From 64ed7a38f98e9b01feae757bec25b81db80b29f6 Mon Sep 17 00:00:00 2001 From: hakatashi Date: Wed, 2 Jul 2014 03:13:12 +0900 Subject: [PATCH 1/2] [niconico] Add support for channel video --- youtube_dl/extractor/niconico.py | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-) diff --git a/youtube_dl/extractor/niconico.py b/youtube_dl/extractor/niconico.py index 517a72561b..43d8644a4b 100644 --- a/youtube_dl/extractor/niconico.py +++ b/youtube_dl/extractor/niconico.py @@ -37,7 +37,7 @@ class NiconicoIE(InfoExtractor): }, } - _VALID_URL = r'^https?://(?:www\.|secure\.)?nicovideo\.jp/watch/([a-z][a-z][0-9]+)(?:.*)$' + _VALID_URL = r'^https?://(?:www\.|secure\.)?nicovideo\.jp/watch/((?:[a-z][a-z])?[0-9]+)(?:.*)$' _NETRC_MACHINE = 'niconico' def _real_initialize(self): @@ -91,20 +91,18 @@ class NiconicoIE(InfoExtractor): 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)) + # No need to fetch extra resources...new API has field for uploader's name + if video_info.find('.//ch_id') is not None: + video_uploader_id = video_info.find('.//ch_id').text + video_uploader = video_info.find('.//ch_name').text + elif video_info.find('.//user_id') is not None: + video_uploader_id = video_info.find('.//user_id').text + video_uploader = video_info.find('.//user_nickname').text return { 'id': video_id, From 15ce1338b42f906a5b9f812b8f8b5287eab8a20a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Fri, 4 Jul 2014 22:05:46 +0700 Subject: [PATCH 2/2] [niconico] Extract more metadata and simplify (Closes #3181) --- youtube_dl/extractor/niconico.py | 57 +++++++++++++++++--------------- 1 file changed, 31 insertions(+), 26 deletions(-) diff --git a/youtube_dl/extractor/niconico.py b/youtube_dl/extractor/niconico.py index 43d8644a4b..31f60041cf 100644 --- a/youtube_dl/extractor/niconico.py +++ b/youtube_dl/extractor/niconico.py @@ -8,10 +8,10 @@ from ..utils import ( compat_urllib_parse, compat_urllib_request, compat_urlparse, - compat_str, - ExtractorError, unified_strdate, + parse_duration, + int_or_none, ) @@ -30,6 +30,7 @@ class NiconicoIE(InfoExtractor): 'uploader_id': '2698420', 'upload_date': '20131123', 'description': '(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org', + 'duration': 33, }, 'params': { 'username': 'ydl.niconico@gmail.com', @@ -37,7 +38,7 @@ class NiconicoIE(InfoExtractor): }, } - _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' def _real_initialize(self): @@ -86,35 +87,39 @@ class NiconicoIE(InfoExtractor): 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_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 + 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 - # uploader - # No need to fetch extra resources...new API has field for uploader's name if video_info.find('.//ch_id') is not None: - video_uploader_id = video_info.find('.//ch_id').text - video_uploader = video_info.find('.//ch_name').text + uploader_id = video_info.find('.//ch_id').text + uploader = video_info.find('.//ch_name').text elif video_info.find('.//user_id') is not None: - video_uploader_id = video_info.find('.//user_id').text - video_uploader = video_info.find('.//user_nickname').text + 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': video_title, - 'ext': video_extension, + 'title': title, + 'ext': 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, + '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, }