[meipai] Fix regular videos extraction and improve (closes #10718)

This commit is contained in:
Sergey M․ 2016-12-17 19:42:34 +07:00
parent 9b785768ac
commit 2786818c33
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D
1 changed files with 79 additions and 74 deletions

View File

@ -1,99 +1,104 @@
# coding: utf-8 # coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
from ..utils import parse_iso8601
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import (
int_or_none,
parse_duration,
unified_timestamp,
)
class MeipaiIE(InfoExtractor): class MeipaiIE(InfoExtractor):
IE_DESC = '美拍' IE_DESC = '美拍'
_VALID_URL = r'https?://(?:www\.)?meipai.com/media/(?P<id>[0-9]+)' _VALID_URL = r'https?://(?:www\.)?meipai.com/media/(?P<id>[0-9]+)'
_TESTS = [ _TESTS = [{
{ # regular uploaded video
'url': 'http://www.meipai.com/media/531697625', 'url': 'http://www.meipai.com/media/531697625',
'md5': 'e3e9600f9e55a302daecc90825854b4f', 'md5': 'e3e9600f9e55a302daecc90825854b4f',
'info_dict': { 'info_dict': {
'id': '531697625', 'id': '531697625',
'ext': 'mp4', 'ext': 'mp4',
'title': '#葉子##阿桑##余姿昀##超級女聲#', 'title': '#葉子##阿桑##余姿昀##超級女聲#',
'description': '#葉子##阿桑##余姿昀##超級女聲#', 'description': '#葉子##阿桑##余姿昀##超級女聲#',
'thumbnail': 're:^https?://.*\.jpg$', 'thumbnail': 're:^https?://.*\.jpg$',
'creator': '她她-TATA', 'duration': 152,
'tags': ['葉子', '阿桑', '余姿昀', '超級女聲'], 'timestamp': 1465492420,
'release_date': 1465492420, 'upload_date': '20160609',
} 'view_count': 35511,
}, 'creator': '她她-TATA',
{ 'tags': ['葉子', '阿桑', '余姿昀', '超級女聲'],
'url': 'http://www.meipai.com/media/576409659', }
'md5': '2e807c16ebe67b8b6b3c8dcacbc32f48', }, {
'info_dict': {
'id': '576409659',
'ext': 'mp4',
'title': '#失語者##蔡健雅##吉他彈唱#',
'description': '#失語者##蔡健雅##吉他彈唱#',
'thumbnail': 're:^https?://.*\.jpg$',
'creator': '她她-TATA',
'tags': ['失語者', '蔡健雅', '吉他彈唱'],
'release_date': 1472534847,
}
},
# record of live streaming # record of live streaming
{ 'url': 'http://www.meipai.com/media/585526361',
'url': 'http://www.meipai.com/media/585526361', 'md5': 'ff7d6afdbc6143342408223d4f5fb99a',
'md5': 'ff7d6afdbc6143342408223d4f5fb99a', 'info_dict': {
'info_dict': { 'id': '585526361',
'id': '585526361', 'ext': 'mp4',
'ext': 'mp4', 'title': '姿昀和善願 練歌練琴啦😁😁😁',
'title': '姿昀和善願 練歌練琴啦😁😁😁', 'description': '姿昀和善願 練歌練琴啦😁😁😁',
'description': '姿昀和善願 練歌練琴啦😁😁😁', 'thumbnail': 're:^https?://.*\.jpg$',
'thumbnail': 're:^https?://.*\.jpg$', 'duration': 5975,
'creator': '她她-TATA', 'timestamp': 1474311799,
'release_date': 1474311799, 'upload_date': '20160919',
} 'view_count': 1215,
}, 'creator': '她她-TATA',
] }
}]
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
title = self._og_search_title(webpage, default=None) title = self._og_search_title(
if title is None: webpage, default=None) or self._html_search_regex(
# fall back to text used in title r'<title[^>]*>([^<]+)</title>', webpage, 'title')
title = self._html_search_regex(
r'<title[^>]*>(.+)</title>', webpage, 'title')
release_date = self._og_search_property( formats = []
'video:release_date', webpage, 'release date', fatal=False)
release_date = parse_iso8601(release_date) # recorded playback of live streaming
m3u8_url = self._html_search_regex(
r'file:\s*encodeURIComponent\((["\'])(?P<url>(?:(?!\1).)+)\1\)',
webpage, 'm3u8 url', group='url', default=None)
if m3u8_url:
formats.extend(self._extract_m3u8_formats(
m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id='hls', fatal=False))
if not formats:
# regular uploaded video
video_url = self._search_regex(
r'data-video=(["\'])(?P<url>(?:(?!\1).)+)\1', webpage, 'video url',
group='url', default=None)
if video_url:
formats.append({
'url': video_url,
'format_id': 'http',
})
timestamp = unified_timestamp(self._og_search_property(
'video:release_date', webpage, 'release date', fatal=False))
tags = self._og_search_property( tags = self._og_search_property(
'video:tag', webpage, 'tags', default='').split(',') 'video:tag', webpage, 'tags', default='').split(',')
info = { view_count = int_or_none(self._html_search_meta(
'interactionCount', webpage, 'view count'))
duration = parse_duration(self._html_search_meta(
'duration', webpage, 'duration'))
creator = self._og_search_property(
'video:director', webpage, 'creator', fatal=False)
return {
'id': video_id, 'id': video_id,
'title': title, 'title': title,
'thumbnail': self._og_search_thumbnail(webpage),
'description': self._og_search_description(webpage), 'description': self._og_search_description(webpage),
'release_date': release_date, 'thumbnail': self._og_search_thumbnail(webpage),
'creator': self._og_search_property( 'duration': duration,
'video:director', webpage, 'creator', fatal=False), 'timestamp': timestamp,
'view_count': view_count,
'creator': creator,
'tags': tags, 'tags': tags,
'formats': formats,
} }
keywords = self._html_search_meta(
'keywords', webpage, 'keywords', default=[])
if '直播回放' in keywords:
# recorded playback of live streaming
m3u8_url = self._html_search_regex(
r'file:\s*encodeURIComponent\(["\'](.+)["\']\)',
webpage,
'm3u8_url')
info['formats'] = self._extract_m3u8_formats(
m3u8_url, video_id, 'mp4', 'm3u8_native')
else:
# regular uploaded video
info['url'] = self._og_search_video_url(webpage)
return info