Improved performance of directory skipping

generatecypher.py

@@ -470,17 +470,12 @@ def main(argv):
 
         dirwalker = os.walk(dirtoscan)
 
-        for direntry in dirwalker:
-            for filename in direntry[2]:
-                fullfilename = os.path.join(direntry[0], filename)
-
-                # Exclude specified directories
-                skip = False
-                for skipdir in skipdirs:
-                    if fullfilename.startswith(skipdir):
-                        skip = True
-                        break
-                if skip:
+        for root, dirs, files in os.walk(dirtoscan, topdown = True):
+            [dirs.remove(d) for d in list(dirs) for s in skipdirs if root.startswith(s)]
+            for filename in files:
+                fullfilename = os.path.join(root, filename)
+
+                if fullfilename == []:
                     continue
 
                 # Exclude broken links
@@ -518,7 +513,7 @@ def main(argv):
                             targetfile = os.path.relpath(targetfile, '/')
                             targetfile = os.path.join(dirtoscan, targetfile)
                         else:
-                            targetfile = os.path.normpath(os.path.join(direntry[0], targetfile))
+                            targetfile = os.path.normpath(os.path.join(root, targetfile))
 
                         # then recursively check whether or not symbolic links
                         # point to other symbolic links or to paths that exist