debian/changelog: Annotate bugs to close.

[youtubedl] / youtube_dl / extractor / niconico.py

diff --git a/youtube_dl/extractor/niconico.py b/youtube_dl/extractor/niconico.py
index c0c139b5df16ce900ba6920a1a004bc433eab4e9..3cecebf95a4acd0a388da5984aebc2822e79b32c 100644 (file)
--- a/youtube_dl/extractor/niconico.py
+++ b/youtube_dl/extractor/niconico.py
@@ -2,15 +2,22 @@
 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 (
+from ..compat import (

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

-    unified_strdate,
-    parse_duration,
+)
+from ..utils import (
+    ExtractorError,

     int_or_none,
     int_or_none,

+    parse_duration,
+    parse_iso8601,
+    xpath_text,
+    determine_ext,

 )
 
 
 )
 
 


@@ -18,7 +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': {


@@ -28,29 +35,65 @@ 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',
             'duration': 33,
         },
             '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',
+        '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,

         },
         },

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

     _NETRC_MACHINE = 'niconico'
     _NETRC_MACHINE = 'niconico'

-    # Determine whether the downloader uses authentication to download video
-    _AUTHENTICATE = False
+    # Determine whether the downloader used authentication to download video
+    _AUTHENTICATED = False

 
     def _real_initialize(self):
 
     def _real_initialize(self):

-        if self._downloader.params.get('username', None) is not None:
-            self._AUTHENTICATE = True
-
-        if self._AUTHENTICATE:
-            self._login()
+        self._login()

 
     def _login(self):
         (username, password) = self._get_login_info()
 
     def _login(self):
         (username, password) = self._get_login_info()

+        # No authentication to be performed
+        if not username:
+            return True

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


@@ -68,24 +111,29 @@ class NiconicoIE(InfoExtractor):
         if re.search(r'(?i)<h1 class="mb8p4">Log in error</h1>', login_results) is not None:
             self._downloader.report_warning('unable to log in: bad username or password')
             return False
         if re.search(r'(?i)<h1 class="mb8p4">Log in error</h1>', login_results) is not None:
             self._downloader.report_warning('unable to log in: bad username or password')
             return False

+        # Successful login
+        self._AUTHENTICATED = True

         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,
             note='Downloading video info page')
 
 
         video_info = self._download_xml(
             'http://ext.nicovideo.jp/api/getthumbinfo/' + video_id, video_id,
             note='Downloading video info page')
 

-        if self._AUTHENTICATE:
+        if self._AUTHENTICATED:

             # 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')
         else:
             # Get external player info
                 video_id, 'Downloading flv info')
         else:
             # Get external player info


@@ -106,19 +154,78 @@ class NiconicoIE(InfoExtractor):
                 flv_info_request, video_id,
                 note='Downloading flv info', errnote='Unable to download flv info')
 
                 flv_info_request, video_id,
                 note='Downloading flv info', errnote='Unable to download 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)
+            else:
+                raise ExtractorError('Unable to find video URL')
+
+        video_real_url = flv_info['url'][0]

 
         # Start extracting information
 
         # Start extracting information

-        title = video_info.find('.//title').text
-        extension = video_info.find('.//movie_type').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)

         video_format = extension.upper()
         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
+
+        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
 
         if video_info.find('.//ch_id') is not None:
             uploader_id = video_info.find('.//ch_id').text


@@ -138,10 +245,44 @@ class NiconicoIE(InfoExtractor):
             'thumbnail': thumbnail,
             'description': description,
             'uploader': uploader,
             'thumbnail': thumbnail,
             'description': description,
             'uploader': uploader,

-            'upload_date': upload_date,
+            'timestamp': timestamp,

             'uploader_id': uploader_id,
             'view_count': view_count,
             'comment_count': comment_count,
             'duration': duration,
             'webpage_url': webpage_url,
         }
             '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,
+        }