X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fniconico.py;h=586e52a4a4f49151c55ed99baf9cadd136863bbd;hb=2e7e561c1d9dedf1a8e5a206e1ef86cfa4599956;hp=0ca046ac2ee1f8b5d8836362e404c465d9967122;hpb=bb865f3a5e7290f0da43d180f068f8715c39e3ae;p=youtube-dl diff --git a/youtube_dl/extractor/niconico.py b/youtube_dl/extractor/niconico.py index 0ca046ac2..586e52a4a 100644 --- a/youtube_dl/extractor/niconico.py +++ b/youtube_dl/extractor/niconico.py @@ -3,18 +3,20 @@ 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_urlparse, ) from ..utils import ( + encode_dict, ExtractorError, int_or_none, parse_duration, parse_iso8601, + sanitized_Request, xpath_text, determine_ext, ) @@ -38,13 +40,10 @@ class NiconicoIE(InfoExtractor): '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', @@ -56,23 +55,30 @@ class NiconicoIE(InfoExtractor): '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', - 'md5': '38e53c9aad548f3ecf01ca7680b59b08', '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, }, - 'params': { - 'username': 'ydl.niconico@gmail.com', - 'password': 'youtube-dl', + }, { + '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', } }] @@ -95,11 +101,8 @@ class NiconicoIE(InfoExtractor): '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 = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('utf-8') + 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') @@ -116,7 +119,10 @@ class NiconicoIE(InfoExtractor): # 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 = self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id) + 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, @@ -139,7 +145,7 @@ class NiconicoIE(InfoExtractor): '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( @@ -158,22 +164,65 @@ class NiconicoIE(InfoExtractor): # Start extracting information 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']+class="videoHeaderTitle"[^>]*>([^<]+)', webpage, 'video title') + watch_api_data_string = self._html_search_regex( + r']+id="watchAPIDataContainer"[^>]+>([^<]+)', + 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) - video_format = extension.upper() - thumbnail = xpath_text(video_info, './/thumbnail_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: ]*>([^<]+)', + 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')) - duration = parse_duration(xpath_text(video_info, './/length')) - webpage_url = xpath_text(video_info, './/watch_url') + if not comment_count: + match = self._html_search_regex( + r'>Comments: ]*>([^<]+)', + 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 @@ -184,12 +233,12 @@ class NiconicoIE(InfoExtractor): else: uploader_id = uploader = None - ret = { + return { 'id': video_id, '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, @@ -200,7 +249,6 @@ class NiconicoIE(InfoExtractor): 'duration': duration, 'webpage_url': webpage_url, } - return dict((k, v) for k, v in ret.items() if v is not None) class NiconicoPlaylistIE(InfoExtractor):