From a39edd21ec784fceae6569e5c85ef82b2d8c3e77 Mon Sep 17 00:00:00 2001
From: "codeflash-ai[bot]"
 <148906541+codeflash-ai[bot]@users.noreply.github.com>
Date: Tue, 25 Feb 2025 23:29:03 +0000
Subject: [PATCH] =?UTF-8?q?=E2=9A=A1=EF=B8=8F=20Speed=20up=20method=20`Ass?=
 =?UTF-8?q?ertCleanup.=5Ftransform=5Fassert=5Fline`=20by=2034%=20in=20PR?=
 =?UTF-8?q?=20#26=20(`clean=5Fconcolic=5Ftests`)=20To=20optimize=20the=20`?=
 =?UTF-8?q?AssertCleanup`=20class,=20we=20can=20improve=20the=20`=5Ftransf?=
 =?UTF-8?q?orm=5Fassert=5Fline`=20method=20by=20reducing=20the=20use=20of?=
 =?UTF-8?q?=20regular=20expressions,=20and=20replacing=20them=20with=20mor?=
 =?UTF-8?q?e=20efficient=20string=20operations=20where=20possible.=20Where?=
 =?UTF-8?q?=20regular=20expressions=20are=20still=20necessary,=20we=20comp?=
 =?UTF-8?q?ile=20them=20once=20and=20reuse=20them.=20Here's=20the=20refact?=
 =?UTF-8?q?ored=20code.?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

### Explanation of Changes.
1. **Regex Compilation in `__init__`**: Compiled the regular expressions in the `__init__` method to avoid recompiling them every time `_transform_assert_line` is called, improving speed.

2. **String Operations for Trailing Characters**: Replaced `re.sub` used to strip trailing commas or semicolons with simpler string operations, improving efficiency.

These improvements help in optimizing the running speed of the program while maintaining the same functionality.
---
 codeflash/code_utils/code_replacer.py | 16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)

diff --git a/codeflash/code_utils/code_replacer.py b/codeflash/code_utils/code_replacer.py
index 269dd4706..28e252c47 100644
--- a/codeflash/code_utils/code_replacer.py
+++ b/codeflash/code_utils/code_replacer.py
@@ -354,21 +354,23 @@ def transform_asserts(self, code: str) -> str:
         return "\n".join(result_lines)
 
     def _transform_assert_line(self, line: str) -> Optional[str]:
-        indent = line[: len(line) - len(line.lstrip())]
+        indent_len = len(line) - len(line.lstrip())
+        indent = line[:indent_len]
 
-        assert_match = re.match(r"\s*assert\s+(.*?)(?:\s*==\s*.*)?$", line)
+        assert_match = self.assert_pattern.match(line)
         if assert_match:
             expression = assert_match.group(1).strip()
             if expression.startswith("not "):
                 return f"{indent}{expression}"
 
-            expression = re.sub(r"[,;]\s*$", "", expression)
+            # Removing trailing commas or semicolons without using regex
+            if expression and expression[-1] in ",;":
+                expression = expression[:-1]
             return f"{indent}{expression}"
 
-        unittest_match = re.match(r"(\s*)self\.assert([A-Za-z]+)\((.*)\)$", line)
+        unittest_match = self.unittest_pattern.match(line)
         if unittest_match:
             indent, assert_method, args = unittest_match.groups()
-
             if args:
                 arg_parts = self._split_top_level_args(args)
                 if arg_parts and arg_parts[0]:
@@ -399,6 +401,10 @@ def _split_top_level_args(self, args_str: str) -> list[str]:
 
         return result
 
+    def __init__(self):
+        self.assert_pattern = re.compile(r"\s*assert\s+(.*?)(?:\s*==\s*.*)?$")
+        self.unittest_pattern = re.compile(r"(\s*)self\.assert([A-Za-z]+)\((.*)\)$")
+
 
 def clean_concolic_tests(test_suite_code: str) -> str:
     try: