Merge bitcoin/bitcoin#24929: lint: convert shell locale linter test to Python

2c838cc309b2e3e1e30344178c17f789381a8b6b lint: convert shell locale linter test to Python (Eunoia)

Pull request description:

  Refs #24783

ACKs for top commit:
  laanwj:
    Code review ACK 2c838cc309b2e3e1e30344178c17f789381a8b6b

Tree-SHA512: 3cb5e7c7cd2acbdf0dc45096055b33cbfa0ec9e47ea567452d23a49a7441b3b62a8416879f234459c86fa892c42205c91d8a575115346c023ab0152cf713e20c

2 files changed, 67 insertions, 25 deletions

diff --git a/test/lint/lint-shell-locale.py b/test/lint/lint-shell-locale.py
new file mode 100755
index 0000000000..f3dfe18a95
--- /dev/null
+++ b/test/lint/lint-shell-locale.py
@@ -0,0 +1,67 @@
+#!/usr/bin/env python3
+#
+# Copyright (c) 2018-2022 The Bitcoin Core developers
+# Distributed under the MIT software license, see the accompanying
+# file COPYING or http://www.opensource.org/licenses/mit-license.php.
+
+"""
+Make sure all shell scripts are:
+a.) explicitly opt out of locale dependence using
+    "export LC_ALL=C" or "export LC_ALL=C.UTF-8", or
+b.) explicitly opt in to locale dependence using the annotation below.
+"""
+
+import subprocess
+import sys
+import re
+
+OPT_IN_LINE = '# This script is intentionally locale dependent by not setting \"export LC_ALL=C\"'
+
+OPT_OUT_LINES = [
+    'export LC_ALL=C',
+    'export LC_ALL=C.UTF-8',
+]
+
+def get_shell_files_list():
+    command = [
+        'git',
+        'ls-files',
+        '--',
+        '*.sh',
+    ]
+    try:
+        return subprocess.check_output(command, stderr = subprocess.STDOUT).decode('utf-8').splitlines()
+    except subprocess.CalledProcessError as e:
+        if e.returncode > 1: # return code is 1 when match is empty
+            print(e.output.decode('utf-8'), end='')
+            sys.exit(1)
+        return []
+
+def main():
+    exit_code = 0
+    shell_files = get_shell_files_list()
+    for file_path in shell_files:
+        if re.search('src/(secp256k1|minisketch|univalue)/', file_path):
+            continue
+
+        with open(file_path, 'r', encoding='utf-8') as file_obj:
+            contents = file_obj.read()
+
+        if OPT_IN_LINE in contents:
+            continue
+
+        non_comment_pattern = re.compile(r'^\s*((?!#).+)$', re.MULTILINE)
+        non_comment_lines = re.findall(non_comment_pattern, contents)
+        if not non_comment_lines:
+            continue
+
+        first_non_comment_line = non_comment_lines[0]
+        if first_non_comment_line not in OPT_OUT_LINES:
+            print(f'Missing "export LC_ALL=C" (to avoid locale dependence) as first non-comment non-empty line in {file_path}')
+            exit_code = 1
+
+    return sys.exit(exit_code)
+
+if __name__ == '__main__':
+    main()
+
diff --git a/test/lint/lint-shell-locale.sh b/test/lint/lint-shell-locale.sh
deleted file mode 100755
index 4c6b8a57e6..0000000000
--- a/test/lint/lint-shell-locale.sh
+++ /dev/null
@@ -1,25 +0,0 @@
-#!/usr/bin/env bash
-#
-# Copyright (c) 2018-2020 The Bitcoin Core developers
-# Distributed under the MIT software license, see the accompanying
-# file COPYING or http://www.opensource.org/licenses/mit-license.php.
-#
-# Make sure all shell scripts:
-# a.) explicitly opt out of locale dependence using
-#     "export LC_ALL=C" or "export LC_ALL=C.UTF-8", or
-# b.) explicitly opt in to locale dependence using the annotation below.
-
-export LC_ALL=C
-
-EXIT_CODE=0
-for SHELL_SCRIPT in $(git ls-files -- "*.sh" | grep -vE "src/(secp256k1|minisketch|univalue)/"); do
-    if grep -q "# This script is intentionally locale dependent by not setting \"export LC_ALL=C\"" "${SHELL_SCRIPT}"; then
-        continue
-    fi
-    FIRST_NON_COMMENT_LINE=$(grep -vE '^(#.*)?$' "${SHELL_SCRIPT}" | head -1)
-    if [[ ${FIRST_NON_COMMENT_LINE} != "export LC_ALL=C" && ${FIRST_NON_COMMENT_LINE} != "export LC_ALL=C.UTF-8" ]]; then
-        echo "Missing \"export LC_ALL=C\" (to avoid locale dependence) as first non-comment non-empty line in ${SHELL_SCRIPT}"
-        EXIT_CODE=1
-    fi
-done
-exit ${EXIT_CODE}