Fixed public list streaming

rainbowstream/draw.py

@@ -210,7 +210,7 @@ def draw(t, keyword=None, humanize=True, noti=False, fil=[], ig=[]):
             t['event'] = 'retweet'
             c['events'].append(t)
             notify_retweet(t)
-            return
+            return True
     except:
         pass
 
@@ -250,9 +250,9 @@ def draw(t, keyword=None, humanize=True, noti=False, fil=[], ig=[]):
     fil = list(set((fil or []) + c['ONLY_LIST']))
     ig = list(set((ig or []) + c['IGNORE_LIST']))
     if fil and screen_name not in fil:
-        return
+        return False
     if ig and screen_name in ig:
-        return
+        return False
 
     # Get rainbow id
     if tid not in c['tweet_dict']:
@@ -375,6 +375,7 @@ def draw(t, keyword=None, humanize=True, noti=False, fil=[], ig=[]):
                 image_to_display(BytesIO(response.content))
             except Exception:
                 printNicely(red('Sorry, image link is broken'))
+    return True
 
 
 def print_threads(d):

rainbowstream/rainbow.py

@@ -2010,20 +2010,22 @@ def stream(domain, args, name='Rainbow Stream'):
     # The Logo
     art_dict = {
         c['USER_DOMAIN']: name,
-        c['PUBLIC_DOMAIN']: args.track_keywords or 'Global',
+        c['PUBLIC_DOMAIN']: args.track_keywords or name or 'Global',
         c['SITE_DOMAIN']: name,
     }
     if c['ASCII_ART']:
         ascii_art(art_dict.get(domain, name))
     # These arguments are optional:
     stream_args = dict(
-        timeout=0.5,  # To check g['stream_stop'] after each 0.5 s
+        timeout=0.25,  # To check g['stream_stop'] after each 0.5 s
         block=True,
         heartbeat_timeout=c['HEARTBEAT_TIMEOUT'] * 60)
     # Track keyword
     query_args = dict()
     if args.track_keywords:
         query_args['track'] = args.track_keywords
+    if args.filter_ids:
+        query_args['follow'] = args.filter_ids
     # Get stream
     stream = TwitterStream(
         auth=authen(),
@@ -2035,7 +2037,7 @@ def stream(domain, args, name='Rainbow Stream'):
         elif domain == c['SITE_DOMAIN']:
             tweet_iter = stream.site(**query_args)
         else:
-            if args.track_keywords:
+            if args.track_keywords or args.filter_ids:
                 tweet_iter = stream.statuses.filter(**query_args)
             else:
                 tweet_iter = stream.statuses.sample()
@@ -2071,22 +2073,22 @@ def stream(domain, args, name='Rainbow Stream'):
                 if g['pause']:
                     continue
                 while c['lock']:
-                    time.sleep(0.5)
+                    time.sleep(0.25)
                 # Draw the tweet
-                draw(
+                did_draw = draw(
                     t=tweet,
                     keyword=args.track_keywords,
                     humanize=False,
                     fil=args.filter,
                     ig=args.ignore,
                 )
                 # Current readline buffer
-                current_buffer = readline.get_line_buffer().strip()
+                current_buffer = readline.get_line_buffer()
                 # There is an unexpected behaviour in MacOSX readline + Python 2:
                 # after completely delete a word after typing it,
                 # somehow readline buffer still contains
                 # the 1st character of that word
-                if current_buffer and g['cmd'] != current_buffer:
+                if current_buffer and g['cmd'] != current_buffer and did_draw:
                     sys.stdout.write(
                         g['decorated_name'](g['PREFIX']) + current_buffer)
                     sys.stdout.flush()
@@ -2098,7 +2100,7 @@ def stream(domain, args, name='Rainbow Stream'):
                 if g['pause']:
                     continue
                 while c['lock']:
-                    time.sleep(0.5)
+                    time.sleep(0.25)
                 print_message(tweet['direct_message'])
             elif tweet.get('event'):
                 c['events'].append(tweet)
@@ -2122,6 +2124,7 @@ def spawn_public_stream(args, keyword=None):
     Spawn a new public stream
     """
     # Only set keyword if specified
+    args.filter_ids = ''
     if keyword:
         if keyword[0] == '#':
             keyword = keyword[1:]
@@ -2131,6 +2134,7 @@ def spawn_public_stream(args, keyword=None):
         g['keyword'] = 'Global'
     g['PREFIX'] = u2str(emojize(format_prefix(keyword=g['keyword'])))
     g['listname'] = ''
+
     # Start new thread
     th = threading.Thread(
         target=stream,
@@ -2162,7 +2166,7 @@ def spawn_list_stream(args, stuff=None):
     printNicely(light_yellow('getting list members ...'))
     # Get members
     t = Twitter(auth=authen())
-    members = []
+    members, member_ids = [], []
     next_cursor = -1
     while next_cursor != 0:
         m = t.lists.members(
@@ -2172,15 +2176,18 @@ def spawn_list_stream(args, stuff=None):
             include_entities=False)
         for u in m['users']:
             members.append('@' + u['screen_name'])
+            member_ids.append(u['id'])
         next_cursor = m['next_cursor']
     printNicely(light_yellow('... done.'))
     # Build thread filter array
     args.filter = members
+    args.filter_ids = ','.join([str(x) for x in member_ids])
+
     # Start new thread
     th = threading.Thread(
         target=stream,
         args=(
-            c['USER_DOMAIN'],
+            c['PUBLIC_DOMAIN'],
             args,
             slug))
     th.daemon = True
@@ -2197,6 +2204,7 @@ def spawn_personal_stream(args, stuff=None):
     """
     Spawn a new personal stream
     """
+    args.filter_ids = ''
     # Reset the tracked keyword and listname
     g['keyword'] = g['listname'] = ''
     # Reset prefix
@@ -2254,7 +2262,7 @@ def fly():
         spawn_dict.get(target)(args, stuff)
 
     # Start listen process
-    time.sleep(0.5)
+    time.sleep(0.25)
     g['reset'] = True
     g['prefix'] = True
     listen()