paradizelost/youtube-dl
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[lynda] Convert to new subtitles system

Browse Source
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
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
test/test_subtitles.py
View File

@ -18,6 +18,7 @@ from youtube_dl.extractor import (
VimeoIE,
WallaIE,
CeskaTelevizeIE,
LyndaIE,
)
@ -304,5 +305,17 @@ class TestCeskaTelevizeSubtitles(BaseTestSubtitles):
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__':
unittest.main()

60
youtube_dl/extractor/lynda.py
View File

@ -3,7 +3,6 @@ from __future__ import unicode_literals
import re
import json
from .subtitles import SubtitlesInfoExtractor
from .common import InfoExtractor
from ..compat import (
compat_str,
@ -16,7 +15,7 @@ from ..utils import (
)
class LyndaIE(SubtitlesInfoExtractor):
class LyndaIE(InfoExtractor):
IE_NAME = 'lynda'
IE_DESC = 'lynda.com videos'
_VALID_URL = r'https?://www\.lynda\.com/[^/]+/[^/]+/\d+/(\d+)-\d\.html'
@ -88,11 +87,7 @@ class LyndaIE(SubtitlesInfoExtractor):
self._check_formats(formats, video_id)
self._sort_formats(formats)
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))
subtitles = self.extract_subtitles(video_id, page)
return {
'id': video_id,
@ -144,38 +139,31 @@ class LyndaIE(SubtitlesInfoExtractor):
if re.search(self._SUCCESSFUL_LOGIN_REGEX, login_page) is None:
raise ExtractorError('Unable to log in')
def _fix_subtitles(self, subtitles):
if subtitles is None:
return subtitles # subtitles not requested
fixed_subtitles = {}
for k, v in subtitles.items():
subs = json.loads(v)
if len(subs) == 0:
def _fix_subtitles(self, subs):
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
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
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:
return srt
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
sub = self._download_webpage(url, None, False)
sub_json = json.loads(sub)
return {'en': url} if len(sub_json) > 0 else {}
subs = self._download_json(url, None, False)
if subs:
return {'en': [{'ext': 'srt', 'data': self._fix_subtitles(subs)}]}
else:
return {}
class LyndaCourseIE(InfoExtractor):