From 774ce35571c08a1532fe4079224239adfdb80e43 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Tue, 22 Dec 2015 21:48:48 +0600 Subject: [PATCH] [imgur] Improve (Closes #7928) --- youtube_dl/extractor/imgur.py | 45 ++++++++++++++++++++++------------- 1 file changed, 29 insertions(+), 16 deletions(-) diff --git a/youtube_dl/extractor/imgur.py b/youtube_dl/extractor/imgur.py index 88423f179..85e9344aa 100644 --- a/youtube_dl/extractor/imgur.py +++ b/youtube_dl/extractor/imgur.py @@ -13,7 +13,7 @@ from ..utils import ( class ImgurIE(InfoExtractor): - _VALID_URL = r'https?://(?:i\.)?imgur\.com/(gallery/)?(?P[a-zA-Z0-9]{6,})' + _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?:(?:gallery|topic/[^/]+)/)?(?P[a-zA-Z0-9]{6,})(?:[/?#&]+|\.[a-z]+)?$' _TESTS = [{ 'url': 'https://i.imgur.com/A61SaA1.gifv', @@ -40,6 +40,9 @@ class ImgurIE(InfoExtractor): 'description': 'Imgur: The most awesome images on the Internet.' } + }, { + 'url': 'http://imgur.com/topic/Funny/N8rOudd', + 'only_matching': True, }] def _real_extract(self, url): @@ -109,28 +112,38 @@ class ImgurIE(InfoExtractor): class ImgurAlbumIE(InfoExtractor): - _VALID_URL = r'https?://(?:i\.)?imgur\.com/(gallery/)?(?P[a-zA-Z0-9]{5})(?![a-zA-Z0-9])' + _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?:(?:a|gallery|topic/[^/]+)/)?(?P[a-zA-Z0-9]{5})(?:[/?#&]+)?$' - _TEST = { + _TESTS = [{ 'url': 'http://imgur.com/gallery/Q95ko', 'info_dict': { 'id': 'Q95ko', }, 'playlist_count': 25, - } + }, { + 'url': 'http://imgur.com/a/j6Orj', + 'only_matching': True, + }, { + 'url': 'http://imgur.com/topic/Aww/ll5Vk', + 'only_matching': True, + }] def _real_extract(self, url): album_id = self._match_id(url) - album_img_data = self._download_json( - 'http://imgur.com/gallery/%s/album_images/hit.json?all=true' % album_id, album_id)['data'] - - if len(album_img_data) == 0: - return self.url_result('http://imgur.com/%s' % album_id) - else: - album_images = album_img_data['images'] - entries = [ - self.url_result('http://imgur.com/%s' % image['hash']) - for image in album_images if image.get('hash')] - - return self.playlist_result(entries, album_id) + album_images = self._download_json( + 'http://imgur.com/gallery/%s/album_images/hit.json?all=true' % album_id, + album_id, fatal=False) + + if album_images: + data = album_images.get('data') + if data and isinstance(data, dict): + images = data.get('images') + if images and isinstance(images, list): + entries = [ + self.url_result('http://imgur.com/%s' % image['hash']) + for image in images if image.get('hash')] + return self.playlist_result(entries, album_id) + + # Fallback to single video + return self.url_result('http://imgur.com/%s' % album_id, ImgurIE.ie_key())