Merge remote-tracking branch 'h-collector/master'

Conflicts:
	youtube_dl/extractor/fc2.py
This commit is contained in:
Philipp Hagemeister 2015-01-25 03:37:51 +01:00
commit 8940b8608e

View File

@ -5,6 +5,7 @@ import hashlib
from .common import InfoExtractor
from ..compat import (
compat_urllib_parse,
compat_urllib_request,
compat_urlparse,
)
@ -16,7 +17,8 @@ from ..utils import (
class FC2IE(InfoExtractor):
_VALID_URL = r'^http://video\.fc2\.com/(?:[^/]+/)?content/(?P<id>[^/]+)'
IE_NAME = 'fc2'
_TEST = {
_NETRC_MACHINE = 'fc2'
_TESTS = [{
'url': 'http://video.fc2.com/en/content/20121103kUan1KHs',
'md5': 'a6ebe8ebe0396518689d963774a54eb7',
'info_dict': {
@ -24,12 +26,57 @@ class FC2IE(InfoExtractor):
'ext': 'flv',
'title': 'Boxing again with Puff',
},
}
}, {
'url': 'http://video.fc2.com/en/content/20150125cEva0hDn/',
'info_dict': {
'id': '20150125cEva0hDn',
'ext': 'mp4',
},
'params': {
'username': 'ytdl@yt-dl.org',
'password': '(snip)',
'skip': 'requires actual password'
}
}]
def _login(self):
(username, password) = self._get_login_info()
if username is None or password is None:
return False
# Log in
login_form_strs = {
'email': username,
'password': password,
'done': 'video',
'Submit': ' Login ',
}
# Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
# chokes on unicode
login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
request = compat_urllib_request.Request(
'https://secure.id.fc2.com/index.php?mode=login&switch_language=en', login_data)
login_results = self._download_webpage(request, None, note='Logging in', errnote='Unable to log in')
if 'mode=redirect&login=done' not in login_results:
self.report_warning('unable to log in: bad username or password')
return False
# this is also needed
login_redir = compat_urllib_request.Request('http://id.fc2.com/?mode=redirect&login=done')
self._download_webpage(
login_redir, None, note='Login redirect', errnote='Login redirect failed')
return True
def _real_extract(self, url):
video_id = self._match_id(url)
self._login()
webpage = self._download_webpage(url, video_id)
self._downloader.cookiejar.clear_session_cookies() # must clear
self._login()
title = self._og_search_title(webpage)
thumbnail = self._og_search_thumbnail(webpage)
@ -46,7 +93,12 @@ class FC2IE(InfoExtractor):
info = compat_urlparse.parse_qs(info_webpage)
if 'err_code' in info:
raise ExtractorError('Error code: %s' % info['err_code'][0])
# most of the time we can still download wideo even if err_code is 403 or 602
self.report_warning(
'Error code was: %s... but still trying' % info['err_code'][0])
if 'filepath' not in info:
raise ExtractorError('Cannot download file. Are you logged in?')
video_url = info['filepath'][0] + '?mid=' + info['mid'][0]
title_info = info.get('title')