[pandoratv] Improve extraction (Closes #7921)

This commit is contained in:
Sergey M․ 2016-01-01 00:40:27 +06:00
parent 9accfed4e7
commit e4bd63f9c0

View file

@ -2,28 +2,36 @@
from __future__ import unicode_literals from __future__ import unicode_literals
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import ( from ..compat import (
compat_str,
compat_urlparse, compat_urlparse,
) )
from ..utils import ( from ..utils import (
ExtractorError, ExtractorError,
float_or_none,
parse_duration,
str_to_int,
) )
class PandoraTVIE(InfoExtractor): class PandoraTVIE(InfoExtractor):
_VALID_URL = r'http://(?:.+?\.)?channel.pandora.tv/channel/video.ptv\?' _VALID_URL = r'https?://(?:.+?\.)?channel\.pandora\.tv/channel/video\.ptv\?'
_TESTS = [{ _TEST = {
'url': 'http://jp.channel.pandora.tv/channel/video.ptv?c1=&prgid=53294230&ch_userid=mikakim&ref=main&lot=cate_01_2', 'url': 'http://jp.channel.pandora.tv/channel/video.ptv?c1=&prgid=53294230&ch_userid=mikakim&ref=main&lot=cate_01_2',
'info_dict': { 'info_dict': {
'description': '\u982d\u3092\u64ab\u3067\u3066\u304f\u308c\u308b\uff1f',
'ext': 'mp4',
'id': '53294230', 'id': '53294230',
'title': '\u982d\u3092\u64ab\u3067\u3066\u304f\u308c\u308b\uff1f', 'ext': 'flv',
'title': '頭を撫でてくれる?',
'description': '頭を撫でてくれる?',
'thumbnail': 're:^https?://.*\.jpg$',
'duration': 39,
'upload_date': '20151218', 'upload_date': '20151218',
'uploader': 'カワイイ動物まとめ',
'uploader_id': 'mikakim',
'view_count': int,
'like_count': int,
}
} }
}]
def _real_extract(self, url): def _real_extract(self, url):
qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
@ -32,25 +40,37 @@ class PandoraTVIE(InfoExtractor):
if any(not f for f in (video_id, user_id,)): if any(not f for f in (video_id, user_id,)):
raise ExtractorError('Invalid URL', expected=True) raise ExtractorError('Invalid URL', expected=True)
data_url ='http://m.pandora.tv/?c=view&m=viewJsonApi&ch_userid={userid}&prgid={prgid}'.format(userid=user_id,prgid=video_id) data = self._download_json(
data = self._download_json(data_url, video_id) 'http://m.pandora.tv/?c=view&m=viewJsonApi&ch_userid=%s&prgid=%s'
% (user_id, video_id), video_id)
info = data['data']['rows']['vod_play_info']['result'] info = data['data']['rows']['vod_play_info']['result']
formats = [] formats = []
for format_id in sorted([k for k in info if k.startswith('v') and k.endswith('Url') and info[k]]): for format_id, format_url in info.items():
if not format_url:
continue
height = self._search_regex(
r'^v(\d+)[Uu]rl$', format_id, 'height', default=None)
if not height:
continue
formats.append({ formats.append({
'format_id': format_id, 'format_id': '%sp' % height,
'url': info[format_id], 'url': format_url,
'ext': 'mp4', 'height': int(height),
'height': int(format_id[1:-3]),
}) })
self._sort_formats(formats)
return { return {
'description': info['body'],
'thumbnail': info['thumbnail'],
'formats': formats,
'id': video_id, 'id': video_id,
'title': info['subject'], 'title': info['subject'],
'upload_date': info['fid'][:8], 'description': info.get('body'),
'view_count': info['hit'], 'thumbnail': info.get('thumbnail') or info.get('poster'),
'duration': float_or_none(info.get('runtime'), 1000) or parse_duration(info.get('time')),
'upload_date': info['fid'][:8] if isinstance(info.get('fid'), compat_str) else None,
'uploader': info.get('nickname'),
'uploader_id': info.get('upload_userid'),
'view_count': str_to_int(info.get('hit')),
'like_count': str_to_int(info.get('likecnt')),
'formats': formats,
} }