diff --git a/py/app_se_cocoa.py b/py/app_se_cocoa.py
index 618ed8b9..a07c7613 100644
--- a/py/app_se_cocoa.py
+++ b/py/app_se_cocoa.py
@@ -8,6 +8,7 @@
 from hsfs.phys import Directory as DirectoryBase
 from hsfs.phys.bundle import Bundle
 from hsutil.path import Path
+from hsutil.misc import extract
 from hsutil.str import get_file_ext
 
 
@@ -15,13 +16,19 @@ from . import app_cocoa, data
 from .directories import Directories as DirectoriesBase, STATE_EXCLUDED
 
 class DGDirectory(DirectoryBase):
-    def _create_sub_dir(self, name, with_parent = True):
+    def _create_sub_file(self, name, with_parent=True):
         ext = get_file_ext(name)
         if ext == 'app':
             parent = self if with_parent else None
             return Bundle(parent, name)
         else:
-            return super(DGDirectory, self)._create_sub_dir(name, with_parent)
+            return super(DGDirectory, self)._create_sub_file(name, with_parent)
+    
+    def _fetch_subitems(self):
+        subdirs, subfiles = super(DGDirectory, self)._fetch_subitems()
+        apps, normal_dirs = extract(lambda name: get_file_ext(name) == 'app', subdirs)
+        subfiles += apps
+        return normal_dirs, subfiles
     
 
 class Directories(DirectoriesBase):
diff --git a/py/scanner.py b/py/scanner.py
index 9a7521c7..69a03b38 100644
--- a/py/scanner.py
+++ b/py/scanner.py
@@ -54,7 +54,11 @@ class Scanner(object):
             SCAN_TYPE_CONTENT_AUDIO: lambda f: [str(f.audiosize)]
         }[self.scan_type]
         for f in j.iter_with_progress(files, 'Read metadata of %d/%d files'):
+            if self.size_threshold:
+                f.size # pre-read, makes a smoother progress if read here (especially for bundles)
             f.words = func(f)
+        if self.size_threshold:
+            files = [f for f in files if f.size >= self.size_threshold]
         return mf.getmatches(files, j)
     
     @staticmethod
@@ -75,8 +79,6 @@ class Scanner(object):
         j = j.start_subjob([8, 2])
         for f in [f for f in files if not hasattr(f, 'is_ref')]:
             f.is_ref = False
-        if self.size_threshold:
-            files = [f for f in files if f.size >= self.size_threshold]
         logging.info('Getting matches')
         if self.match_factory is None:
             matches = self._getmatches(files, j)