SplitFolder

繁體中文 | 简体中文 | 日本語 | Español | Deutsch

---

Split files in an oversized folder into evenly distributed subfolders by filename — built for folders too large to load smoothly in Finder / Explorer. Works for any file type: photos, documents, archives, logs, datasets, and more. Supports SMB shares, dry-run preview, and re-splitting of existing partitions.

Features

• Name-based grouping — Related files sharing the same prefix+number (e.g., report_001.pdf, report_001.xlsx) stay together
• Natural sort — file2 sorts before file10, not lexicographically
• Even distribution — File groups are spread as evenly as possible across target folders
• Dry-run mode — Preview the result without moving any files
• Re-splitting — Already-split part_* folders are detected and can be re-distributed
• Recursive scanning — Optionally scan the entire directory tree
• SMB / network share safe — Uses os.rename for same-filesystem moves, falling back to copy+delete
• macOS junk filtering — Automatically skips .DS_Store and ._* files
• Duplicate protection — If a filename already exists in the target, renames to xxx__dupN.ext
• Empty folder cleanup — Removes empty old part_* folders after redistribution

Use Cases

• Oversized folders — A single folder with 10,000+ files that takes forever to open? Split it into smaller chunks.
• Photo collections — Distribute RAW/ARW/HIF/XMP files across multiple drives or network shares.
• Log archives — Partition millions of log files by name range for easier browsing.
• Dataset preparation — Split training data into balanced shards.
• Any large flat folder — If a folder is too big to load, SplitFolder can help.

Requirements

• Python 3.7+

No external dependencies — uses only the standard library.

Installation

git clone https://github.com/LeonJian/SplitFolder.git
cd SplitFolder

Usage

python3 main.py /path/to/large/folder -n 10

Splits all files into 10 subfolders: part_001_*, part_002_*, ..., part_010_*.

Examples

# Split into 5 parts with dry-run preview
python3 main.py /Volumes/Data/Archive -n 5 --dry-run

# Split into 20 parts with custom prefix
python3 main.py /Volumes/Data/Archive -n 20 --prefix batch_

# Recursively scan all subdirectories
python3 main.py /Volumes/Data/Archive -n 10 --recursive

# Re-split existing part folders (detected by default)
python3 main.py /Volumes/Data/Archive -n 40

# Skip empty folder cleanup
python3 main.py /Volumes/Data/Archive -n 10 --no-clean-empty

Options

Option	Default	Description
source	(required)	Path to the source folder
-n, --parts	(required)	Number of target subfolders
--prefix	part_	Prefix for output folder names
--recursive	off	Recursively scan the entire directory tree
--dry-run	off	Preview only — do not move files
--no-clean-empty	off	Skip cleaning empty old part_* folders

How It Works

1. Scan — Collects all files from the source folder (and optionally subdirectories)
2. Group — Groups files by their name prefix+number (e.g., report_001.pdf + report_001.xlsx → group report_001)
3. Sort — Sorts groups using natural number ordering
4. Distribute — Evenly spreads groups across N target folders
5. Move — Moves files into their target folders, preserving filenames
6. Cleanup — Removes empty old part_* folders

Folder Naming Convention

part_001_report_0001-report_0500/
part_002_report_0501-report_1000/
part_003_report_1001-report_1500/
...

Each folder name shows the range of file groups it contains.

License

Apache License 2.0