1
0
mirror of https://github.com/l1ving/youtube-dl synced 2020-11-18 19:53:54 -08:00

use regex instead

This commit is contained in:
Julien Lhermitte 2019-04-07 14:35:57 -04:00
parent 6af62a164c
commit d4651f3926

View File

@ -9,9 +9,7 @@ from ..utils import (
int_or_none, int_or_none,
url_or_none, url_or_none,
) )
from urllib.parse import urlparse
from urllib.request import urlopen from urllib.request import urlopen
import lxml.html
class AolIE(InfoExtractor): class AolIE(InfoExtractor):
@ -69,27 +67,15 @@ class AolIE(InfoExtractor):
return urlopen(url).read() return urlopen(url).read()
def _download_and_extract_video_id_from_page(self, url): def _download_and_extract_video_id_from_page(self, url):
page_bytes = self._download_page(url) video_id_regex = r'<script [^>]*src="[^"]*vid=(?P<vid>[^?/]*)'
tree = lxml.html.fromstring(page_bytes) page_bytes = self._download_page(url).decode()
src_xpath = tree.xpath("//script[contains(@src, " mobj = re.search(video_id_regex, page_bytes)
"'delivery.vidible.tv')]") return mobj.group('vid')
src_tag = src_xpath[0].attrib.get('src')
parsed_vid_url = urlparse(src_tag)
vid_url_path = parsed_vid_url.path
vid_url_params = vid_url_path.split('/')
return self._find_vid_param(vid_url_params)
def _find_vid_param(self, vid_url_params: [str]):
for param in vid_url_params:
if param.startswith('vid='):
return param.split('=')[1]
return None
def _real_extract(self, url): def _real_extract(self, url):
# video_id = self._match_id(url)
print(f'getting video_id')
video_id = self._download_and_extract_video_id_from_page(url) video_id = self._download_and_extract_video_id_from_page(url)
print(f'video_id: {video_id}') if video_id is None:
raise ExtractorError('Could not find video_id')
response = self._download_json( response = self._download_json(
'https://feedapi.b2c.on.aol.com/v1.0/app/videos/aolon/%s/details' % video_id, 'https://feedapi.b2c.on.aol.com/v1.0/app/videos/aolon/%s/details' % video_id,