
         post_url = self._search_regex(
         })
 
         post_url = self._search_regex(


@@ -60,8 +75,8 @@ class PluralsightIE(InfoExtractor):
         if not post_url.startswith('http'):
             post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
 
         if not post_url.startswith('http'):
             post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
 

-        request = compat_urllib_request.Request(
-            post_url, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
+        request = sanitized_Request(
+            post_url, urlencode_postdata(login_form))

         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 
         response = self._download_webpage(
         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 
         response = self._download_webpage(


@@ -73,30 +88,47 @@ class PluralsightIE(InfoExtractor):
         if error:
             raise ExtractorError('Unable to login: %s' % error, expected=True)
 
         if error:
             raise ExtractorError('Unable to login: %s' % error, expected=True)
 

+        if all(p not in response for p in ('__INITIAL_STATE__', '"currentUser"')):
+            raise ExtractorError('Unable to log in')
+

     def _real_extract(self, url):
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        author = mobj.group('author')
-        name = mobj.group('name')
-        clip_id = mobj.group('clip')
-        course = mobj.group('course')
+        qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
+
+        author = qs.get('author', [None])[0]
+        name = qs.get('name', [None])[0]
+        clip_id = qs.get('clip', [None])[0]
+        course = qs.get('course', [None])[0]
+
+        if any(not f for f in (author, name, clip_id, course,)):
+            raise ExtractorError('Invalid URL', expected=True)

 
         display_id = '%s-%s' % (name, clip_id)
 
         webpage = self._download_webpage(url, display_id)
 
 
         display_id = '%s-%s' % (name, clip_id)
 
         webpage = self._download_webpage(url, display_id)
 

-        collection = self._parse_json(
-            self._search_regex(
-                r'moduleCollection\s*:\s*new\s+ModuleCollection\((\[.+?\])\s*,\s*\$rootScope\)',
-                webpage, 'modules'),
-            display_id)
+        modules = self._search_regex(
+            r'moduleCollection\s*:\s*new\s+ModuleCollection\((\[.+?\])\s*,\s*\$rootScope\)',
+            webpage, 'modules', default=None)
+
+        if modules:
+            collection = self._parse_json(modules, display_id)
+        else:
+            # Webpage may be served in different layout (see
+            # https://github.com/rg3/youtube-dl/issues/7607)
+            collection = self._parse_json(
+                self._search_regex(
+                    r'var\s+initialState\s*=\s*({.+?});\n', webpage, 'initial state'),
+                display_id)['course']['modules']

 
         module, clip = None, None
 
         for module_ in collection:
 
         module, clip = None, None
 
         for module_ in collection:

-            if module_.get('moduleName') == name:
+            if name in (module_.get('moduleName'), module_.get('name')):

                 module = module_
                 for clip_ in module_.get('clips', []):
                     clip_index = clip_.get('clipIndex')
                 module = module_
                 for clip_ in module_.get('clips', []):
                     clip_index = clip_.get('clipIndex')

+                    if clip_index is None:
+                        clip_index = clip_.get('index')

                     if clip_index is None:
                         continue
                     if compat_str(clip_index) == clip_id:
                     if clip_index is None:
                         continue
                     if compat_str(clip_index) == clip_id:


@@ -110,16 +142,49 @@ class PluralsightIE(InfoExtractor):
             'low': {'width': 640, 'height': 480},
             'medium': {'width': 848, 'height': 640},
             'high': {'width': 1024, 'height': 768},
             'low': {'width': 640, 'height': 480},
             'medium': {'width': 848, 'height': 640},
             'high': {'width': 1024, 'height': 768},

+            'high-widescreen': {'width': 1280, 'height': 720},

         }
 
         }
 

+        QUALITIES_PREFERENCE = ('low', 'medium', 'high', 'high-widescreen',)
+        quality_key = qualities(QUALITIES_PREFERENCE)
+
+        AllowedQuality = collections.namedtuple('AllowedQuality', ['ext', 'qualities'])
+

         ALLOWED_QUALITIES = (
         ALLOWED_QUALITIES = (

-            ('webm', ('high',)),
-            ('mp4', ('low', 'medium', 'high',)),
+            AllowedQuality('webm', ['high', ]),
+            AllowedQuality('mp4', ['low', 'medium', 'high', ]),

         )
 
         )
 

+        # Some courses also offer widescreen resolution for high quality (see
+        # https://github.com/rg3/youtube-dl/issues/7766)
+        widescreen = True if re.search(
+            r'courseSupportsWidescreenVideoFormats\s*:\s*true', webpage) else False
+        best_quality = 'high-widescreen' if widescreen else 'high'
+        if widescreen:
+            for allowed_quality in ALLOWED_QUALITIES:
+                allowed_quality.qualities.append(best_quality)
+
+        # In order to minimize the number of calls to ViewClip API and reduce
+        # the probability of being throttled or banned by Pluralsight we will request
+        # only single format until formats listing was explicitly requested.
+        if self._downloader.params.get('listformats', False):
+            allowed_qualities = ALLOWED_QUALITIES
+        else:
+            def guess_allowed_qualities():
+                req_format = self._downloader.params.get('format') or 'best'
+                req_format_split = req_format.split('-', 1)
+                if len(req_format_split) > 1:
+                    req_ext, req_quality = req_format_split
+                    for allowed_quality in ALLOWED_QUALITIES:
+                        if req_ext == allowed_quality.ext and req_quality in allowed_quality.qualities:
+                            return (AllowedQuality(req_ext, (req_quality, )), )
+                req_ext = 'webm' if self._downloader.params.get('prefer_free_formats') else 'mp4'
+                return (AllowedQuality(req_ext, (best_quality, )), )
+            allowed_qualities = guess_allowed_qualities()
+

         formats = []
         formats = []

-        for ext, qualities in ALLOWED_QUALITIES:
-            for quality in qualities:
+        for ext, qualities_ in allowed_qualities:
+            for quality in qualities_:

                 f = QUALITIES[quality].copy()
                 clip_post = {
                     'a': author,
                 f = QUALITIES[quality].copy()
                 clip_post = {
                     'a': author,


@@ -131,19 +196,31 @@ class PluralsightIE(InfoExtractor):
                     'mt': ext,
                     'q': '%dx%d' % (f['width'], f['height']),
                 }
                     'mt': ext,
                     'q': '%dx%d' % (f['width'], f['height']),
                 }

-                request = compat_urllib_request.Request(
-                    'http://www.pluralsight.com/training/Player/ViewClip',
+                request = sanitized_Request(
+                    '%s/training/Player/ViewClip' % self._API_BASE,

                     json.dumps(clip_post).encode('utf-8'))
                 request.add_header('Content-Type', 'application/json;charset=utf-8')
                 format_id = '%s-%s' % (ext, quality)
                 clip_url = self._download_webpage(
                     request, display_id, 'Downloading %s URL' % format_id, fatal=False)
                     json.dumps(clip_post).encode('utf-8'))
                 request.add_header('Content-Type', 'application/json;charset=utf-8')
                 format_id = '%s-%s' % (ext, quality)
                 clip_url = self._download_webpage(
                     request, display_id, 'Downloading %s URL' % format_id, fatal=False)

+
+                # Pluralsight tracks multiple sequential calls to ViewClip API and start
+                # to return 429 HTTP errors after some time (see
+                # https://github.com/rg3/youtube-dl/pull/6989). Moreover it may even lead
+                # to account ban (see https://github.com/rg3/youtube-dl/issues/6842).
+                # To somewhat reduce the probability of these consequences
+                # we will sleep random amount of time before each call to ViewClip.
+                self._sleep(
+                    random.randint(2, 5), display_id,
+                    '%(video_id)s: Waiting for %(timeout)s seconds to avoid throttling')
+

                 if not clip_url:
                     continue
                 f.update({
                     'url': clip_url,
                     'ext': ext,
                     'format_id': format_id,
                 if not clip_url:
                     continue
                 f.update({
                     'url': clip_url,
                     'ext': ext,
                     'format_id': format_id,

+                    'quality': quality_key(quality),

                 })
                 formats.append(f)
         self._sort_formats(formats)
                 })
                 formats.append(f)
         self._sort_formats(formats)


@@ -155,7 +232,7 @@ class PluralsightIE(InfoExtractor):
         # { a = author, cn = clip_id, lc = end, m = name }
 
         return {
         # { a = author, cn = clip_id, lc = end, m = name }
 
         return {

-            'id': clip['clipName'],
+            'id': clip.get('clipName') or clip['name'],

             'title': '%s - %s' % (module['title'], clip['title']),
             'duration': int_or_none(clip.get('duration')) or parse_duration(clip.get('formattedDuration')),
             'creator': author,
             'title': '%s - %s' % (module['title'], clip['title']),
             'duration': int_or_none(clip.get('duration')) or parse_duration(clip.get('formattedDuration')),
             'creator': author,


@@ -163,10 +240,10 @@ class PluralsightIE(InfoExtractor):
         }
 
 
         }
 
 

-class PluralsightCourseIE(InfoExtractor):
+class PluralsightCourseIE(PluralsightBaseIE):

     IE_NAME = 'pluralsight:course'
     IE_NAME = 'pluralsight:course'

-    _VALID_URL = r'https?://(?:www\.)?pluralsight\.com/courses/(?P<id>[^/]+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/(?:library/)?courses/(?P<id>[^/]+)'
+    _TESTS = [{

         # Free course from Pluralsight Starter Subscription for Microsoft TechNet
         # https://offers.pluralsight.com/technet?loc=zTS3z&prod=zOTprodz&tech=zOttechz&prog=zOTprogz&type=zSOz&media=zOTmediaz&country=zUSz
         'url': 'http://www.pluralsight.com/courses/hosting-sql-server-windows-azure-iaas',
         # Free course from Pluralsight Starter Subscription for Microsoft TechNet
         # https://offers.pluralsight.com/technet?loc=zTS3z&prod=zOTprodz&tech=zOttechz&prog=zOTprogz&type=zSOz&media=zOTmediaz&country=zUSz
         'url': 'http://www.pluralsight.com/courses/hosting-sql-server-windows-azure-iaas',


@@ -176,7 +253,14 @@ class PluralsightCourseIE(InfoExtractor):
             'description': 'md5:61b37e60f21c4b2f91dc621a977d0986',
         },
         'playlist_count': 31,
             'description': 'md5:61b37e60f21c4b2f91dc621a977d0986',
         },
         'playlist_count': 31,

-    }
+    }, {
+        # available without pluralsight account
+        'url': 'https://www.pluralsight.com/courses/angularjs-get-started',
+        'only_matching': True,
+    }, {
+        'url': 'https://app.pluralsight.com/library/courses/understanding-microsoft-azure-amazon-aws/table-of-contents',
+        'only_matching': True,
+    }]

 
     def _real_extract(self, url):
         course_id = self._match_id(url)
 
     def _real_extract(self, url):
         course_id = self._match_id(url)


@@ -184,24 +268,29 @@ class PluralsightCourseIE(InfoExtractor):
         # TODO: PSM cookie
 
         course = self._download_json(
         # TODO: PSM cookie
 
         course = self._download_json(

-            'http://www.pluralsight.com/data/course/%s' % course_id,
+            '%s/data/course/%s' % (self._API_BASE, course_id),

             course_id, 'Downloading course JSON')
 
         title = course['title']
         description = course.get('description') or course.get('shortDescription')
 
         course_data = self._download_json(
             course_id, 'Downloading course JSON')
 
         title = course['title']
         description = course.get('description') or course.get('shortDescription')
 
         course_data = self._download_json(

-            'http://www.pluralsight.com/data/course/content/%s' % course_id,
+            '%s/data/course/content/%s' % (self._API_BASE, course_id),

             course_id, 'Downloading course data JSON')
 
         entries = []
             course_id, 'Downloading course data JSON')
 
         entries = []

-        for module in course_data:
+        for num, module in enumerate(course_data, 1):

             for clip in module.get('clips', []):
                 player_parameters = clip.get('playerParameters')
                 if not player_parameters:
                     continue
             for clip in module.get('clips', []):
                 player_parameters = clip.get('playerParameters')
                 if not player_parameters:
                     continue

-                entries.append(self.url_result(
-                    'http://www.pluralsight.com/training/player?%s' % player_parameters,
-                    'Pluralsight'))
+                entries.append({
+                    '_type': 'url_transparent',
+                    'url': '%s/training/player?%s' % (self._API_BASE, player_parameters),
+                    'ie_key': PluralsightIE.ie_key(),
+                    'chapter': module.get('title'),
+                    'chapter_number': num,
+                    'chapter_id': module.get('moduleRef'),
+                })

 
         return self.playlist_result(entries, course_id, title, description)
 
         return self.playlist_result(entries, course_id, title, description)