From 7474b96d5ade2c1b569617253225f4bc139da926 Mon Sep 17 00:00:00 2001
From: hx507 <72583014+hx507@users.noreply.github.com>
Date: Mon, 20 Jun 2022 11:18:34 -0700
Subject: [PATCH] Parallelize clang-tidy hook

Currently clang-tidy hook is running in serial, which can be slow if multiple large files are modified.
Clang-tidy is trivially parallelize-able per file, so why don't we take advantage of that?
Locally it speeds up a ~8 sec clang-tidy run to ~3 sec, which is bottle-necked on largest file changed -- still much faster.
---
 hooks/clang_tidy.py | 32 +++++++++++++++-----------------
 1 file changed, 15 insertions(+), 17 deletions(-)

diff --git a/hooks/clang_tidy.py b/hooks/clang_tidy.py
index d280269..2aa3bc8 100755
--- a/hooks/clang_tidy.py
+++ b/hooks/clang_tidy.py
@@ -1,8 +1,7 @@
-#!/usr/bin/env python3
-"""Wrapper script for clang-tidy."""
 import re
 import sys
 from typing import List
+import multiprocessing as mp
 
 from hooks.utils import StaticAnalyzerCmd
 
@@ -18,21 +17,20 @@ def __init__(self, args: List[str]):
         self.parse_args(args)
         self.edit_in_place = "-fix" in self.args or "--fix-errors" in self.args
 
+    def tidy_file(self, filename: List[str]):
+        self.run_command([filename] + self.args)
+        # Warnings generated aren't important.
+        self.stderr = re.sub(rb"[\d,]+ warning \S+\s+", b"", self.stderr)
+        if len(self.stderr) > 0 and "--fix-errors" in self.args:
+            self.returncode = 1
+        return self.returncode, self.stderr
+
     def run(self):
         """Run clang-tidy. If --fix-errors is passed in, then return code will be 0, even if there are errors."""
-        for filename in self.files:
-            self.run_command([filename] + self.args)
-            # Warnings generated aren't important.
-            self.stderr = re.sub(rb"[\d,]+ warning \S+\s+", b"", self.stderr)
-            if len(self.stderr) > 0 and "--fix-errors" in self.args:
-                self.returncode = 1
-            self.exit_on_error()
-
+        with mp.Pool(mp.cpu_count()) as p:
+            results = p.map(self.tidy_file, self.files)
+            ret_codes, stderrs = zip(*results)
 
-def main(argv: List[str] = sys.argv):
-    cmd = ClangTidyCmd(argv)
-    cmd.run()
-
-
-if __name__ == "__main__":
-    main()
+            self.returncode = max(ret_codes)
+            self.stderr = b"\n".join(stderrs)
+            self.exit_on_error()