[fivemin] improve extraction

- skip m3u8 formats(404 error)
- skip unavailable test
- download embed page only when it's needed
- update _VALID_URL regex(joystiq.com redirect to engadget.com)
This commit is contained in:
remitamine 2016-03-04 10:25:16 +01:00
parent e1dd521e49
commit 6d6536acb2

View File

@ -1,5 +1,7 @@
from __future__ import unicode_literals from __future__ import unicode_literals
import re
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import ( from ..compat import (
compat_urllib_parse, compat_urllib_parse,
@ -16,12 +18,7 @@
class FiveMinIE(InfoExtractor): class FiveMinIE(InfoExtractor):
IE_NAME = '5min' IE_NAME = '5min'
_VALID_URL = r'''(?x) _VALID_URL = r'(?:5min:(?P<id>\d+)(?::(?P<sid>\d+))?|https?://[^/]*?5min\.com/Scripts/PlayerSeed\.js\?(?P<query>.*))'
(?:https?://[^/]*?5min\.com/Scripts/PlayerSeed\.js\?(?:.*?&)?playList=|
https?://(?:(?:massively|www)\.)?joystiq\.com/video/|
5min:)
(?P<id>\d+)
'''
_TESTS = [ _TESTS = [
{ {
@ -45,6 +42,7 @@ class FiveMinIE(InfoExtractor):
'title': 'How to Make a Next-Level Fruit Salad', 'title': 'How to Make a Next-Level Fruit Salad',
'duration': 184, 'duration': 184,
}, },
'skip': 'no longer available',
}, },
] ]
_ERRORS = { _ERRORS = {
@ -91,20 +89,33 @@ class FiveMinIE(InfoExtractor):
} }
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
sid = mobj.group('sid')
if mobj.group('query'):
qs = compat_parse_qs(mobj.group('query'))
if not qs.get('playList'):
raise ExtractorError('Invalid URL', expected=True)
video_id = qs['playList'][0]
if qs.get('sid'):
sid = qs['sid'][0]
embed_url = 'https://embed.5min.com/playerseed/?playList=%s' % video_id embed_url = 'https://embed.5min.com/playerseed/?playList=%s' % video_id
if not sid:
embed_page = self._download_webpage(embed_url, video_id, embed_page = self._download_webpage(embed_url, video_id,
'Downloading embed page') 'Downloading embed page')
sid = self._search_regex(r'sid=(\d+)', embed_page, 'sid') sid = self._search_regex(r'sid=(\d+)', embed_page, 'sid')
query = compat_urllib_parse.urlencode({
response = self._download_json(
'https://syn.5min.com/handlers/SenseHandler.ashx?' +
compat_urllib_parse.urlencode({
'func': 'GetResults', 'func': 'GetResults',
'playlist': video_id, 'playlist': video_id,
'sid': sid, 'sid': sid,
'isPlayerSeed': 'true', 'isPlayerSeed': 'true',
'url': embed_url, 'url': embed_url,
}) }),
response = self._download_json(
'https://syn.5min.com/handlers/SenseHandler.ashx?' + query,
video_id) video_id)
if not response['success']: if not response['success']:
raise ExtractorError( raise ExtractorError(
@ -118,9 +129,7 @@ def _real_extract(self, url):
parsed_video_url = compat_urllib_parse_urlparse(compat_parse_qs( parsed_video_url = compat_urllib_parse_urlparse(compat_parse_qs(
compat_urllib_parse_urlparse(info['EmbededURL']).query)['videoUrl'][0]) compat_urllib_parse_urlparse(info['EmbededURL']).query)['videoUrl'][0])
for rendition in info['Renditions']: for rendition in info['Renditions']:
if rendition['RenditionType'] == 'm3u8': if rendition['RenditionType'] == 'aac' or rendition['RenditionType'] == 'm3u8':
formats.extend(self._extract_m3u8_formats(rendition['Url'], video_id, m3u8_id='hls'))
elif rendition['RenditionType'] == 'aac':
continue continue
else: else:
rendition_url = compat_urlparse.urlunparse(parsed_video_url._replace(path=replace_extension(parsed_video_url.path.replace('//', '/%s/' % rendition['ID']), rendition['RenditionType']))) rendition_url = compat_urlparse.urlunparse(parsed_video_url._replace(path=replace_extension(parsed_video_url.path.replace('//', '/%s/' % rendition['ID']), rendition['RenditionType'])))