Compare commits

...

4 Commits

Author SHA1 Message Date
Philipp Hagemeister
dae313e725 release 2014.03.07.1 2014-03-07 15:59:10 +01:00
Jaime Marquínez Ferrándiz
b74fa8cd2c [facebook] Fix login process
It was broken and didn't work in python 3.
And use `_download_webpage` instead of `compat_urllib_request.urlopen`.
2014-03-07 15:25:33 +01:00
Philipp Hagemeister
94eae04c94 release 2014.03.07 2014-03-07 06:41:48 +01:00
Sergey M․
16ff7ebc77 [lynda] Fix successful login regex and fix formats extraction (Closes #2520) 2014-03-07 06:56:48 +07:00
5 changed files with 45 additions and 13 deletions

View File

@@ -33,6 +33,7 @@ from youtube_dl.utils import (
unified_strdate,
unsmuggle_url,
url_basename,
urlencode_postdata,
xpath_with_ns,
)
@@ -261,5 +262,9 @@ class TestUtil(unittest.TestCase):
bam''')
self.assertEqual(read_batch_urls(f), [u'foo', u'bar', u'baz', u'bam'])
def test_urlencode_postdata(self):
data = urlencode_postdata({'username': 'foo@bar.com', 'password': '1234'})
self.assertTrue(isinstance(data, bytes))
if __name__ == '__main__':
unittest.main()

View File

@@ -11,6 +11,7 @@ from ..utils import (
compat_urllib_error,
compat_urllib_parse,
compat_urllib_request,
urlencode_postdata,
ExtractorError,
)
@@ -51,8 +52,8 @@ class FacebookIE(InfoExtractor):
login_page_req = compat_urllib_request.Request(self._LOGIN_URL)
login_page_req.add_header('Cookie', 'locale=en_US')
self.report_login()
login_page = self._download_webpage(login_page_req, None, note=False,
login_page = self._download_webpage(login_page_req, None,
note='Downloading login page',
errnote='Unable to download login page')
lsd = self._search_regex(
r'<input type="hidden" name="lsd" value="([^"]*)"',
@@ -70,23 +71,25 @@ class FacebookIE(InfoExtractor):
'timezone': '-60',
'trynum': '1',
}
request = compat_urllib_request.Request(self._LOGIN_URL, compat_urllib_parse.urlencode(login_form))
request = compat_urllib_request.Request(self._LOGIN_URL, urlencode_postdata(login_form))
request.add_header('Content-Type', 'application/x-www-form-urlencoded')
try:
login_results = compat_urllib_request.urlopen(request).read()
login_results = self._download_webpage(request, None,
note='Logging in', errnote='unable to fetch login page')
if re.search(r'<form(.*)name="login"(.*)</form>', login_results) is not None:
self._downloader.report_warning('unable to log in: bad username/password, or exceded login rate limit (~3/min). Check credentials or wait.')
return
check_form = {
'fb_dtsg': self._search_regex(r'"fb_dtsg":"(.*?)"', login_results, 'fb_dtsg'),
'fb_dtsg': self._search_regex(r'name="fb_dtsg" value="(.+?)"', login_results, 'fb_dtsg'),
'nh': self._search_regex(r'name="nh" value="(\w*?)"', login_results, 'nh'),
'name_action_selected': 'dont_save',
'submit[Continue]': self._search_regex(r'<input value="(.*?)" name="submit\[Continue\]"', login_results, 'continue'),
'submit[Continue]': self._search_regex(r'<button[^>]+value="(.*?)"[^>]+name="submit\[Continue\]"', login_results, 'continue'),
}
check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, compat_urllib_parse.urlencode(check_form))
check_req = compat_urllib_request.Request(self._CHECKPOINT_URL, urlencode_postdata(check_form))
check_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
check_response = compat_urllib_request.urlopen(check_req).read()
check_response = self._download_webpage(check_req, None,
note='Confirming login')
if re.search(r'id="checkpointSubmitButton"', check_response) is not None:
self._downloader.report_warning('Unable to confirm login, you have to login in your brower and authorize the login.')
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:

View File

@@ -8,7 +8,8 @@ from .common import InfoExtractor
from ..utils import (
compat_urllib_parse,
compat_urllib_request,
ExtractorError
ExtractorError,
int_or_none,
)
@@ -19,7 +20,7 @@ class LyndaIE(SubtitlesInfoExtractor):
_LOGIN_URL = 'https://www.lynda.com/login/login.aspx'
_NETRC_MACHINE = 'lynda'
_SUCCESSFUL_LOGIN_REGEX = r'<a href="https://www.lynda.com/home/userAccount/ChangeContactInfo.aspx" data-qa="eyebrow_account_menu">My account'
_SUCCESSFUL_LOGIN_REGEX = r'isLoggedIn: true'
_TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
@@ -55,13 +56,29 @@ class LyndaIE(SubtitlesInfoExtractor):
duration = video_json['DurationInSeconds']
title = video_json['Title']
formats = [{'url': fmt['Url'],
formats = []
fmts = video_json.get('Formats')
if fmts:
formats.extend([
{
'url': fmt['Url'],
'ext': fmt['Extension'],
'width': fmt['Width'],
'height': fmt['Height'],
'filesize': fmt['FileSize'],
'format_id': str(fmt['Resolution'])
} for fmt in video_json['Formats']]
} for fmt in fmts])
prioritized_streams = video_json.get('PrioritizedStreams')
if prioritized_streams:
formats.extend([
{
'url': video_url,
'width': int_or_none(format_id),
'format_id': format_id,
} for format_id, video_url in prioritized_streams['0'].items()
])
self._sort_formats(formats)
@@ -179,6 +196,9 @@ class LyndaCourseIE(InfoExtractor):
videos = []
(username, _) = self._get_login_info()
# Might want to extract videos right here from video['Formats'] as it seems 'Formats' is not provided
# by single video API anymore
for chapter in course_json['Chapters']:
for video in chapter['Videos']:
if username is None and video['HasAccess'] is False:

View File

@@ -1263,3 +1263,7 @@ def read_batch_urls(batch_fd):
with contextlib.closing(batch_fd) as fd:
return [url for url in map(fixup, fd) if url]
def urlencode_postdata(*args, **kargs):
return compat_urllib_parse.urlencode(*args, **kargs).encode('ascii')

View File

@@ -1,2 +1,2 @@
__version__ = '2014.03.06'
__version__ = '2014.03.07.1'