[lynda] Convert to new subtitles system

This commit is contained in:
Jaime Marquínez Ferrándiz 2015-02-18 17:46:33 +01:00
parent 18c1c42405
commit 311c393838
2 changed files with 37 additions and 36 deletions

View file

@ -18,6 +18,7 @@
VimeoIE, VimeoIE,
WallaIE, WallaIE,
CeskaTelevizeIE, CeskaTelevizeIE,
LyndaIE,
) )
@ -304,5 +305,17 @@ def test_nosubtitles(self):
self.assertEqual(len(subtitles), 0) self.assertEqual(len(subtitles), 0)
class TestLyndaSubtitles(BaseTestSubtitles):
url = 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html'
IE = LyndaIE
def test_allsubtitles(self):
self.DL.params['writesubtitles'] = True
self.DL.params['allsubtitles'] = True
subtitles = self.getSubtitles()
self.assertEqual(set(subtitles.keys()), set(['en']))
self.assertEqual(md5(subtitles['en']), '09bbe67222259bed60deaa26997d73a7')
if __name__ == '__main__': if __name__ == '__main__':
unittest.main() unittest.main()

View file

@ -3,7 +3,6 @@
import re import re
import json import json
from .subtitles import SubtitlesInfoExtractor
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import ( from ..compat import (
compat_str, compat_str,
@ -16,7 +15,7 @@
) )
class LyndaIE(SubtitlesInfoExtractor): class LyndaIE(InfoExtractor):
IE_NAME = 'lynda' IE_NAME = 'lynda'
IE_DESC = 'lynda.com videos' IE_DESC = 'lynda.com videos'
_VALID_URL = r'https?://www\.lynda\.com/[^/]+/[^/]+/\d+/(\d+)-\d\.html' _VALID_URL = r'https?://www\.lynda\.com/[^/]+/[^/]+/\d+/(\d+)-\d\.html'
@ -88,11 +87,7 @@ def _real_extract(self, url):
self._check_formats(formats, video_id) self._check_formats(formats, video_id)
self._sort_formats(formats) self._sort_formats(formats)
if self._downloader.params.get('listsubtitles', False): subtitles = self.extract_subtitles(video_id, page)
self._list_available_subtitles(video_id, page)
return
subtitles = self._fix_subtitles(self.extract_subtitles(video_id, page))
return { return {
'id': video_id, 'id': video_id,
@ -144,38 +139,31 @@ def _login(self):
if re.search(self._SUCCESSFUL_LOGIN_REGEX, login_page) is None: if re.search(self._SUCCESSFUL_LOGIN_REGEX, login_page) is None:
raise ExtractorError('Unable to log in') raise ExtractorError('Unable to log in')
def _fix_subtitles(self, subtitles): def _fix_subtitles(self, subs):
if subtitles is None: srt = ''
return subtitles # subtitles not requested for pos in range(0, len(subs) - 1):
seq_current = subs[pos]
fixed_subtitles = {} m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode'])
for k, v in subtitles.items(): if m_current is None:
subs = json.loads(v)
if len(subs) == 0:
continue continue
srt = '' seq_next = subs[pos + 1]
for pos in range(0, len(subs) - 1): m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode'])
seq_current = subs[pos] if m_next is None:
m_current = re.match(self._TIMECODE_REGEX, seq_current['Timecode']) continue
if m_current is None: appear_time = m_current.group('timecode')
continue disappear_time = m_next.group('timecode')
seq_next = subs[pos + 1] text = seq_current['Caption']
m_next = re.match(self._TIMECODE_REGEX, seq_next['Timecode']) srt += '%s\r\n%s --> %s\r\n%s' % (str(pos), appear_time, disappear_time, text)
if m_next is None: if srt:
continue return srt
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): def _get_subtitles(self, video_id, webpage):
url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id url = 'http://www.lynda.com/ajax/player?videoId=%s&type=transcript' % video_id
sub = self._download_webpage(url, None, False) subs = self._download_json(url, None, False)
sub_json = json.loads(sub) if subs:
return {'en': url} if len(sub_json) > 0 else {} return {'en': [{'ext': 'srt', 'data': self._fix_subtitles(subs)}]}
else:
return {}
class LyndaCourseIE(InfoExtractor): class LyndaCourseIE(InfoExtractor):