Practracker: improve exclude-directory logic

Instead of excluding directories at the last minute if they happen
to appear in our filenames, we exclude them early, before recursing
into all their subdirectories.

Part of 29746.

diff --git a/scripts/maint/practracker/util.py b/scripts/maint/practracker/util.py
index b0ca73b997f87d8b1d7eda2febc70a39c4a6a098..5a8876a0f65bd1cb2bfe95950b0de98d9fa82bbc 100644 (file)
--- a/scripts/maint/practracker/util.py
+++ b/scripts/maint/practracker/util.py
@@ -2,15 +2,24 @@ import os
 
 # We don't want to run metrics for unittests, automatically-generated C files,
 # external libraries or git leftovers.
-EXCLUDE_SOURCE_DIRS = {"/src/test/", "/src/trunnel/", "/src/ext/", "/.git/"}
+EXCLUDE_SOURCE_DIRS = {"src/test/", "src/trunnel/", "src/rust/",
+                       "src/ext/", ".git/"}
+
+def _norm(p):
+    return os.path.normcase(os.path.normpath(p))
 
 def get_tor_c_files(tor_topdir):
     """
     Return a list with the .c filenames we want to get metrics of.
     """
     files_list = []
+    exclude_dirs = { _norm(os.path.join(tor_topdir, p)) for p in EXCLUDE_SOURCE_DIRS }
+
 
     for root, directories, filenames in os.walk(tor_topdir):
+        # Remove all the directories that are excluded.
+        directories[:] = [ d for d in directories
+                           if _norm(os.path.join(root,d)) not in exclude_dirs ]
         directories.sort()
         filenames.sort()
         for filename in filenames:
@@ -18,10 +27,7 @@ def get_tor_c_files(tor_topdir):
             if not filename.endswith(".c"):
                 continue
 
-            # Exclude the excluded paths
             full_path = os.path.join(root,filename)
-            if any(os.path.normcase(exclude_dir) in full_path for exclude_dir in EXCLUDE_SOURCE_DIRS):
-                continue
 
             files_list.append(full_path)