import json
import os
import random
-import re
from .common import InfoExtractor
from ..compat import (
compat_urlparse,
)
from ..utils import (
+ dict_get,
ExtractorError,
float_or_none,
int_or_none,
parse_duration,
qualities,
srt_subtitles_timecode,
+ try_get,
+ update_url_query,
urlencode_postdata,
)
class PluralsightBaseIE(InfoExtractor):
- _API_BASE = 'http://app.pluralsight.com'
+ _API_BASE = 'https://app.pluralsight.com'
+
+ def _download_course(self, course_id, url, display_id):
+ try:
+ return self._download_course_rpc(course_id, url, display_id)
+ except ExtractorError:
+ # Old API fallback
+ return self._download_json(
+ 'https://app.pluralsight.com/player/user/api/v1/player/payload',
+ display_id, data=urlencode_postdata({'courseId': course_id}),
+ headers={'Referer': url})
+
+ def _download_course_rpc(self, course_id, url, display_id):
+ response = self._download_json(
+ '%s/player/functions/rpc' % self._API_BASE, display_id,
+ 'Downloading course JSON',
+ data=json.dumps({
+ 'fn': 'bootstrapPlayer',
+ 'payload': {
+ 'courseId': course_id,
+ },
+ }).encode('utf-8'),
+ headers={
+ 'Content-Type': 'application/json;charset=utf-8',
+ 'Referer': url,
+ })
+
+ course = try_get(response, lambda x: x['payload']['course'], dict)
+ if course:
+ return course
+
+ raise ExtractorError(
+ '%s said: %s' % (self.IE_NAME, response['error']['message']),
+ expected=True)
class PluralsightIE(PluralsightBaseIE):
'info_dict': {
'id': 'hosting-sql-server-windows-azure-iaas-m7-mgmt-04',
'ext': 'mp4',
- 'title': 'Management of SQL Server - Demo Monitoring',
+ 'title': 'Demo Monitoring',
'duration': 338,
},
'skip': 'Requires pluralsight account credentials',
post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
response = self._download_webpage(
- post_url, None, 'Logging in as %s' % username,
+ post_url, None, 'Logging in',
data=urlencode_postdata(login_form),
headers={'Content-Type': 'application/x-www-form-urlencoded'})
raise ExtractorError('Unable to login: %s' % error, expected=True)
if all(p not in response for p in ('__INITIAL_STATE__', '"currentUser"')):
+ BLOCKED = 'Your account has been blocked due to suspicious activity'
+ if BLOCKED in response:
+ raise ExtractorError(
+ 'Unable to login: %s' % BLOCKED, expected=True)
+ MUST_AGREE = 'To continue using Pluralsight, you must agree to'
+ if any(p in response for p in (MUST_AGREE, '>Disagree<', '>Agree<')):
+ raise ExtractorError(
+ 'Unable to login: %s some documents. Go to pluralsight.com, '
+ 'log in and agree with what Pluralsight requires.'
+ % MUST_AGREE, expected=True)
+
raise ExtractorError('Unable to log in')
def _get_subtitles(self, author, clip_id, lang, name, duration, video_id):
'm': name,
}
captions = self._download_json(
- '%s/training/Player/Captions' % self._API_BASE, video_id,
+ '%s/player/retrieve-captions' % self._API_BASE, video_id,
'Downloading captions JSON', 'Unable to download captions JSON',
fatal=False, data=json.dumps(captions_post).encode('utf-8'),
headers={'Content-Type': 'application/json;charset=utf-8'})
@staticmethod
def _convert_subtitles(duration, subs):
srt = ''
+ TIME_OFFSET_KEYS = ('displayTimeOffset', 'DisplayTimeOffset')
+ TEXT_KEYS = ('text', 'Text')
for num, current in enumerate(subs):
current = subs[num]
- start, text = float_or_none(
- current.get('DisplayTimeOffset')), current.get('Text')
+ start, text = (
+ float_or_none(dict_get(current, TIME_OFFSET_KEYS, skip_false_values=False)),
+ dict_get(current, TEXT_KEYS))
if start is None or text is None:
continue
end = duration if num == len(subs) - 1 else float_or_none(
- subs[num + 1].get('DisplayTimeOffset'))
+ dict_get(subs[num + 1], TIME_OFFSET_KEYS, skip_false_values=False))
if end is None:
continue
srt += os.linesep.join(
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]
+ course_name = qs.get('course', [None])[0]
- if any(not f for f in (author, name, clip_id, course,)):
+ if any(not f for f in (author, name, clip_id, course_name,)):
raise ExtractorError('Invalid URL', expected=True)
display_id = '%s-%s' % (name, clip_id)
- webpage = self._download_webpage(url, display_id)
-
- modules = self._search_regex(
- r'moduleCollection\s*:\s*new\s+ModuleCollection\((\[.+?\])\s*,\s*\$rootScope\)',
- webpage, 'modules', default=None)
+ course = self._download_course(course_name, url, display_id)
- 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']
+ collection = course['modules']
- module, clip = None, None
+ clip = None
for module_ in collection:
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:
if not clip:
raise ExtractorError('Unable to resolve clip')
- title = '%s - %s' % (module['title'], clip['title'])
+ title = clip['title']
QUALITIES = {
'low': {'width': 640, 'height': 480},
# 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
+ widescreen = course.get('supportsWideScreenVideoFormats') is True
best_quality = 'high-widescreen' if widescreen else 'high'
if widescreen:
for allowed_quality in ALLOWED_QUALITIES:
req_format_split = req_format.split('-', 1)
if len(req_format_split) > 1:
req_ext, req_quality = req_format_split
+ req_quality = '-'.join(req_quality.split('-')[:2])
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, )), )
for quality in qualities_:
f = QUALITIES[quality].copy()
clip_post = {
- 'a': author,
- 'cap': 'false',
- 'cn': clip_id,
- 'course': course,
- 'lc': 'en',
- 'm': name,
- 'mt': ext,
- 'q': '%dx%d' % (f['width'], f['height']),
+ 'author': author,
+ 'includeCaptions': False,
+ 'clipIndex': int(clip_id),
+ 'courseName': course_name,
+ 'locale': 'en',
+ 'moduleName': name,
+ 'mediaType': ext,
+ 'quality': '%dx%d' % (f['width'], f['height']),
}
format_id = '%s-%s' % (ext, quality)
- clip_url = self._download_webpage(
- '%s/training/Player/ViewClip' % self._API_BASE, display_id,
- 'Downloading %s URL' % format_id, fatal=False,
+ viewclip = self._download_json(
+ '%s/video/clips/viewclip' % self._API_BASE, display_id,
+ 'Downloading %s viewclip JSON' % format_id, fatal=False,
data=json.dumps(clip_post).encode('utf-8'),
headers={'Content-Type': 'application/json;charset=utf-8'})
random.randint(2, 5), display_id,
'%(video_id)s: Waiting for %(timeout)s seconds to avoid throttling')
- if not clip_url:
+ if not viewclip:
continue
- f.update({
- 'url': clip_url,
- 'ext': ext,
- 'format_id': format_id,
- 'quality': quality_key(quality),
- })
- formats.append(f)
+
+ clip_urls = viewclip.get('urls')
+ if not isinstance(clip_urls, list):
+ continue
+
+ for clip_url_data in clip_urls:
+ clip_url = clip_url_data.get('url')
+ if not clip_url:
+ continue
+ cdn = clip_url_data.get('cdn')
+ clip_f = f.copy()
+ clip_f.update({
+ 'url': clip_url,
+ 'ext': ext,
+ 'format_id': '%s-%s' % (format_id, cdn) if cdn else format_id,
+ 'quality': quality_key(quality),
+ 'source_preference': int_or_none(clip_url_data.get('rank')),
+ })
+ formats.append(clip_f)
+
self._sort_formats(formats)
duration = int_or_none(
# TODO: PSM cookie
- course = self._download_json(
- '%s/data/course/%s' % (self._API_BASE, course_id),
- course_id, 'Downloading course JSON')
+ course = self._download_course(course_id, url, course_id)
title = course['title']
+ course_name = course['name']
+ course_data = course['modules']
description = course.get('description') or course.get('shortDescription')
- course_data = self._download_json(
- '%s/data/course/content/%s' % (self._API_BASE, course_id),
- course_id, 'Downloading course data JSON')
-
entries = []
for num, module in enumerate(course_data, 1):
+ author = module.get('author')
+ module_name = module.get('name')
+ if not author or not module_name:
+ continue
for clip in module.get('clips', []):
- player_parameters = clip.get('playerParameters')
- if not player_parameters:
+ clip_index = int_or_none(clip.get('index'))
+ if clip_index is None:
continue
+ clip_url = update_url_query(
+ '%s/player' % self._API_BASE, query={
+ 'mode': 'live',
+ 'course': course_name,
+ 'author': author,
+ 'name': module_name,
+ 'clip': clip_index,
+ })
entries.append({
'_type': 'url_transparent',
- 'url': '%s/training/player?%s' % (self._API_BASE, player_parameters),
+ 'url': clip_url,
'ie_key': PluralsightIE.ie_key(),
'chapter': module.get('title'),
'chapter_number': num,