Files

73 lines
2.0 KiB
Python
Raw Permalink Normal View History

2025-05-07 13:09:41 +04:00
import os
import subprocess
from collections import Counter
2026-03-17 17:58:01 -05:00
CONFIG_FILE_EXTENSIONS = ('.json', '.yml', '.yaml', '.ini', '.conf', '.toml')
2025-05-07 13:09:41 +04:00
def is_text_file(filepath):
# Check for binary file by scanning for null bytes.
try:
2026-03-17 17:58:01 -05:00
with open(filepath, 'rb') as f:
2025-05-07 13:09:41 +04:00
chunk = f.read(4096)
2026-03-17 17:58:01 -05:00
if b'\0' in chunk:
2025-05-07 13:09:41 +04:00
return False
return True
except Exception:
return False
def should_skip_file(path):
base = os.path.basename(path)
# Skip dotfiles and dotdirs
2026-03-17 17:58:01 -05:00
if base.startswith('.'):
2025-05-07 13:09:41 +04:00
return True
# Skip config files by extension
if base.lower().endswith(CONFIG_FILE_EXTENSIONS):
return True
return False
def get_tracked_files():
try:
2026-03-17 17:58:01 -05:00
output = subprocess.check_output(['git', 'ls-files'], text=True)
files = output.strip().split('\n')
2025-05-07 13:09:41 +04:00
files = [f for f in files if f and os.path.isfile(f)]
return files
except subprocess.CalledProcessError:
2026-03-17 17:58:01 -05:00
print('Error: Are you in a git repository?')
2025-05-07 13:09:41 +04:00
return []
def main():
files = get_tracked_files()
email_counter = Counter()
total_lines = 0
for file in files:
if should_skip_file(file):
continue
if not is_text_file(file):
continue
try:
2026-03-17 17:58:01 -05:00
blame = subprocess.check_output(['git', 'blame', '-e', file], text=True, errors='replace')
2025-05-07 13:09:41 +04:00
for line in blame.splitlines():
# The email always inside <>
2026-03-17 17:58:01 -05:00
if '<' in line and '>' in line:
2025-05-07 13:09:41 +04:00
try:
2026-03-17 17:58:01 -05:00
email = line.split('<')[1].split('>')[0].strip()
2025-05-07 13:09:41 +04:00
except Exception:
continue
email_counter[email] += 1
total_lines += 1
except subprocess.CalledProcessError:
continue
for email, lines in email_counter.most_common():
percent = (lines / total_lines * 100) if total_lines else 0
2026-03-17 17:58:01 -05:00
print(f'{email}: {lines}/{total_lines} {percent:.2f}%')
2025-05-07 13:09:41 +04:00
2026-03-17 17:58:01 -05:00
if __name__ == '__main__':
2025-05-07 13:09:41 +04:00
main()