Simple Duplicate Finder

Introduction

This program can find and display files that share one of these features:

• A SHA-1 hash;
• An Adler-32 checksum;
• A modification date, or
• A file name.

Choosing SHA-1 and Adler-32 will open all files in the file system under the chosen directory, altering their access dates, but checking the modification time or file name will only use a file's metadata, which is much faster to retrieve. Other considerations:

• SHA-1 was originally intended as a cryptographic hashing algorithm and is the only method that should always and only return duplicates. It is also the slowest of all methods.
• Adler-32 was made for data compression rather than hashing, and it is particularly not reliable with small files.
• A modification date shared by two or more files is no guarantee that the files are the same. For instance, they could have been decompressed at the same time.
• If two files share the name, it is all that is meant. This option does not check for anything else.

Controls

• Double click: opens a file.
• Delete key: deletes a file. By default, there is a confirmation dialog.
• The tree allows multiple selection, with Shift + click or Ctrl + click.

Coming soon

• Context menu on right click with the options:
  • Open the containing directory (with one file only);
  • Rename (id.);
  • Swap file names (with two files only);
  • Move... (one to multiple files);
  • Delete selected (one to multiple files);
  • Keep the oldest (with two to multiple files),
  • Bulk rename... (id.)

Installation

Linux

This program will run, provided that all dependencies are installed (e.g. PyGObject), in this way:

git clone # paste the URL to this repository
cd simple-duplicate-finder
make translations # for German, Italian, Spanish and Portuguese
src/main.pyw

Windows

An executable file is included for convenience. It was made with PyInstaller and may take a few seconds to start.

Notes

This program was meant as a GUI alternative to the Unix command below, which is actually faster, as a sort command is only run once, but the list of files is not shown until it finishes running (unlike the tree view that is updated with each file that is found).

find . -type f -exec shasum {} \; | sort | uniq -w 40 -D

A very similar output can be achieved by saving the tree view on this program as a CSV file, with the button on the bottom right.