-# encoding: utf-8
+# coding: utf-8
from __future__ import unicode_literals
import re
import json
+import datetime
from .common import InfoExtractor
from ..compat import (
- compat_urllib_parse,
- compat_urllib_request,
+ compat_urllib_parse_urlencode,
compat_urlparse,
)
from ..utils import (
ExtractorError,
int_or_none,
parse_duration,
- unified_strdate,
+ parse_iso8601,
+ sanitized_Request,
+ xpath_text,
+ determine_ext,
+ urlencode_postdata,
)
'uploader': 'takuya0301',
'uploader_id': '2698420',
'upload_date': '20131123',
+ 'timestamp': 1385182762,
'description': '(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org',
'duration': 33,
},
- 'params': {
- 'username': 'ydl.niconico@gmail.com',
- 'password': 'youtube-dl',
- },
}, {
+ # File downloaded with and without credentials are different, so omit
+ # the md5 field
'url': 'http://www.nicovideo.jp/watch/nm14296458',
- 'md5': '8db08e0158457cf852a31519fceea5bc',
'info_dict': {
'id': 'nm14296458',
'ext': 'swf',
'title': '【鏡音リン】Dance on media【オリジナル】take2!',
- 'description': 'md5:',
+ 'description': 'md5:689f066d74610b3b22e0f1739add0f58',
'uploader': 'りょうた',
'uploader_id': '18822557',
'upload_date': '20110429',
+ 'timestamp': 1304065916,
'duration': 209,
},
- 'params': {
- 'username': 'ydl.niconico@gmail.com',
- 'password': 'youtube-dl',
+ }, {
+ # 'video exists but is marked as "deleted"
+ # md5 is unstable
+ 'url': 'http://www.nicovideo.jp/watch/sm10000',
+ 'info_dict': {
+ 'id': 'sm10000',
+ 'ext': 'unknown_video',
+ 'description': 'deleted',
+ 'title': 'ドラえもんエターナル第3話「決戦第3新東京市」＜前編＞',
+ 'upload_date': '20071224',
+ 'timestamp': 1198527840, # timestamp field has different value if logged in
+ 'duration': 304,
},
+ }, {
+ 'url': 'http://www.nicovideo.jp/watch/so22543406',
+ 'info_dict': {
+ 'id': '1388129933',
+ 'ext': 'mp4',
+ 'title': '【第1回】RADIOアニメロミックス ラブライブ！～のぞえりRadio Garden～',
+ 'description': 'md5:b27d224bb0ff53d3c8269e9f8b561cf1',
+ 'timestamp': 1388851200,
+ 'upload_date': '20140104',
+ 'uploader': 'アニメロチャンネル',
+ 'uploader_id': '312',
+ }
}]
_VALID_URL = r'https?://(?:www\.|secure\.)?nicovideo\.jp/watch/(?P<id>(?:[a-z]{2})?[0-9]+)'
'mail': username,
'password': password,
}
- # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
- # chokes on unicode
- login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
- login_data = compat_urllib_parse.urlencode(login_form).encode('utf-8')
- request = compat_urllib_request.Request(
+ login_data = urlencode_postdata(login_form_strs)
+ request = sanitized_Request(
'https://secure.nicovideo.jp/secure/login', login_data)
login_results = self._download_webpage(
request, None, note='Logging in', errnote='Unable to log in')
def _real_extract(self, url):
video_id = self._match_id(url)
- # Get video webpage. We are not actually interested in it, but need
- # the cookies in order to be able to download the info webpage
- self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id)
+ # Get video webpage. We are not actually interested in it for normal
+ # cases, but need the cookies in order to be able to download the
+ # info webpage
+ webpage, handle = self._download_webpage_handle(
+ 'http://www.nicovideo.jp/watch/' + video_id, video_id)
+ if video_id.startswith('so'):
+ video_id = self._match_id(handle.geturl())
video_info = self._download_xml(
'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id, video_id,
r'\'thumbPlayKey\'\s*:\s*\'(.*?)\'', ext_player_info, 'thumbPlayKey')
# Get flv info
- flv_info_data = compat_urllib_parse.urlencode({
+ flv_info_data = compat_urllib_parse_urlencode({
'k': thumb_play_key,
'v': video_id
})
- flv_info_request = compat_urllib_request.Request(
+ flv_info_request = sanitized_Request(
'http://ext.nicovideo.jp/thumb_watch', flv_info_data,
{'Content-Type': 'application/x-www-form-urlencoded'})
flv_info_webpage = self._download_webpage(
flv_info_request, video_id,
note='Downloading flv info', errnote='Unable to download flv info')
- if 'deleted=' in flv_info_webpage:
- raise ExtractorError('The video has been deleted.',
- expected=True)
- video_real_url = compat_urlparse.parse_qs(flv_info_webpage)['url'][0]
+ flv_info = compat_urlparse.parse_qs(flv_info_webpage)
+ if 'url' not in flv_info:
+ if 'deleted' in flv_info:
+ raise ExtractorError('The video has been deleted.',
+ expected=True)
+ else:
+ raise ExtractorError('Unable to find video URL')
+
+ video_real_url = flv_info['url'][0]
# Start extracting information
- title = video_info.find('.//title').text
- extension = video_info.find('.//movie_type').text
- video_format = extension.upper()
- thumbnail = video_info.find('.//thumbnail_url').text
- description = video_info.find('.//description').text
- upload_date = unified_strdate(video_info.find('.//first_retrieve').text.split('+')[0])
- view_count = int_or_none(video_info.find('.//view_counter').text)
- comment_count = int_or_none(video_info.find('.//comment_num').text)
- duration = parse_duration(video_info.find('.//length').text)
- webpage_url = video_info.find('.//watch_url').text
+ title = xpath_text(video_info, './/title')
+ if not title:
+ title = self._og_search_title(webpage, default=None)
+ if not title:
+ title = self._html_search_regex(
+ r'<span[^>]+class="videoHeaderTitle"[^>]*>([^<]+)</span>',
+ webpage, 'video title')
+
+ watch_api_data_string = self._html_search_regex(
+ r'<div[^>]+id="watchAPIDataContainer"[^>]+>([^<]+)</div>',
+ webpage, 'watch api data', default=None)
+ watch_api_data = self._parse_json(watch_api_data_string, video_id) if watch_api_data_string else {}
+ video_detail = watch_api_data.get('videoDetail', {})
+
+ extension = xpath_text(video_info, './/movie_type')
+ if not extension:
+ extension = determine_ext(video_real_url)
+
+ thumbnail = (
+ xpath_text(video_info, './/thumbnail_url') or
+ self._html_search_meta('image', webpage, 'thumbnail', default=None) or
+ video_detail.get('thumbnail'))
+
+ description = xpath_text(video_info, './/description')
+
+ timestamp = parse_iso8601(xpath_text(video_info, './/first_retrieve'))
+ if not timestamp:
+ match = self._html_search_meta('datePublished', webpage, 'date published', default=None)
+ if match:
+ timestamp = parse_iso8601(match.replace('+', ':00+'))
+ if not timestamp and video_detail.get('postedAt'):
+ timestamp = parse_iso8601(
+ video_detail['postedAt'].replace('/', '-'),
+ delimiter=' ', timezone=datetime.timedelta(hours=9))
+
+ view_count = int_or_none(xpath_text(video_info, './/view_counter'))
+ if not view_count:
+ match = self._html_search_regex(
+ r'>Views: <strong[^>]*>([^<]+)</strong>',
+ webpage, 'view count', default=None)
+ if match:
+ view_count = int_or_none(match.replace(',', ''))
+ view_count = view_count or video_detail.get('viewCount')
+
+ comment_count = int_or_none(xpath_text(video_info, './/comment_num'))
+ if not comment_count:
+ match = self._html_search_regex(
+ r'>Comments: <strong[^>]*>([^<]+)</strong>',
+ webpage, 'comment count', default=None)
+ if match:
+ comment_count = int_or_none(match.replace(',', ''))
+ comment_count = comment_count or video_detail.get('commentCount')
+
+ duration = (parse_duration(
+ xpath_text(video_info, './/length') or
+ self._html_search_meta(
+ 'video:duration', webpage, 'video duration', default=None)) or
+ video_detail.get('length'))
+
+ webpage_url = xpath_text(video_info, './/watch_url') or url
if video_info.find('.//ch_id') is not None:
uploader_id = video_info.find('.//ch_id').text
'url': video_real_url,
'title': title,
'ext': extension,
- 'format': video_format,
+ 'format_id': 'economy' if video_real_url.endswith('low') else 'normal',
'thumbnail': thumbnail,
'description': description,
'uploader': uploader,
- 'upload_date': upload_date,
+ 'timestamp': timestamp,
'uploader_id': uploader_id,
'view_count': view_count,
'comment_count': comment_count,
class NiconicoPlaylistIE(InfoExtractor):
- _VALID_URL = r'https?://www\.nicovideo\.jp/mylist/(?P<id>\d+)'
+ _VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/mylist/(?P<id>\d+)'
_TEST = {
'url': 'http://www.nicovideo.jp/mylist/27411728',
projects / youtube-dl / blobdiff