Merge tag 'upstream/2017.05.18.1'

[youtubedl] / youtube_dl / extractor / niconico.py

diff --git a/youtube_dl/extractor/niconico.py b/youtube_dl/extractor/niconico.py
index 517a72561bbaf444c54daabde4bae61f341086b3..8baac23e4b16643a59e6a83570862b6a5afd2b45 100644 (file)
--- a/youtube_dl/extractor/niconico.py
+++ b/youtube_dl/extractor/niconico.py
@@ -1,17 +1,23 @@
-# encoding: utf-8
+# coding: utf-8

 from __future__ import unicode_literals
 
 import re
 from __future__ import unicode_literals
 
 import re

+import json
+import datetime

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

-from ..utils import (
-    compat_urllib_parse,
-    compat_urllib_request,
+from ..compat import (

     compat_urlparse,
     compat_urlparse,

-    compat_str,
-
+)
+from ..utils import (

     ExtractorError,
     ExtractorError,

-    unified_strdate,
+    int_or_none,
+    parse_duration,
+    parse_iso8601,
+    sanitized_Request,
+    xpath_text,
+    determine_ext,
+    urlencode_postdata,

 )
 
 
 )
 
 


@@ -19,7 +25,7 @@ class NiconicoIE(InfoExtractor):
     IE_NAME = 'niconico'
     IE_DESC = 'ニコニコ動画'
 
     IE_NAME = 'niconico'
     IE_DESC = 'ニコニコ動画'
 

-    _TEST = {
+    _TESTS = [{

         'url': 'http://www.nicovideo.jp/watch/sm22312215',
         'md5': 'd1a75c0823e2f629128c43e1212760f9',
         'info_dict': {
         'url': 'http://www.nicovideo.jp/watch/sm22312215',
         'md5': 'd1a75c0823e2f629128c43e1212760f9',
         'info_dict': {


@@ -29,15 +35,57 @@ class NiconicoIE(InfoExtractor):
             'uploader': 'takuya0301',
             'uploader_id': '2698420',
             'upload_date': '20131123',
             'uploader': 'takuya0301',
             'uploader_id': '2698420',
             'upload_date': '20131123',

+            'timestamp': 1385182762,

             'description': '(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org',
             'description': '(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org',

+            'duration': 33,

         },
         },

-        'params': {
-            'username': 'ydl.niconico@gmail.com',
-            'password': 'youtube-dl',
+        'skip': 'Requires an account',
+    }, {
+        # File downloaded with and without credentials are different, so omit
+        # the md5 field
+        'url': 'http://www.nicovideo.jp/watch/nm14296458',
+        'info_dict': {
+            'id': 'nm14296458',
+            'ext': 'swf',
+            'title': '【鏡音リン】Dance on media【オリジナル】take2!',
+            'description': 'md5:689f066d74610b3b22e0f1739add0f58',
+            'uploader': 'りょうた',
+            'uploader_id': '18822557',
+            'upload_date': '20110429',
+            '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',
+        '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.',
+    }]

 
 

-    _VALID_URL = r'^https?://(?:www\.|secure\.)?nicovideo\.jp/watch/([a-z][a-z][0-9]+)(?:.*)$'
+    _VALID_URL = r'https?://(?:www\.|secure\.)?nicovideo\.jp/watch/(?P<id>(?:[a-z]{2})?[0-9]+)'

     _NETRC_MACHINE = 'niconico'
 
     def _real_initialize(self):
     _NETRC_MACHINE = 'niconico'
 
     def _real_initialize(self):


@@ -45,20 +93,17 @@ class NiconicoIE(InfoExtractor):
 
     def _login(self):
         (username, password) = self._get_login_info()
 
     def _login(self):
         (username, password) = self._get_login_info()

-        if username is None:
-            # Login is required
-            raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
+        # No authentication to be performed
+        if not username:
+            return True

 
         # Log in
         login_form_strs = {
             'mail': username,
             'password': password,
         }
 
         # Log in
         login_form_strs = {
             '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')
             'https://secure.nicovideo.jp/secure/login', login_data)
         login_results = self._download_webpage(
             request, None, note='Logging in', errnote='Unable to log in')


@@ -68,12 +113,15 @@ class NiconicoIE(InfoExtractor):
         return True
 
     def _real_extract(self, url):
         return True
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1)
+        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,
 
         video_info = self._download_xml(
             'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id, video_id,


@@ -81,42 +129,140 @@ class NiconicoIE(InfoExtractor):
 
         # Get flv info
         flv_info_webpage = self._download_webpage(
 
         # Get flv info
         flv_info_webpage = self._download_webpage(

-            'http://flapi.nicovideo.jp/api/getflv?v=' + video_id,
+            'http://flapi.nicovideo.jp/api/getflv/' + video_id + '?as3=1',

             video_id, 'Downloading flv info')
             video_id, 'Downloading flv info')

-        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)
+            elif 'closed' in flv_info:
+                raise ExtractorError('Niconico videos now require logging in',
+                                     expected=True)
+            else:
+                raise ExtractorError('Unable to find video URL')
+
+        video_real_url = flv_info['url'][0]

 
         # Start extracting information
 
         # Start extracting information

-        video_title = video_info.find('.//title').text
-        video_extension = video_info.find('.//movie_type').text
-        video_format = video_extension.upper()
-        video_thumbnail = video_info.find('.//thumbnail_url').text
-        video_description = video_info.find('.//description').text
-        video_uploader_id = video_info.find('.//user_id').text
-        video_upload_date = unified_strdate(video_info.find('.//first_retrieve').text.split('+')[0])
-        video_view_count = video_info.find('.//view_counter').text
-        video_webpage_url = video_info.find('.//watch_url').text
-
-        # uploader
-        video_uploader = video_uploader_id
-        url = 'http://seiga.nicovideo.jp/api/user/info?id=' + video_uploader_id
-        try:
-            user_info = self._download_xml(
-                url, video_id, note='Downloading user information')
-            video_uploader = user_info.find('.//nickname').text
-        except ExtractorError as err:
-            self._downloader.report_warning('Unable to download user info webpage: %s' % compat_str(err))
+        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
+            uploader = video_info.find('.//ch_name').text
+        elif video_info.find('.//user_id') is not None:
+            uploader_id = video_info.find('.//user_id').text
+            uploader = video_info.find('.//user_nickname').text
+        else:
+            uploader_id = uploader = None

 
         return {
             'id': video_id,
             'url': video_real_url,
 
         return {
             'id': video_id,
             'url': video_real_url,

-            'title': video_title,
-            'ext': video_extension,
-            'format': video_format,
-            'thumbnail': video_thumbnail,
-            'description': video_description,
-            'uploader': video_uploader,
-            'upload_date': video_upload_date,
-            'uploader_id': video_uploader_id,
-            'view_count': video_view_count,
-            'webpage_url': video_webpage_url,
+            'title': title,
+            'ext': extension,
+            'format_id': 'economy' if video_real_url.endswith('low') else 'normal',
+            'thumbnail': thumbnail,
+            'description': description,
+            'uploader': uploader,
+            'timestamp': timestamp,
+            'uploader_id': uploader_id,
+            'view_count': view_count,
+            'comment_count': comment_count,
+            'duration': duration,
+            'webpage_url': webpage_url,
+        }
+
+
+class NiconicoPlaylistIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/mylist/(?P<id>\d+)'
+
+    _TEST = {
+        'url': 'http://www.nicovideo.jp/mylist/27411728',
+        'info_dict': {
+            'id': '27411728',
+            'title': 'AKB48のオールナイトニッポン',
+        },
+        'playlist_mincount': 225,
+    }
+
+    def _real_extract(self, url):
+        list_id = self._match_id(url)
+        webpage = self._download_webpage(url, list_id)
+
+        entries_json = self._search_regex(r'Mylist\.preload\(\d+, (\[.*\])\);',
+                                          webpage, 'entries')
+        entries = json.loads(entries_json)
+        entries = [{
+            '_type': 'url',
+            'ie_key': NiconicoIE.ie_key(),
+            'url': ('http://www.nicovideo.jp/watch/%s' %
+                    entry['item_data']['video_id']),
+        } for entry in entries]
+
+        return {
+            '_type': 'playlist',
+            'title': self._search_regex(r'\s+name: "(.*?)"', webpage, 'title'),
+            'id': list_id,
+            'entries': entries,

         }
         }