from __future__ import unicode_literals
+import collections
import json
+import os
import random
-import collections
+import re
from .common import InfoExtractor
from ..compat import (
compat_str,
- compat_urllib_parse,
- compat_urllib_request,
compat_urlparse,
)
from ..utils import (
ExtractorError,
+ float_or_none,
int_or_none,
parse_duration,
+ qualities,
+ srt_subtitles_timecode,
+ urlencode_postdata,
)
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(
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.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/training/Player/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 = ''
+ for num, current in enumerate(subs):
+ current = subs[num]
+ start, text = float_or_none(
+ current.get('DisplayTimeOffset')), current.get('Text')
+ 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'))
+ 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)
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},
'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.
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,
'mt': ext,
'q': '%dx%d' % (f['width'], f['height']),
}
- request = compat_urllib_request.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)
+ '%s/training/Player/ViewClip' % self._API_BASE, display_id,
+ 'Downloading %s URL' % 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
'url': clip_url,
'ext': ext,
'format_id': format_id,
+ 'quality': quality_key(quality),
})
formats.append(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['clipName'],
- 'title': '%s - %s' % (module['title'], clip['title']),
- 'duration': int_or_none(clip.get('duration')) or parse_duration(clip.get('formattedDuration')),
+ 'id': clip.get('clipName') or clip['name'],
+ 'title': title,
+ 'duration': duration,
'creator': author,
- 'formats': formats
+ 'formats': formats,
+ 'subtitles': subtitles,
}
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
- entries.append(self.url_result(
- '%s/training/player?%s' % (self._API_BASE, 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)
projects / youtube-dl / blobdiff