Sync only changed files between two folders in Python

This code compares two folders and copies only the new or modified files from source to destination, skipping unchanged ones by comparing SHA-256 hashes.

Easy Python 3.9+ Aug 9, 2026 Files & data 16 views 0 copies

files sync hashing pathlib shutil

Python code

36 lines

Python 3.9+

import hashlib
from pathlib import Path
import shutil

def file_hash(path: Path, chunk_size: int = 8192) -> str:
    hasher = hashlib.sha256()
    with path.open("rb") as f:
        for chunk in iter(lambda: f.read(chunk_size), b""):
            hasher.update(chunk)
    return hasher.hexdigest()

def sync_files(src: str, dst: str) -> list[str]:
    src_path = Path(src)
    dst_path = Path(dst)
    dst_path.mkdir(parents=True, exist_ok=True)
    synced = []

    for src_file in src_path.rglob("*"):
        if src_file.is_file():
            rel_path = src_file.relative_to(src_path)
            dst_file = dst_path / rel_path
            if not dst_file.exists() or file_hash(src_file) != file_hash(dst_file):
                dst_file.parent.mkdir(parents=True, exist_ok=True)
                shutil.copy2(src_file, dst_file)
                synced.append(str(rel_path))

    return synced

if __name__ == "__main__":
    changed = sync_files("source_folder", "backup_folder")
    if changed:
        print("Synced files:")
        for path in changed:
            print(f"  {path}")
    else:
        print("No changes detected — everything is up to date.")

Output

stdout

Synced files:
  data.txt
  reports/q1.csv

How it works

The script walks the entire source tree with Path.rglob("*"), so it captures files in subfolders too. For each file, it computes a SHA-256 hash in chunks, which avoids loading large files into memory all at once. It only copies when the destination file is missing or its hash differs, so unchanged files are skipped. Directories under the destination are created on demand with mkdir(parents=True, exist_ok=True). Using shutil.copy2 preserves metadata like timestamps, which is useful for backups.

Common mistakes

Using `os.path.exists` on directories and forgetting that only files should be compared — always check `is_file()`.
Hashing large files without chunking can exhaust memory; use a loop that reads fixed-size blocks.
Forgetting to create destination subdirectories before copying, which raises `FileNotFoundError`.
Comparing file modification times instead of content hashes may miss changes when timestamps are preserved but content differs.

Variations

Use `filecmp.cmp(src_file, dst_file, shallow=False)` to compare content directly without computing hashes yourself.
For very large trees, use `os.walk` combined with a set of relative paths to speed up the check.

Real-world use cases

Incremental backups that copy only new or changed files to a backup location, saving time and storage.
Deploying website assets or configs to a server by syncing only files that changed between the local and remote directory.
Mirroring a dataset folder to a shared network drive, skipping unchanged files to minimize transfer time.

Sync only changed files between two folders in Python

Python code

Output

How it works

Common mistakes

Variations

Real-world use cases

More from Files & data

Tutorials

Quizzes

Python code

Output

How it works

Common mistakes

Variations

Real-world use cases

More from Files & data

Keep learning

Tutorials

Quizzes