from __future__ import unicode_literals
-import re
+import collections
import json
+import os
import random
-import collections
from .common import InfoExtractor
from ..compat import (
compat_urlparse,
)
from ..utils import (
+ dict_get,
ExtractorError,
+ float_or_none,
int_or_none,
parse_duration,
qualities,
- sanitized_Request,
+ srt_subtitles_timecode,
urlencode_postdata,
)
class PluralsightBaseIE(InfoExtractor):
- _API_BASE = 'http://app.pluralsight.com'
+ _API_BASE = 'https://app.pluralsight.com'
class PluralsightIE(PluralsightBaseIE):
IE_NAME = 'pluralsight'
- _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/training/player\?'
+ _VALID_URL = r'https?://(?:(?:www|app)\.)?pluralsight\.com/(?:training/)?player\?'
_LOGIN_URL = 'https://app.pluralsight.com/id/'
_NETRC_MACHINE = 'pluralsight'
# 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,
+ }, {
+ 'url': 'https://app.pluralsight.com/player?course=ccna-intro-networking&author=ross-bagurdes&name=ccna-intro-networking-m06&clip=0',
+ 'only_matching': True,
}]
def _real_initialize(self):
if not post_url.startswith('http'):
post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
- request = sanitized_Request(
- post_url, urlencode_postdata(login_form))
- request.add_header('Content-Type', 'application/x-www-form-urlencoded')
-
response = self._download_webpage(
- request, None, 'Logging in as %s' % username)
+ post_url, None, 'Logging in as %s' % username,
+ data=urlencode_postdata(login_form),
+ headers={'Content-Type': 'application/x-www-form-urlencoded'})
error = self._search_regex(
r'<span[^>]+class="field-validation-error"[^>]*>([^<]+)</span>',
if all(p not in response for p in ('__INITIAL_STATE__', '"currentUser"')):
raise ExtractorError('Unable to log in')
+ def _get_subtitles(self, author, clip_id, lang, name, duration, video_id):
+ captions_post = {
+ 'a': author,
+ 'cn': clip_id,
+ 'lc': lang,
+ 'm': name,
+ }
+ captions = self._download_json(
+ '%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'})
+ if captions:
+ return {
+ lang: [{
+ 'ext': 'json',
+ 'data': json.dumps(captions),
+ }, {
+ 'ext': 'srt',
+ 'data': self._convert_subtitles(duration, captions),
+ }]
+ }
+
+ @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(dict_get(current, TIME_OFFSET_KEYS)),
+ 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(
+ dict_get(subs[num + 1], TIME_OFFSET_KEYS))
+ if end is None:
+ continue
+ srt += os.linesep.join(
+ (
+ '%d' % num,
+ '%s --> %s' % (
+ srt_subtitles_timecode(start),
+ srt_subtitles_timecode(end)),
+ text,
+ os.linesep,
+ ))
+ return srt
+
def _real_extract(self, url):
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]
+ 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)
+ parsed_url = compat_urlparse.urlparse(url)
- modules = self._search_regex(
- r'moduleCollection\s*:\s*new\s+ModuleCollection\((\[.+?\])\s*,\s*\$rootScope\)',
- webpage, 'modules', default=None)
+ payload_url = compat_urlparse.urlunparse(parsed_url._replace(
+ netloc='app.pluralsight.com', path='player/api/v1/payload'))
- 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']
+ course = self._download_json(
+ payload_url, display_id, headers={'Referer': url})['payload']['course']
+
+ collection = course['modules']
module, clip = None, None
if not clip:
raise ExtractorError('Unable to resolve clip')
+ title = '%s - %s' % (module['title'], clip['title'])
+
QUALITIES = {
'low': {'width': 640, 'height': 480},
'medium': {'width': 848, 'height': 640},
# 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:
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']),
}
- 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)
+ 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'})
# Pluralsight tracks multiple sequential calls to ViewClip API and start
# to return 429 HTTP errors after some time (see
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)
- # TODO: captions
- # http://www.pluralsight.com/training/Player/ViewClip + cap = true
- # or
- # http://www.pluralsight.com/training/Player/Captions
- # { a = author, cn = clip_id, lc = end, m = name }
+ duration = int_or_none(
+ clip.get('duration')) or parse_duration(clip.get('formattedDuration'))
+
+ # TODO: other languages?
+ subtitles = self.extract_subtitles(
+ author, clip_id, 'en', name, duration, display_id)
return {
'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')),
+ 'title': title,
+ 'duration': duration,
'creator': author,
- 'formats': formats
+ 'formats': formats,
+ 'subtitles': subtitles,
}
Rapsys Git / youtubedl / blobdiff