mirror of
https://github.com/ytdl-org/youtube-dl.git
synced 2024-12-20 03:54:01 +00:00
[ora] Add new extractor(closes #7732)
This commit is contained in:
parent
80b8b72cb8
commit
9d46608efa
@ -491,6 +491,7 @@ from .ooyala import (
|
||||
OoyalaIE,
|
||||
OoyalaExternalIE,
|
||||
)
|
||||
from .ora import OraTVIE
|
||||
from .orf import (
|
||||
ORFTVthekIE,
|
||||
ORFOE1IE,
|
||||
|
75
youtube_dl/extractor/ora.py
Normal file
75
youtube_dl/extractor/ora.py
Normal file
@ -0,0 +1,75 @@
|
||||
# coding: utf-8
|
||||
from __future__ import unicode_literals
|
||||
|
||||
import re
|
||||
from .common import InfoExtractor
|
||||
from ..compat import compat_urlparse
|
||||
from ..utils import (
|
||||
get_element_by_attribute,
|
||||
qualities,
|
||||
unescapeHTML,
|
||||
)
|
||||
|
||||
|
||||
class OraTVIE(InfoExtractor):
|
||||
_VALID_URL = r'https?://(?:www\.)?ora\.tv/([^/]+/)*(?P<id>[^/\?#]+)'
|
||||
_TEST = {
|
||||
'url': 'https://www.ora.tv/larrykingnow/2015/12/16/vine-youtube-stars-zach-king-king-bach-on-their-viral-videos-0_36jupg6090pq',
|
||||
'md5': 'fa33717591c631ec93b04b0e330df786',
|
||||
'info_dict': {
|
||||
'id': '50178',
|
||||
'ext': 'mp4',
|
||||
'title': 'Vine & YouTube Stars Zach King & King Bach On Their Viral Videos!',
|
||||
'description': 'md5:ebbc5b1424dd5dba7be7538148287ac1',
|
||||
'duration': 1477,
|
||||
}
|
||||
}
|
||||
|
||||
def _real_extract(self, url):
|
||||
display_id = self._match_id(url)
|
||||
webpage = self._download_webpage(url, display_id)
|
||||
|
||||
video_data = self._search_regex(
|
||||
r'"current"\s*:\s*({[^}]+?})', webpage, 'current video')
|
||||
m3u8_url = self._search_regex(
|
||||
r'"hls_stream"\s*:\s*"([^"]+)', video_data, 'm3u8 url', None)
|
||||
if m3u8_url:
|
||||
formats = self._extract_m3u8_formats(
|
||||
m3u8_url, display_id, 'mp4', 'm3u8_native',
|
||||
m3u8_id='hls', fatal=False)
|
||||
# simular to GameSpotIE
|
||||
m3u8_path = compat_urlparse.urlparse(m3u8_url).path
|
||||
QUALITIES_RE = r'((,[a-z]+\d+)+,?)'
|
||||
available_qualities = self._search_regex(
|
||||
QUALITIES_RE, m3u8_path, 'qualities').strip(',').split(',')
|
||||
http_path = m3u8_path[1:].split('/', 1)[1]
|
||||
http_template = re.sub(QUALITIES_RE, r'%s', http_path)
|
||||
http_template = http_template.replace('.csmil/master.m3u8', '')
|
||||
http_template = compat_urlparse.urljoin(
|
||||
'http://videocdn-pmd.ora.tv/', http_template)
|
||||
preference = qualities(
|
||||
['mobile400', 'basic400', 'basic600', 'sd900', 'sd1200', 'sd1500', 'hd720', 'hd1080'])
|
||||
for q in available_qualities:
|
||||
formats.append({
|
||||
'url': http_template % q,
|
||||
'format_id': q,
|
||||
'preference': preference(q),
|
||||
})
|
||||
self._sort_formats(formats)
|
||||
else:
|
||||
return self.url_result(self._search_regex(
|
||||
r'"youtube_id"\s*:\s*"([^"]+)', webpage, 'youtube id'), 'Youtube')
|
||||
|
||||
return {
|
||||
'id': self._search_regex(
|
||||
r'"video_id"\s*:\s*(\d+)', video_data, 'video id'),
|
||||
'display_id': display_id,
|
||||
'title': unescapeHTML(self._og_search_title(webpage)),
|
||||
'description': get_element_by_attribute(
|
||||
'class', 'video_txt_decription', webpage),
|
||||
'thumbnail': self._proto_relative_url(self._search_regex(
|
||||
r'"thumb"\s*:\s*"([^"]+)', video_data, 'thumbnail', None)),
|
||||
'duration': int(self._search_regex(
|
||||
r'"duration"\s*:\s*(\d+)', video_data, 'duration')),
|
||||
'formats': formats,
|
||||
}
|
Loading…
Reference in New Issue
Block a user