[ora] minimise fragile regex shenanigans; recognise unsafespeech.com URLs

This commit is contained in:
felix 2016-03-20 12:17:57 +01:00 committed by Sergey M․
parent bd1e484448
commit 640eea0a0c
No known key found for this signature in database
GPG key ID: 2C393E0F18A9236D

View file

@ -6,13 +6,14 @@ from .common import InfoExtractor
from ..compat import compat_urlparse from ..compat import compat_urlparse
from ..utils import ( from ..utils import (
get_element_by_attribute, get_element_by_attribute,
js_to_json,
qualities, qualities,
unescapeHTML, unescapeHTML,
) )
class OraTVIE(InfoExtractor): class OraTVIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?ora\.tv/([^/]+/)*(?P<id>[^/\?#]+)' _VALID_URL = r'https?://(?:www\.)?(ora\.tv|unsafespeech\.com)/([^/]+/)*(?P<id>[^/\?#]+)'
_TEST = { _TEST = {
'url': 'https://www.ora.tv/larrykingnow/2015/12/16/vine-youtube-stars-zach-king-king-bach-on-their-viral-videos-0_36jupg6090pq', 'url': 'https://www.ora.tv/larrykingnow/2015/12/16/vine-youtube-stars-zach-king-king-bach-on-their-viral-videos-0_36jupg6090pq',
'md5': 'fa33717591c631ec93b04b0e330df786', 'md5': 'fa33717591c631ec93b04b0e330df786',
@ -28,10 +29,13 @@ class OraTVIE(InfoExtractor):
display_id = self._match_id(url) display_id = self._match_id(url)
webpage = self._download_webpage(url, display_id) webpage = self._download_webpage(url, display_id)
video_data = self._search_regex( ora_meta = self._parse_json(self._search_regex(
r'"(?:video|current)"\s*:\s*({[^}]+?})', webpage, 'current video') r'(?s);\s*ora_meta = ({.*?});</script>', webpage, 'ora_meta'), display_id,
m3u8_url = self._search_regex( transform_source=lambda data: js_to_json(re.sub('":(document|\().*?(:false|\(\)),', '":null,', data)))
r'hls_stream"?\s*:\s*"([^"]+)', video_data, 'm3u8 url', None)
video_data = ora_meta.get('video', ora_meta.get('current'))
m3u8_url = video_data['hls_stream']
if m3u8_url: if m3u8_url:
formats = self._extract_m3u8_formats( formats = self._extract_m3u8_formats(
m3u8_url, display_id, 'mp4', 'm3u8_native', m3u8_url, display_id, 'mp4', 'm3u8_native',
@ -60,13 +64,11 @@ class OraTVIE(InfoExtractor):
r'"youtube_id"\s*:\s*"([^"]+)', webpage, 'youtube id'), 'Youtube') r'"youtube_id"\s*:\s*"([^"]+)', webpage, 'youtube id'), 'Youtube')
return { return {
'id': self._search_regex( 'id': video_data.get('id', display_id),
r'"id"\s*:\s*(\d+)', video_data, 'video id', default=display_id),
'display_id': display_id, 'display_id': display_id,
'title': unescapeHTML(self._og_search_title(webpage)), 'title': unescapeHTML(self._og_search_title(webpage)),
'description': get_element_by_attribute( 'description': get_element_by_attribute(
'class', 'video_txt_decription', webpage), 'class', 'video_txt_decription', webpage),
'thumbnail': self._proto_relative_url(self._search_regex( 'thumbnail': self._proto_relative_url(video_data.get('thumb')),
r'"thumb"\s*:\s*"([^"]+)', video_data, 'thumbnail', None)),
'formats': formats, 'formats': formats,
} }