X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fniconico.py;h=79b9952c36c212b4321b01408672675e191d99a2;hb=daaaf5f5942252e9fbc367957bd3b8a96d0dd5bb;hp=a9b770cb321ef3ab2f3cc3096698ded1e1801922;hpb=59d814f79341341e6390392a09e628ee12a6f18d;p=youtube-dl diff --git a/youtube_dl/extractor/niconico.py b/youtube_dl/extractor/niconico.py index a9b770cb3..79b9952c3 100644 --- a/youtube_dl/extractor/niconico.py +++ b/youtube_dl/extractor/niconico.py @@ -1,22 +1,22 @@ -# 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_parse_qs, compat_urlparse, ) from ..utils import ( + determine_ext, ExtractorError, int_or_none, parse_duration, parse_iso8601, + urlencode_postdata, xpath_text, - determine_ext, ) @@ -38,6 +38,7 @@ class NiconicoIE(InfoExtractor): 'description': '(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org', 'duration': 33, }, + 'skip': 'Requires an account', }, { # File downloaded with and without credentials are different, so omit # the md5 field @@ -53,22 +54,41 @@ class NiconicoIE(InfoExtractor): 'timestamp': 1304065916, 'duration': 209, }, + 'skip': 'Requires an account', }, { # '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': int, # timestamp field has different value if logged in + 'duration': 304, + }, + 'skip': 'Requires an account', + }, { + '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', }, + 'skip': 'The viewing period of the video you were searching for has expired.', + }, { + 'url': 'http://sp.nicovideo.jp/watch/sm28964488?ss_pos=1&cp_in=wt_tg', + 'only_matching': True, }] - _VALID_URL = r'https?://(?:www\.|secure\.)?nicovideo\.jp/watch/(?P(?:[a-z]{2})?[0-9]+)' + _VALID_URL = r'https?://(?:www\.|secure\.|sp\.)?nicovideo\.jp/watch/(?P(?:[a-z]{2})?[0-9]+)' _NETRC_MACHINE = 'niconico' - # Determine whether the downloader used authentication to download video - _AUTHENTICATED = False def _real_initialize(self): self._login() @@ -80,24 +100,24 @@ class NiconicoIE(InfoExtractor): return True # Log in + login_ok = True login_form_strs = { - 'mail': username, + 'mail_tel': 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( - 'https://secure.nicovideo.jp/secure/login', login_data) - login_results = self._download_webpage( - request, None, note='Logging in', errnote='Unable to log in') - if re.search(r'(?i)

Log in error

', login_results) is not None: + urlh = self._request_webpage( + 'https://account.nicovideo.jp/api/v1/login', None, + note='Logging in', errnote='Unable to log in', + data=urlencode_postdata(login_form_strs)) + if urlh is False: + login_ok = False + else: + parts = compat_urlparse.urlparse(urlh.geturl()) + if compat_parse_qs(parts.query).get('message', [None])[0] == 'cant_login': + login_ok = False + if not login_ok: self._downloader.report_warning('unable to log in: bad username or password') - return False - # Successful login - self._AUTHENTICATED = True - return True + return login_ok def _real_extract(self, url): video_id = self._match_id(url) @@ -105,41 +125,31 @@ 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, note='Downloading video info page') - if self._AUTHENTICATED: - # Get flv info - flv_info_webpage = self._download_webpage( - 'http://flapi.nicovideo.jp/api/getflv/' + video_id + '?as3=1', - video_id, 'Downloading flv info') - else: - # Get external player info - ext_player_info = self._download_webpage( - 'http://ext.nicovideo.jp/thumb_watch/' + video_id, video_id) - thumb_play_key = self._search_regex( - r'\'thumbPlayKey\'\s*:\s*\'(.*?)\'', ext_player_info, 'thumbPlayKey') - - # Get flv info - flv_info_data = compat_urllib_parse.urlencode({ - 'k': thumb_play_key, - 'v': video_id - }) - flv_info_request = compat_urllib_request.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') + # Get flv info + flv_info_webpage = self._download_webpage( + 'http://flapi.nicovideo.jp/api/getflv/' + video_id + '?as3=1', + video_id, 'Downloading flv info') 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) + elif 'closed' in flv_info: + raise ExtractorError('Niconico videos now require logging in', + expected=True) + elif 'error' in flv_info: + raise ExtractorError('%s reports error: %s' % ( + self.IE_NAME, flv_info['error'][0]), expected=True) else: raise ExtractorError('Unable to find video URL') @@ -154,17 +164,58 @@ class NiconicoIE(InfoExtractor): 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 @@ -175,12 +226,12 @@ 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, @@ -191,11 +242,10 @@ 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): - _VALID_URL = r'https?://www\.nicovideo\.jp/mylist/(?P\d+)' + _VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/mylist/(?P\d+)' _TEST = { 'url': 'http://www.nicovideo.jp/mylist/27411728',