Merge pull request #7045 from remitamine/ign

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

diff --git a/youtube_dl/extractor/pluralsight.py b/youtube_dl/extractor/pluralsight.py
index a11cebaf1a8a234046daa53397509e349ef2e03f..55c11b3bf89be7196c3c95d0a3fb25dae27ce51b 100644 (file)
--- a/youtube_dl/extractor/pluralsight.py
+++ b/youtube_dl/extractor/pluralsight.py
@@ -1,5 +1,6 @@
 from __future__ import unicode_literals
 
+import re
 import json
 import random
 import collections
@@ -8,13 +9,14 @@ 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,
+    qualities,
+    sanitized_Request,
 )
 
 
@@ -73,7 +75,7 @@ class PluralsightIE(PluralsightBaseIE):
         if not post_url.startswith('http'):
             post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
 
-        request = compat_urllib_request.Request(
+        request = sanitized_Request(
             post_url, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 
@@ -104,19 +106,29 @@ class PluralsightIE(PluralsightBaseIE):
 
         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:
-            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')
+                    if clip_index is None:
+                        clip_index = clip_.get('index')
                     if clip_index is None:
                         continue
                     if compat_str(clip_index) == clip_id:
@@ -130,15 +142,28 @@ class PluralsightIE(PluralsightBaseIE):
             '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 = (
-            AllowedQuality('webm', ('high',)),
-            AllowedQuality('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.
@@ -147,19 +172,19 @@ class PluralsightIE(PluralsightBaseIE):
         else:
             def guess_allowed_qualities():
                 req_format = self._downloader.params.get('format') or 'best'
-                req_format_split = req_format.split('-')
+                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, ('high', )), )
+                return (AllowedQuality(req_ext, (best_quality, )), )
             allowed_qualities = guess_allowed_qualities()
 
         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,
@@ -171,7 +196,7 @@ class PluralsightIE(PluralsightBaseIE):
                     'mt': ext,
                     'q': '%dx%d' % (f['width'], f['height']),
                 }
-                request = compat_urllib_request.Request(
+                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')
@@ -195,6 +220,7 @@ class PluralsightIE(PluralsightBaseIE):
                     'url': clip_url,
                     'ext': ext,
                     'format_id': format_id,
+                    'quality': quality_key(quality),
                 })
                 formats.append(f)
         self._sort_formats(formats)