youtube-dl/youtube_dl/extractor/lynda.py

143 lines
5.2 KiB
Python
Raw Normal View History

2014-01-03 12:24:29 +00:00
from __future__ import unicode_literals
import re
import json
2014-01-05 16:59:33 +00:00
from .subtitles import SubtitlesInfoExtractor
from .common import InfoExtractor
from ..utils import ExtractorError
2014-01-05 16:59:33 +00:00
class LyndaIE(SubtitlesInfoExtractor):
2014-01-03 12:24:29 +00:00
IE_NAME = 'lynda'
IE_DESC = 'lynda.com videos'
_VALID_URL = r'https?://www\.lynda\.com/[^/]+/[^/]+/\d+/(\d+)-\d\.html'
_TEST = {
2014-01-03 12:24:29 +00:00
'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
'file': '114408.mp4',
'md5': 'ecfc6862da89489161fb9cd5f5a6fac1',
2014-01-07 06:14:12 +00:00
'info_dict': {
2014-01-03 12:24:29 +00:00
'title': 'Using the exercise files',
'duration': 68
}
}
2014-01-05 16:59:33 +00:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group(1)
page = self._download_webpage('http://www.lynda.com/ajax/player?videoId=%s&type=video' % video_id,
2014-01-03 12:24:29 +00:00
video_id, 'Downloading video JSON')
video_json = json.loads(page)
2014-01-03 12:24:29 +00:00
if 'Status' in video_json and video_json['Status'] == 'NotFound':
raise ExtractorError('Video %s does not exist' % video_id, expected=True)
2014-01-03 12:24:29 +00:00
if video_json['HasAccess'] is False:
raise ExtractorError('Video %s is only available for members' % video_id, expected=True)
2014-01-03 12:24:29 +00:00
video_id = video_json['ID']
duration = video_json['DurationInSeconds']
title = video_json['Title']
2014-01-03 12:24:29 +00:00
formats = [{'url': fmt['Url'],
'ext': fmt['Extension'],
'width': fmt['Width'],
'height': fmt['Height'],
'filesize': fmt['FileSize'],
2014-01-05 16:59:33 +00:00
'format_id': str(fmt['Resolution'])
2014-01-03 12:24:29 +00:00
} for fmt in video_json['Formats']]
self._sort_formats(formats)
2014-01-05 16:59:33 +00:00
if self._downloader.params.get('listsubtitles', False):
self._list_available_subtitles(video_id, page)
return
subtitles = self._fix_subtitles(self.extract_subtitles(video_id, page))
return {
'id': video_id,
'title': title,
'duration': duration,
2014-01-05 16:59:33 +00:00
'subtitles': subtitles,
'formats': formats
}
2014-01-05 16:59:33 +00:00
_TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
def _fix_subtitles(self, subtitles):
fixed_subtitles = {}
for k, v in subtitles.items():
subs = json.loads(v)
if len(subs) == 0:
continue
srt = ''
for pos in range(0, len(subs) - 1):
seq_current = subs[pos]
m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode'])
if m_current is None:
continue
seq_next = subs[pos+1]
m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode'])
if m_next is None:
continue
appear_time = m_current.group('timecode')
disappear_time = m_next.group('timecode')
text = seq_current['Caption']
srt += '%s\r\n%s --> %s\r\n%s' % (str(pos), appear_time, disappear_time, text)
if srt:
fixed_subtitles[k] = srt
return fixed_subtitles
def _get_available_subtitles(self, video_id, webpage):
url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
sub = self._download_webpage(url, None, note=False)
sub_json = json.loads(sub)
return {'en': url} if len(sub_json) > 0 else {}
class LyndaCourseIE(InfoExtractor):
2014-01-03 12:24:29 +00:00
IE_NAME = 'lynda:course'
IE_DESC = 'lynda.com online courses'
# Course link equals to welcome/introduction video link of same course
# We will recognize it as course link
_VALID_URL = r'https?://(?:www|m)\.lynda\.com/(?P<coursepath>[^/]+/[^/]+/(?P<courseid>\d+))-\d\.html'
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
course_path = mobj.group('coursepath')
course_id = mobj.group('courseid')
page = self._download_webpage('http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
2014-01-03 12:24:29 +00:00
course_id, 'Downloading course JSON')
course_json = json.loads(page)
2014-01-03 12:24:29 +00:00
if 'Status' in course_json and course_json['Status'] == 'NotFound':
raise ExtractorError('Course %s does not exist' % course_id, expected=True)
unaccessible_videos = 0
videos = []
2014-01-03 12:24:29 +00:00
for chapter in course_json['Chapters']:
for video in chapter['Videos']:
if video['HasAccess'] is not True:
unaccessible_videos += 1
continue
2014-01-03 12:24:29 +00:00
videos.append(video['ID'])
if unaccessible_videos > 0:
2014-01-03 12:24:29 +00:00
self._downloader.report_warning('%s videos are only available for members and will not be downloaded' % unaccessible_videos)
2014-01-03 12:24:29 +00:00
entries = [
self.url_result('http://www.lynda.com/%s/%s-4.html' %
(course_path, video_id),
'Lynda')
for video_id in videos]
2014-01-03 12:24:29 +00:00
course_title = course_json['Title']
2014-01-03 12:24:29 +00:00
return self.playlist_result(entries, course_id, course_title)