kumi/yt-dlp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

[globo] Fix GloboIE (#994)

Browse source 
Closes #991 
Authored by: Ashish0804
This commit is contained in:
Ashish Gupta 2021-09-16 23:01:39 +05:30 committed by GitHub
parent 0001fcb586
commit b89378a69a
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 63 additions and 90 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

111
yt_dlp/extractor/globo.py
View file

@ -9,15 +9,14 @@ import re
from .common import InfoExtractor
from ..compat import (
compat_HTTPError,
compat_str,
)
from ..utils import (
ExtractorError,
float_or_none,
int_or_none,
orderedSet,
str_or_none,
try_get,
)
@ -26,18 +25,19 @@ class GloboIE(InfoExtractor):
_NETRC_MACHINE = 'globo'
_TESTS = [{
'url': 'http://g1.globo.com/carros/autoesporte/videos/t/exclusivos-do-g1/v/mercedes-benz-gla-passa-por-teste-de-colisao-na-europa/3607726/',
'md5': 'b3ccc801f75cd04a914d51dadb83a78d',
'info_dict': {
'id': '3607726',
'ext': 'mp4',
'title': 'Mercedes-Benz GLA passa por teste de colisão na Europa',
'duration': 103.204,
'uploader': 'Globo.com',
'uploader_id': '265',
'uploader': 'G1',
'uploader_id': '2015',
},
'params': {
'skip_download': True,
},
}, {
'url': 'http://globoplay.globo.com/v/4581987/',
'md5': 'f36a1ecd6a50da1577eee6dd17f67eff',
'info_dict': {
'id': '4581987',
'ext': 'mp4',
@ -46,6 +46,9 @@ class GloboIE(InfoExtractor):
'uploader': 'Rede Globo',
'uploader_id': '196',
},
'params': {
'skip_download': True,
},
}, {
'url': 'http://canalbrasil.globo.com/programas/sangue-latino/videos/3928201.html',
'only_matching': True,
@ -66,30 +69,6 @@ class GloboIE(InfoExtractor):
'only_matching': True,
}]
def _real_initialize(self):
email, password = self._get_login_info()
if email is None:
return
try:
glb_id = (self._download_json(
'https://login.globo.com/api/authentication', None, data=json.dumps({
'payload': {
'email': email,
'password': password,
'serviceId': 4654,
},
}).encode(), headers={
'Content-Type': 'application/json; charset=utf-8',
}) or {}).get('glbId')
if glb_id:
self._set_cookie('.globo.com', 'GLBID', glb_id)
except ExtractorError as e:
if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
resp = self._parse_json(e.cause.read(), None)
raise ExtractorError(resp.get('userMessage') or resp['id'], expected=True)
raise
def _real_extract(self, url):
video_id = self._match_id(url)
@ -102,35 +81,23 @@ class GloboIE(InfoExtractor):
title = video['title']
formats = []
subtitles = {}
for resource in video['resources']:
resource_id = resource.get('_id')
resource_url = resource.get('url')
resource_type = resource.get('type')
if not resource_url or (resource_type == 'media' and not resource_id) or resource_type not in ('subtitle', 'media'):
continue
if resource_type == 'subtitle':
subtitles.setdefault(resource.get('language') or 'por', []).append({
'url': resource_url,
})
continue
security = self._download_json(
'http://security.video.globo.com/videos/%s/hash' % video_id,
video_id, 'Downloading security hash for %s' % resource_id, query={
'player': 'desktop',
'version': '5.19.1',
'resource_id': resource_id,
})
'https://playback.video.globo.com/v1/video-session', video_id, 'Downloading security hash for %s' % video_id,
headers={'content-type': 'application/json'}, data=json.dumps({
"player_type": "desktop",
"video_id": video_id,
"quality": "max",
"content_protection": "widevine",
"vsid": "581b986b-4c40-71f0-5a58-803e579d5fa2",
"tz": "-3.0:00"
}).encode())
security_hash = security.get('hash')
security_hash = security['source']['token']
if not security_hash:
message = security.get('message')
if message:
raise ExtractorError(
'%s returned error: %s' % (self.IE_NAME, message), expected=True)
continue
hash_code = security_hash[:2]
padding = '%010d' % random.randint(1, 10000000000)
@ -148,27 +115,33 @@ class GloboIE(InfoExtractor):
md5_data = (received_md5 + padded_sign_time + '0xAC10FD').encode()
signed_md5 = base64.urlsafe_b64encode(hashlib.md5(md5_data).digest()).decode().strip('=')
signed_hash = hash_prefix + padded_sign_time + signed_md5
signed_url = '%s?h=%s&k=html5&a=%s&u=%s' % (resource_url, signed_hash, 'F' if video.get('subscriber_only') else 'A', security.get('user') or '')
source = security['source']['url_parts']
resource_url = source['scheme'] + '://' + source['domain'] + source['path']
signed_url = '%s?h=%s&k=html5&a=%s' % (resource_url, signed_hash, 'F' if video.get('subscriber_only') else 'A')
if resource_id.endswith('m3u8') or resource_url.endswith('.m3u8'):
formats.extend(self._extract_m3u8_formats(
signed_url, resource_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id='hls', fatal=False))
elif resource_id.endswith('mpd') or resource_url.endswith('.mpd'):
formats.extend(self._extract_mpd_formats(
signed_url, resource_id, mpd_id='dash', fatal=False))
elif resource_id.endswith('manifest') or resource_url.endswith('/manifest'):
formats.extend(self._extract_ism_formats(
signed_url, resource_id, ism_id='mss', fatal=False))
else:
formats.append({
'url': signed_url,
'format_id': 'http-%s' % resource_id,
'height': int_or_none(resource.get('height')),
})
signed_url, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
self._sort_formats(formats)
subtitles = {}
for resource in video['resources']:
if resource.get('type') == 'subtitle':
subtitles.setdefault(resource.get('language') or 'por', []).append({
'url': resource.get('url'),
})
subs = try_get(security, lambda x: x['source']['subtitles'], expected_type=dict) or {}
for sub_lang, sub_url in subs.items():
if sub_url:
subtitles.setdefault(sub_lang or 'por', []).append({
'url': sub_url,
})
subs = try_get(security, lambda x: x['source']['subtitles_webvtt'], expected_type=dict) or {}
for sub_lang, sub_url in subs.items():
if sub_url:
subtitles.setdefault(sub_lang or 'por', []).append({
'url': sub_url,
})
duration = float_or_none(video.get('duration'), 1000)
uploader = video.get('channel')
uploader_id = str_or_none(video.get('channel_id'))