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

[mgtv] fix extraction

Browse source
This commit is contained in:
Remita Amine 2017-02-23 11:50:04 +01:00
parent c59f703610
commit 30eaa3a702
1 changed files with 24 additions and 26 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

46
youtube_dl/extractor/mgtv.py
View file

@ -2,16 +2,17 @@
from __future__ import unicode_literals from __future__ import unicode_literals
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import compat_str
from ..utils import int_or_none from ..utils import int_or_none
class MGTVIE(InfoExtractor): class MGTVIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?mgtv\.com/v/(?:[^/]+/)*(?P<id>\d+)\.html' _VALID_URL = r'https?://(?:www\.)?mgtv\.com/(v|b)/(?:[^/]+/)*(?P<id>\d+)\.html'
IE_DESC = '芒果TV' IE_DESC = '芒果TV'
_TESTS = [{ _TESTS = [{
'url': 'http://www.mgtv.com/v/1/290525/f/3116640.html', 'url': 'http://www.mgtv.com/v/1/290525/f/3116640.html',
'md5': '1bdadcf760a0b90946ca68ee9a2db41a', 'md5': 'b1ffc0fc163152acf6beaa81832c9ee7',
'info_dict': { 'info_dict': {
'id': '3116640', 'id': '3116640',
'ext': 'mp4', 'ext': 'mp4',
@ -21,48 +22,45 @@ class MGTVIE(InfoExtractor):
'thumbnail': r're:^https?://.*\.jpg$', 'thumbnail': r're:^https?://.*\.jpg$',
}, },
}, { }, {
# no tbr extracted from stream_url 'url': 'http://www.mgtv.com/b/301817/3826653.html',
'url': 'http://www.mgtv.com/v/1/1/f/3324755.html',
'only_matching': True, 'only_matching': True,
}] }]
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
api_data = self._download_json( api_data = self._download_json(
'http://v.api.mgtv.com/player/video', video_id, 'http://pcweb.api.mgtv.com/player/video', video_id,
query={'video_id': video_id}, query={'video_id': video_id},
headers=self.geo_verification_headers())['data'] headers=self.geo_verification_headers())['data']
info = api_data['info'] info = api_data['info']
title = info['title'].strip()
stream_domain = api_data['stream_domain'][0]
formats = [] formats = []
for idx, stream in enumerate(api_data['stream']): for idx, stream in enumerate(api_data['stream']):
stream_url = stream.get('url') stream_path = stream.get('url')
if not stream_url: if not stream_path:
continue
format_data = self._download_json(
stream_domain + stream_path, video_id,
note='Download video info for format #%d' % idx)
format_url = format_data.get('info')
if not format_url:
continue continue
tbr = int_or_none(self._search_regex( tbr = int_or_none(self._search_regex(
r'(\d+)\.mp4', stream_url, 'tbr', default=None)) r'_(\d+)_mp4/', format_url, 'tbr', default=None))
formats.append({
def extract_format(stream_url, format_id, idx, query={}): 'format_id': compat_str(tbr or idx),
format_info = self._download_json( 'url': format_url,
stream_url, video_id,
note='Download video info for format %s' % (format_id or '#%d' % idx),
query=query)
return {
'format_id': format_id,
'url': format_info['info'],
'ext': 'mp4', 'ext': 'mp4',
'tbr': tbr, 'tbr': tbr,
} 'protocol': 'm3u8_native',
})
formats.append(extract_format(
stream_url, 'hls-%d' % tbr if tbr else None, idx * 2))
formats.append(extract_format(stream_url.replace(
'/playlist.m3u8', ''), 'http-%d' % tbr if tbr else None, idx * 2 + 1, {'pno': 1031}))
self._sort_formats(formats) self._sort_formats(formats)
return { return {
'id': video_id, 'id': video_id,
'title': info['title'].strip(), 'title': title,
'formats': formats, 'formats': formats,
'description': info.get('desc'), 'description': info.get('desc'),
'duration': int_or_none(info.get('duration')), 'duration': int_or_none(info.get('duration')),