[extractor/common] Filter out unsupported encrypted media for f4m formats (Closes...

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

diff --git a/youtube_dl/extractor/instagram.py b/youtube_dl/extractor/instagram.py
index ddcaf76d11ec5c5b7e2ef05fb76afd1b804391df..4e62098b05fa1ec753f240adc5a27b7da57e2e95 100644 (file)
--- a/youtube_dl/extractor/instagram.py
+++ b/youtube_dl/extractor/instagram.py
@@ -4,14 +4,16 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
+    get_element_by_attribute,
     int_or_none,
     limit_length,
+    lowercase_escape,
 )
 
 
 class InstagramIE(InfoExtractor):
-    _VALID_URL = r'https://instagram\.com/p/(?P<id>[^/?#&]+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?instagram\.com/p/(?P<id>[^/?#&]+)'
+    _TESTS = [{
         'url': 'https://instagram.com/p/aye83DjauH/?foo=bar#abc',
         'md5': '0d2da106a9d2631273e192b372806516',
         'info_dict': {
@@ -21,7 +23,34 @@ class InstagramIE(InfoExtractor):
             'title': 'Video by naomipq',
             'description': 'md5:1f17f0ab29bd6fe2bfad705f58de3cb8',
         }
-    }
+    }, {
+        # missing description
+        'url': 'https://www.instagram.com/p/BA-pQFBG8HZ/?taken-by=britneyspears',
+        'info_dict': {
+            'id': 'BA-pQFBG8HZ',
+            'ext': 'mp4',
+            'uploader_id': 'britneyspears',
+            'title': 'Video by britneyspears',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://instagram.com/p/-Cmh1cukG2/',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_embed_url(webpage):
+        blockquote_el = get_element_by_attribute(
+            'class', 'instagram-media', webpage)
+        if blockquote_el is None:
+            return
+
+        mobj = re.search(
+            r'<a[^>]+href=([\'"])(?P<link>[^\'"]+)\1', blockquote_el)
+        if mobj:
+            return mobj.group('link')
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -29,8 +58,10 @@ class InstagramIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
         uploader_id = self._search_regex(r'"owner":{"username":"(.+?)"',
                                          webpage, 'uploader id', fatal=False)
-        desc = self._search_regex(r'"caption":"(.*?)"', webpage, 'description',
-                                  fatal=False)
+        desc = self._search_regex(
+            r'"caption":"(.+?)"', webpage, 'description', default=None)
+        if desc is not None:
+            desc = lowercase_escape(desc)
 
         return {
             'id': video_id,
@@ -44,7 +75,7 @@ class InstagramIE(InfoExtractor):
 
 
 class InstagramUserIE(InfoExtractor):
-    _VALID_URL = r'https://instagram\.com/(?P<username>[^/]{2,})/?(?:$|[?#])'
+    _VALID_URL = r'https?://(?:www\.)?instagram\.com/(?P<username>[^/]{2,})/?(?:$|[?#])'
     IE_DESC = 'Instagram user profile'
     IE_NAME = 'instagram:user'
     _TEST = {