fix(general): escape default ignored directories (#6738)

* Escape `ignored_directories`

* Implement windows compatability in `re_dir`

* refactor `EXCLUDED_PATHS`

* Add test for `re_dir`

* Add test: `TestBaseRunner::tests_re_dir_test_pattern`

* typo

---------

Co-authored-by: pazbec <paz8097@gmail.com>

Author: tpvasconcelos

checkov/common/runners/base_runner.py

@@ -44,7 +44,16 @@ def strtobool(val: str) -> int:
         raise ValueError("invalid boolean value %r for environment variable CKV_IGNORE_HIDDEN_DIRECTORIES" % (val,))
 
 
-IGNORED_DIRECTORIES_ENV = os.getenv("CKV_IGNORED_DIRECTORIES", "node_modules,.terraform,.serverless")
+def re_dir(path: str) -> str:
+    """Compile a regex pattern that matches paths containing the given directory at any level."""
+    sep = re.escape(os.sep)  # windows compatibility
+    return rf"(^|.*{sep}){re.escape(path)}($|{sep}.*)"
+
+
+IGNORED_DIRECTORIES_ENV = os.getenv(
+    "CKV_IGNORED_DIRECTORIES",
+    ",".join(re_dir(p) for p in ["node_modules", ".terraform", ".serverless"])
+)
 IGNORE_HIDDEN_DIRECTORY_ENV = strtobool(os.getenv("CKV_IGNORE_HIDDEN_DIRECTORIES", "True"))
 
 ignored_directories = IGNORED_DIRECTORIES_ENV.split(",")

checkov/secrets/utils.py

@@ -4,10 +4,16 @@
 import re
 from collections.abc import Iterable
 
-from checkov.common.runners.base_runner import ignored_directories, safe_remove
+from checkov.common.runners.base_runner import ignored_directories, safe_remove, re_dir
 from checkov.common.util.consts import DEFAULT_EXTERNAL_MODULES_DIR
 
-EXCLUDED_PATHS = [*ignored_directories, DEFAULT_EXTERNAL_MODULES_DIR, ".idea", ".git", "venv"]
+EXCLUDED_PATHS = [
+    *ignored_directories,
+    re_dir(DEFAULT_EXTERNAL_MODULES_DIR),
+    re_dir(".idea"),
+    re_dir(".git"),
+    re_dir("venv"),
+]
 
 
 def filter_excluded_paths(

tests/common/runners/test_base_runner.py

@@ -1,14 +1,47 @@
 import os
+import re
 import unittest
 from typing import Optional, List
 
 from checkov.common.output.report import Report
-from checkov.common.runners.base_runner import filter_ignored_paths, BaseRunner
+from checkov.common.runners.base_runner import filter_ignored_paths, BaseRunner, re_dir
 from checkov.runner_filter import RunnerFilter
 
 
 class TestBaseRunner(unittest.TestCase):
 
+    def test_re_dir(self):
+        sep = '\\' if os.name == 'nt' else '/'
+        # add regex prefix and suffix to the (unmodified) directory name
+        self.assertEqual(re_dir('dir'), fr'(^|.*{sep})dir($|{sep}.*)')
+        # escape the directory name (but leave the os separator unaltered)
+        self.assertEqual(re_dir('.dir1/.dir2'), fr'(^|.*{sep})\.dir1/\.dir2($|{sep}.*)')
+
+    def tests_re_dir_pattern(self):
+        dir_name_to_ignore = ".hidden"
+        dir_name_to_ignore_re = re.compile(re_dir(dir_name_to_ignore))
+        paths_to_ignore = [
+            ".hidden",
+            "/.hidden",
+            "/path/.hidden",
+            "path/.hidden",
+            ".hidden/path",
+            "path/.hidden/path",
+            "path/.hidden/path/",
+        ]
+        paths_to_keep = [
+            ".hidden1",
+            "not.hidden",
+            "nothidden",
+            "not.hidden/path",
+            "path/not.hidden",
+            "also/nothidden",
+            "hidden/not",
+            "also/hidden/not",
+        ]
+        self.assertTrue(all(dir_name_to_ignore_re.match(p) for p in paths_to_ignore))
+        self.assertFalse(any(dir_name_to_ignore_re.match(p) for p in paths_to_keep))
+
     def test_filter_ignored_directories_regex_legacy(self):
         d_names = ['bin', 'integration_tests', 'tests', 'docs', '.github', 'checkov', 'venv', '.git', 'kubernetes', '.idea']
         expected = ['bin', 'docs', 'checkov', 'venv', 'kubernetes']