Update upstream source from tag 'upstream/2018.09.10'

[youtubedl] / youtube_dl / extractor / imgur.py

diff --git a/youtube_dl/extractor/imgur.py b/youtube_dl/extractor/imgur.py
index 70c8ca64e6346d86ef5ecea02722597829db1316..ecc958a1717d35ad37cc68c3b44147f3d0db6161 100644 (file)
--- a/youtube_dl/extractor/imgur.py
+++ b/youtube_dl/extractor/imgur.py
@@ -3,7 +3,6 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

-from ..compat import compat_urlparse

 from ..utils import (
     int_or_none,
     js_to_json,
 from ..utils import (
     int_or_none,
     js_to_json,


@@ -13,7 +12,7 @@ from ..utils import (
 
 
 class ImgurIE(InfoExtractor):
 
 
 class ImgurIE(InfoExtractor):

-    _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?!gallery)(?P<id>[a-zA-Z0-9]+)'
+    _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?:(?:gallery|(?:topic|r)/[^/]+)/)?(?P<id>[a-zA-Z0-9]{6,})(?:[/?#&]+|\.[a-z0-9]+)?$'

 
     _TESTS = [{
         'url': 'https://i.imgur.com/A61SaA1.gifv',
 
     _TESTS = [{
         'url': 'https://i.imgur.com/A61SaA1.gifv',


@@ -21,7 +20,7 @@ class ImgurIE(InfoExtractor):
             'id': 'A61SaA1',
             'ext': 'mp4',
             'title': 're:Imgur GIF$|MRW gifv is up and running without any bugs$',
             'id': 'A61SaA1',
             'ext': 'mp4',
             'title': 're:Imgur GIF$|MRW gifv is up and running without any bugs$',

-            'description': 're:The origin of the Internet\'s most viral images$|The Internet\'s visual storytelling community\. Explore, share, and discuss the best visual stories the Internet has to offer\.$',
+            'description': 'Imgur: The magic of the Internet',

         },
     }, {
         'url': 'https://imgur.com/A61SaA1',
         },
     }, {
         'url': 'https://imgur.com/A61SaA1',


@@ -29,21 +28,35 @@ class ImgurIE(InfoExtractor):
             'id': 'A61SaA1',
             'ext': 'mp4',
             'title': 're:Imgur GIF$|MRW gifv is up and running without any bugs$',
             'id': 'A61SaA1',
             'ext': 'mp4',
             'title': 're:Imgur GIF$|MRW gifv is up and running without any bugs$',

-            'description': 're:The origin of the Internet\'s most viral images$|The Internet\'s visual storytelling community\. Explore, share, and discuss the best visual stories the Internet has to offer\.$',
+            'description': 'Imgur: The magic of 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....',
+        }
+    }, {
+        'url': 'http://imgur.com/topic/Funny/N8rOudd',
+        'only_matching': True,
+    }, {
+        'url': 'http://imgur.com/r/aww/VQcQPhM',
+        'only_matching': True,
+    }, {
+        'url': 'https://i.imgur.com/crGpqCV.mp4',
+        'only_matching': True,

     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

-        webpage = self._download_webpage(
-            compat_urlparse.urljoin(url, video_id), video_id)
+        gifv_url = 'https://i.imgur.com/{id}.gifv'.format(id=video_id)
+        webpage = self._download_webpage(gifv_url, video_id)

 
 

-        width = int_or_none(self._search_regex(
-            r'<param name="width" value="([0-9]+)"',
-            webpage, 'width', fatal=False))
-        height = int_or_none(self._search_regex(
-            r'<param name="height" value="([0-9]+)"',
-            webpage, 'height', fatal=False))
+        width = int_or_none(self._og_search_property(
+            'video:width', webpage, default=None))
+        height = int_or_none(self._og_search_property(
+            'video:height', webpage, default=None))

 
         video_elements = self._search_regex(
             r'(?s)<div class="video-elements">(.*?)</div>',
 
         video_elements = self._search_regex(
             r'(?s)<div class="video-elements">(.*?)</div>',


@@ -94,31 +107,44 @@ class ImgurIE(InfoExtractor):
         return {
             'id': video_id,
             'formats': formats,
         return {
             'id': video_id,
             'formats': formats,

-            'description': self._og_search_description(webpage),
+            'description': self._og_search_description(webpage, default=None),

             'title': self._og_search_title(webpage),
         }
 
 
 class ImgurAlbumIE(InfoExtractor):
             'title': self._og_search_title(webpage),
         }
 
 
 class ImgurAlbumIE(InfoExtractor):

-    _VALID_URL = r'https?://(?:i\.)?imgur\.com/gallery/(?P<id>[a-zA-Z0-9]+)'
+    _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?:(?:a|gallery|topic/[^/]+)/)?(?P<id>[a-zA-Z0-9]{5})(?:[/?#&]+)?$'

 
 

-    _TEST = {
+    _TESTS = [{

         'url': 'http://imgur.com/gallery/Q95ko',
         'info_dict': {
             'id': 'Q95ko',
         },
         'playlist_count': 25,
         '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_images = self._download_json(
             'http://imgur.com/gallery/%s/album_images/hit.json?all=true' % album_id,
 
     def _real_extract(self, url):
         album_id = self._match_id(url)
 
         album_images = self._download_json(
             'http://imgur.com/gallery/%s/album_images/hit.json?all=true' % album_id,

-            album_id)['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_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())