[newstube] extract http formats(closes #9253)

This commit is contained in:
remitamine 2016-04-22 11:26:43 +01:00
parent 7c36ea7d54
commit e962ae15d3
1 changed files with 34 additions and 9 deletions

View File

@ -4,24 +4,24 @@ from __future__ import unicode_literals
import re import re
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ExtractorError from ..utils import (
ExtractorError,
int_or_none,
)
class NewstubeIE(InfoExtractor): class NewstubeIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?newstube\.ru/media/(?P<id>.+)' _VALID_URL = r'https?://(?:www\.)?newstube\.ru/media/(?P<id>.+)'
_TEST = { _TEST = {
'url': 'http://www.newstube.ru/media/telekanal-cnn-peremestil-gorod-slavyansk-v-krym', 'url': 'http://www.newstube.ru/media/telekanal-cnn-peremestil-gorod-slavyansk-v-krym',
'md5': '801eef0c2a9f4089fa04e4fe3533abdc',
'info_dict': { 'info_dict': {
'id': '728e0ef2-e187-4012-bac0-5a081fdcb1f6', 'id': '728e0ef2-e187-4012-bac0-5a081fdcb1f6',
'ext': 'flv', 'ext': 'mp4',
'title': 'Телеканал CNN переместил город Славянск в Крым', 'title': 'Телеканал CNN переместил город Славянск в Крым',
'description': 'md5:419a8c9f03442bc0b0a794d689360335', 'description': 'md5:419a8c9f03442bc0b0a794d689360335',
'duration': 31.05, 'duration': 31.05,
}, },
'params': {
# rtmp download
'skip_download': True,
},
} }
def _real_extract(self, url): def _real_extract(self, url):
@ -62,7 +62,6 @@ class NewstubeIE(InfoExtractor):
server = media_location.find(ns('./Server')).text server = media_location.find(ns('./Server')).text
app = media_location.find(ns('./App')).text app = media_location.find(ns('./App')).text
media_id = stream_info.find(ns('./Id')).text media_id = stream_info.find(ns('./Id')).text
quality_id = stream_info.find(ns('./QualityId')).text
name = stream_info.find(ns('./Name')).text name = stream_info.find(ns('./Name')).text
width = int(stream_info.find(ns('./Width')).text) width = int(stream_info.find(ns('./Width')).text)
height = int(stream_info.find(ns('./Height')).text) height = int(stream_info.find(ns('./Height')).text)
@ -74,12 +73,38 @@ class NewstubeIE(InfoExtractor):
'rtmp_conn': ['S:%s' % session_id, 'S:%s' % media_id, 'S:n2'], 'rtmp_conn': ['S:%s' % session_id, 'S:%s' % media_id, 'S:n2'],
'page_url': url, 'page_url': url,
'ext': 'flv', 'ext': 'flv',
'format_id': quality_id, 'format_id': 'rtmp' + ('-%s' % name if name else ''),
'format_note': name,
'width': width, 'width': width,
'height': height, 'height': height,
}) })
sources_data = self._download_json(
'http://www.newstube.ru/player2/getsources?guid=%s' % video_guid,
video_guid, fatal=False)
if sources_data:
for source in sources_data.get('Sources', []):
source_url = source.get('Src')
if not source_url:
continue
height = int_or_none(source.get('Height'))
f = {
'format_id': 'http' + ('-%dp' % height if height else ''),
'url': source_url,
'width': int_or_none(source.get('Width')),
'height': height,
}
source_type = source.get('Type')
if source_type:
mobj = re.search(r'codecs="([^,]+),\s*([^"]+)"', source_type)
if mobj:
vcodec, acodec = mobj.groups()
f.update({
'vcodec': vcodec,
'acodec': acodec,
})
formats.append(f)
self._check_formats(formats, video_guid)
self._sort_formats(formats) self._sort_formats(formats)
return { return {