Merge tag 'upstream/2014.10.30'

[youtubedl] / youtube_dl / extractor / niconico.py

diff --git a/youtube_dl/extractor/niconico.py b/youtube_dl/extractor/niconico.py
index 46774317c9f6b47a337b20b70100a4d56e7fe022..62d5707fee87302b5f69544f768b1a56a5e3976b 100644 (file)
--- a/youtube_dl/extractor/niconico.py
+++ b/youtube_dl/extractor/niconico.py
@@ -1,47 +1,47 @@
 # encoding: utf-8
 # encoding: utf-8

+from __future__ import unicode_literals

 
 import re
 
 import re

-import socket
+import json

 
 from .common import InfoExtractor
 from ..utils import (
 
 from .common import InfoExtractor
 from ..utils import (

-    compat_http_client,
-    compat_urllib_error,

     compat_urllib_parse,
     compat_urllib_request,
     compat_urlparse,
     compat_urllib_parse,
     compat_urllib_request,
     compat_urlparse,

-    compat_str,
-
-    ExtractorError,

     unified_strdate,
     unified_strdate,

+    parse_duration,
+    int_or_none,

 )
 
 
 class NiconicoIE(InfoExtractor):
 )
 
 
 class NiconicoIE(InfoExtractor):

-    IE_NAME = u'niconico'
-    IE_DESC = u'ニコニコ動画'
+    IE_NAME = 'niconico'
+    IE_DESC = 'ニコニコ動画'

 
     _TEST = {
 
     _TEST = {

-        u'url': u'http://www.nicovideo.jp/watch/sm22312215',
-        u'file': u'sm22312215.mp4',
-        u'md5': u'd1a75c0823e2f629128c43e1212760f9',
-        u'info_dict': {
-            u'title': u'Big Buck Bunny',
-            u'uploader': u'takuya0301',
-            u'uploader_id': u'2698420',
-            u'upload_date': u'20131123',
-            u'description': u'(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org',
+        'url': 'http://www.nicovideo.jp/watch/sm22312215',
+        'md5': 'd1a75c0823e2f629128c43e1212760f9',
+        'info_dict': {
+            'id': 'sm22312215',
+            'ext': 'mp4',
+            'title': 'Big Buck Bunny',
+            'uploader': 'takuya0301',
+            'uploader_id': '2698420',
+            'upload_date': '20131123',
+            'description': '(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org',
+            'duration': 33,

         },
         },

-        u'params': {
-            u'username': u'ydl.niconico@gmail.com',
-            u'password': u'youtube-dl',
+        'params': {
+            'username': 'ydl.niconico@gmail.com',
+            'password': 'youtube-dl',

         },
     }
 
         },
     }
 

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

     _NETRC_MACHINE = 'niconico'
     _NETRC_MACHINE = 'niconico'

-    # If True it will raise an error if no login info is provided
-    _LOGIN_REQUIRED = True
+    # Determine whether the downloader used authentication to download video
+    _AUTHENTICATED = False

 
     def _real_initialize(self):
         self._login()
 
     def _real_initialize(self):
         self._login()


@@ -49,27 +49,27 @@ class NiconicoIE(InfoExtractor):
     def _login(self):
         (username, password) = self._get_login_info()
         # No authentication to be performed
     def _login(self):
         (username, password) = self._get_login_info()
         # No authentication to be performed

-        if username is None:
-            if self._LOGIN_REQUIRED:
-                raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True)
-            return False
+        if not username:
+            return True

 
         # Log in
         login_form_strs = {
 
         # Log in
         login_form_strs = {

-            u'mail': username,
-            u'password': password,
+            'mail': username,
+            'password': password,

         }
         # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
         # chokes on unicode
         }
         # 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_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(login_form).encode('utf-8')
         request = compat_urllib_request.Request(

-            u'https://secure.nicovideo.jp/secure/login', login_data)
+            'https://secure.nicovideo.jp/secure/login', login_data)

         login_results = self._download_webpage(
         login_results = self._download_webpage(

-            request, u'', note=u'Logging in', errnote=u'Unable to log in')
+            request, None, note='Logging in', errnote='Unable to log in')

         if re.search(r'(?i)<h1 class="mb8p4">Log in error</h1>', login_results) is not None:
         if re.search(r'(?i)<h1 class="mb8p4">Log in error</h1>', login_results) is not None:

-            self._downloader.report_warning(u'unable to log in: bad username or password')
+            self._downloader.report_warning('unable to log in: bad username or password')

             return False
             return False

+        # Successful login
+        self._AUTHENTICATED = True

         return True
 
     def _real_extract(self, url):
         return True
 
     def _real_extract(self, url):


@@ -82,46 +82,101 @@ class NiconicoIE(InfoExtractor):
 
         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,

-            note=u'Downloading video info page')
+            note='Downloading video info page')
+
+        if self._AUTHENTICATED:
+            # Get flv info
+            flv_info_webpage = self._download_webpage(
+                'http://flapi.nicovideo.jp/api/getflv?v=' + video_id,
+                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(
-            u'http://flapi.nicovideo.jp/api/getflv?v=' + video_id,
-            video_id, u'Downloading flv info')

         video_real_url = compat_urlparse.parse_qs(flv_info_webpage)['url'][0]
 
         # Start extracting information
         video_real_url = compat_urlparse.parse_qs(flv_info_webpage)['url'][0]
 
         # 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=u'Downloading user information')
-            video_uploader = user_info.find('.//nickname').text
-        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            self._downloader.report_warning(u'Unable to download user info webpage: %s' % compat_str(err))
+        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
+
+        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,
+            'title': title,
+            'ext': extension,
+            'format': video_format,
+            'thumbnail': thumbnail,
+            'description': description,
+            'uploader': uploader,
+            'upload_date': upload_date,
+            '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_id'],
+        } for entry in entries]

 
         return {
 
         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,
+            '_type': 'playlist',
+            'title': self._search_regex(r'\s+name: "(.*?)"', webpage, 'title'),
+            'id': list_id,
+            'entries': entries,

         }
         }