Improve extraction (Closes #7918)

[youtube-dl] / youtube_dl / extractor / imgur.py

diff --git a/youtube_dl/extractor/imgur.py b/youtube_dl/extractor/imgur.py
index 38c9617731cf1b1a708939e46009b0bdba88774d..88423f1798a9cdd830962e2e1b976d1f48776bfc 100644 (file)
--- a/youtube_dl/extractor/imgur.py
+++ b/youtube_dl/extractor/imgur.py
@@ -3,6 +3,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_urlparse
 from ..utils import (
     int_or_none,
     js_to_json,
@@ -10,35 +11,41 @@ from ..utils import (
     ExtractorError,
 )
 
+
 class ImgurIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?P<id>[a-zA-Z0-9]+)(?:\.mp4|\.gifv)?'
+    _VALID_URL = r'https?://(?:i\.)?imgur\.com/(gallery/)?(?P<id>[a-zA-Z0-9]{6,})'
 
     _TESTS = [{
         'url': 'https://i.imgur.com/A61SaA1.gifv',
         'info_dict': {
             'id': 'A61SaA1',
             'ext': 'mp4',
-            'title': 'MRW gifv is up and running without any bugs',
-            'description': 'The Internet\'s visual storytelling community. Explore, share, and discuss the best visual stories the Internet has to offer.',
+            'title': 're:Imgur GIF$|MRW gifv is up and running without any bugs$',
+            'description': 'Imgur: The most awesome images on the Internet.',
         },
     }, {
         'url': 'https://imgur.com/A61SaA1',
         'info_dict': {
             'id': 'A61SaA1',
             'ext': 'mp4',
-            'title': 'MRW gifv is up and running without any bugs',
-            'description': 'The Internet\'s visual storytelling community. Explore, share, and discuss the best visual stories the Internet has to offer.',
+            'title': 're:Imgur GIF$|MRW gifv is up and running without any bugs$',
+            'description': 'Imgur: The most awesome images on the Internet.',
         },
+    }, {
+        'url': 'https://imgur.com/gallery/YcAQlkx',
+        'info_dict': {
+            'id': 'YcAQlkx',
+            'ext': 'mp4',
+            'title': 'Classic Steve Carell gif...cracks me up everytime....damn the repost downvotes....',
+            'description': 'Imgur: The most awesome images on the Internet.'
+
+        }
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-
-        sources = re.findall(r'<source src="([^"]+)" type="([^"]+)"', webpage)
-        if not sources:
-            raise ExtractorError(
-                'No sources found for video %s' % video_id, expected=True)
+        webpage = self._download_webpage(
+            compat_urlparse.urljoin(url, video_id), video_id)
 
         width = int_or_none(self._search_regex(
             r'<param name="width" value="([0-9]+)"',
@@ -47,10 +54,14 @@ class ImgurIE(InfoExtractor):
             r'<param name="height" value="([0-9]+)"',
             webpage, 'height', fatal=False))
 
-        formats = []
         video_elements = self._search_regex(
             r'(?s)<div class="video-elements">(.*?)</div>',
-            webpage, 'video elements')
+            webpage, 'video elements', default=None)
+        if not video_elements:
+            raise ExtractorError(
+                'No sources found for video %s. Maybe an image?' % video_id,
+                expected=True)
+
         formats = []
         for m in re.finditer(r'<source\s+src="(?P<src>[^"]+)"\s+type="(?P<type>[^"]+)"', video_elements):
             formats.append({
@@ -95,3 +106,31 @@ class ImgurIE(InfoExtractor):
             'description': self._og_search_description(webpage),
             'title': self._og_search_title(webpage),
         }
+
+
+class ImgurAlbumIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:i\.)?imgur\.com/(gallery/)?(?P<id>[a-zA-Z0-9]{5})(?![a-zA-Z0-9])'
+
+    _TEST = {
+        'url': 'http://imgur.com/gallery/Q95ko',
+        'info_dict': {
+            'id': 'Q95ko',
+        },
+        'playlist_count': 25,
+    }
+
+    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)