diff --git a/YoutubeDL.py b/YoutubeDL.py
index 1789cb46304f..3ab59ea31e65 100644
--- a/YoutubeDL.py
+++ b/YoutubeDL.py
@@ -2036,7 +2036,7 @@ def sanitize_numeric_fields(info):
         elif thumbnails:
             info_dict['thumbnail'] = thumbnails[-1]['url']
 
-        if 'display_id' not in info_dict and 'id' in info_dict:
+        if info_dict.get('display_id') is None and 'id' in info_dict:
             info_dict['display_id'] = info_dict['id']
 
         for ts_key, date_key in (
@@ -2052,6 +2052,23 @@ def sanitize_numeric_fields(info):
                 except (ValueError, OverflowError, OSError):
                     pass
 
+        live_keys = ('is_live', 'was_live')
+        live_status = info_dict.get('live_status')
+        if live_status is None:
+            for key in live_keys:
+                if info_dict.get(key) is False:
+                    continue
+                if info_dict.get(key):
+                    live_status = key
+                break
+            if all(info_dict.get(key) is False for key in live_keys):
+                live_status = 'not_live'
+        if live_status:
+            info_dict['live_status'] = live_status
+            for key in live_keys:
+                if info_dict.get(key) is None:
+                    info_dict[key] = (live_status == key)
+
         # Auto generate title fields corresponding to the *_number fields when missing
         # in order to always have clean titles. This is very common for TV series.
         for field in ('chapter', 'season', 'episode'):
diff --git a/extractor/common.py b/extractor/common.py
index a6fc5d11a3e1..e892ff3cbcc0 100644
--- a/extractor/common.py
+++ b/extractor/common.py
@@ -297,6 +297,8 @@ class InfoExtractor(object):
                     live stream that goes on instead of a fixed-length video.
     was_live:       True, False, or None (=unknown). Whether this video was
                     originally a live stream.
+    live_status:    'is_live', 'upcoming', 'was_live', 'not_live' or None (=unknown)
+                    If absent, automatically set from is_live, was_live
     start_time:     Time in seconds where the reproduction should start, as
                     specified in the URL.
     end_time:       Time in seconds where the reproduction should end, as