dups

dups is a CLI tool to find and remove duplicate files using different hash algorithms (MD5, SHA256 and XXHash) with multi-core support.

Install

Download binaries:

Release Page

To use in a go project:

go get github.com/Navid2zp/dups

Usage

CLI

Available Commands:

Command	Description
clean	Finds duplicate files in a given path and deletes them.
scan	Finds duplicate files in a given path but doesn't delete them.
help	Help about any command

Flags:

flag	Description
--algorithm string	algorithm to use (md5/sha256/xxhash) (default "md5")
-f, --flat	flat output, no extra info (only prints duplicate files)
-r, --full	full search (search in sub-directories too)
--min-size int	minimum file size to scan in bytes (default 10)
-s, --single-core	use single cpu core

Examples:

Remove duplicates bigger than 1KB using multiple cpu cores:

dups clean path/to/directory --min-size 1024

Find duplicates and write them into file.txt:

dups scan path/to/directory -f >> file.txt

Find and list duplicates using single cpu core and XXHash algorithm:

dups scan path/to/directory -s --algorithm xxhash

Go code:

package main

import (
	"fmt"
	"github.com/Navid2zp/dups"
)

func main()  {
	// list all files including files in any sub-directory
	files, err := dups.GetFiles("path/to/directory", true)
	if err != nil {
		panic(err)
	}

        // group files based on their file size
        fileGroups, totalFiles := dups.GroupFiles(files, 128)

	// collect hashes for groups with more than one file
	// singleThread: use a single thread
	// flatt: don't print the process bar or any other information
	hashes := dups.CollectHashes(fileGroups, false, dups.XXHash, false, totalFiles)
	duplicates, filesCount, duplicatesCount := dups.GetDuplicates(hashes)
	fmt.Println("total number of files with duplicates:", filesCount)
	fmt.Println("total number of duplicate files:", duplicatesCount)

	freedSize, deletedCount, err := dups.RemoveDuplicates(duplicates)
	if err != nil {
		panic(err)
	}
	fmt.Println("remove", deletedCount, "files")
	fmt.Println("freed a total of ", freedSize, "bytes")
}

Notes:

• Use single core option (-s) if files are big (depending on your disk type).
• Use XXHash algorithm for fast scanning and MD5/SHA256 for safest scanning or if the number of files is huge.

Build from source:

go build -tags multicore if you are building using Go < 1.5 or edit runtime.GOMAXPROCS() manually to support multi-core.

License

Apache