From 84769e708c81a1600ff24e7140c889fefc4aa6ef Mon Sep 17 00:00:00 2001 From: Philipp Hagemeister Date: Thu, 13 Mar 2014 16:40:53 +0100 Subject: [PATCH] [ninegag] Fix extraction --- test/test_download.py | 4 ++++ youtube_dl/extractor/ninegag.py | 32 ++++++++++++++++++++------------ 2 files changed, 24 insertions(+), 12 deletions(-) diff --git a/test/test_download.py b/test/test_download.py index bbbb6b78a..c241d9324 100644 --- a/test/test_download.py +++ b/test/test_download.py @@ -144,6 +144,10 @@ def generator(test_case): self.assertTrue( isinstance(got, compat_str) and match_rex.match(got), u'field %s (value: %r) should match %r' % (info_field, got, match_str)) + elif isinstance(expected, type): + got = info_dict.get(info_field) + self.assertTrue(isinstance(got, expected), + u'Expected type %r, but got value %r of type %r' % (expected, got, type(got))) else: if isinstance(expected, compat_str) and expected.startswith('md5:'): got = 'md5:' + md5(info_dict.get(info_field)) diff --git a/youtube_dl/extractor/ninegag.py b/youtube_dl/extractor/ninegag.py index 2b7236be5..1d7aa40ed 100644 --- a/youtube_dl/extractor/ninegag.py +++ b/youtube_dl/extractor/ninegag.py @@ -15,7 +15,9 @@ class NineGagIE(InfoExtractor): "file": "1912.mp4", "info_dict": { "description": "This 3-minute video will make you smile and then make you feel untalented and insignificant. Anyway, you should share this awesomeness. (Thanks, Dino!)", - "title": "\"People Are Awesome 2013\" Is Absolutely Awesome" + "title": "\"People Are Awesome 2013\" Is Absolutely Awesome", + "view_count": int, + "thumbnail": "re:^https?://", }, 'add_ie': ['Youtube'] } @@ -25,21 +27,27 @@ class NineGagIE(InfoExtractor): video_id = mobj.group('id') webpage = self._download_webpage(url, video_id) - data_json = self._html_search_regex(r'''(?x) - .*?

(.*?)

', webpage, + 'description', fatal=False) + view_count_str = self._html_search_regex( + r'

([0-9][0-9,]*) views

', webpage, 'view count', + fatal=False) + view_count = ( + None if view_count_str is None + else int(view_count_str.replace(',', ''))) return { '_type': 'url_transparent', - 'url': data['youtubeVideoId'], + 'url': youtube_id, 'ie_key': 'Youtube', 'id': video_id, - 'title': data['title'], - 'description': data['description'], - 'view_count': int(data['view_count']), - 'like_count': int(data['statistic']['like']), - 'dislike_count': int(data['statistic']['dislike']), - 'thumbnail': data['thumbnail_url'], + 'title': self._og_search_title(webpage), + 'description': description, + 'view_count': view_count, + 'thumbnail': self._og_search_thumbnail(webpage), }