add: Duplicate Check

This commit is contained in:
inference 2025-06-25 22:08:20 +00:00
parent f8c2f1eb81
commit 1c5b5ff52b
Signed by: inference
SSH Key Fingerprint: SHA256:/O3c09/4f1lh4zrhFs2qvQEDda6dZbTwG9xEcj8OfWo

45
duplicate_check.py Normal file
View File

@ -0,0 +1,45 @@
#!/usr/bin/env python
# Duplicate Check
# Version: 1.0.0-beta.1
# Copyright 2025 Jake Winters
# SPDX-License-Identifier: BSD-3-Clause
import os
import hashlib
import argparse
parser = argparse.ArgumentParser(description='Scan directory for duplicate files and delete them.')
parser.add_argument('--dry-run', '-d', action='store_true', help='Detect duplicates without deletion.')
parser.add_argument('directory', type=str, help='The directory to scan for duplicate files.')
args = parser.parse_args()
def hash_file(file_path):
sha256_hash = hashlib.sha256()
with open(file_path, 'rb') as f:
for byte_block in iter(lambda: f.read(65536), b''):
sha256_hash.update(byte_block)
return sha256_hash.hexdigest()
def find_and_delete_duplicates(directory, dry_run):
os.chdir(directory)
file_hashes = {}
for file in os.listdir():
file_path = os.path.abspath(file)
if os.path.isfile(file_path):
file_hash = hash_file(file_path)
if file_hash in file_hashes:
if not dry_run:
os.remove(file_path)
print(f"Duplicate detected: {file_path}")
if not dry_run:
print(f"Duplicate deleted: {file_path}")
else:
file_hashes[file_hash] = file_hash
os.chdir('..')
find_and_delete_duplicates(args.directory, args.dry_run)