Lethe

Lethe is a cross-platform OCaml script designed to anonymize files in a specified directory by renaming them to random strings. Inspired by the mythological river of forgetfulness, Lethe erases identifiable information from files, ensuring greater privacy and security.

Features

• Random Renaming: Each file in the target directory is renamed to a unique random string, preserving the original file extension.
• Timestamp Handling: The script updates the access and modification timestamps of each file to the current time. Additionally, by deleting the original file and creating a new one, it also sets the creation time to the moment of creation.
• Cross-Platform Compatibility: Works seamlessly on both Linux and Windows platforms.

Usage

1. Ensure you have OCaml and Dune installed.

2. Clone the repository:

   git clone https://github.com/paulo-granthon/lethe
   cd lethe

3. Build the project using Dune:

   dune build

4. Run the script with the target directory as an argument:

   dune exec lethe path/to/directory

Example

Consider the following directory structure:

Date Modified Date Created Date Accessed Name
26 Oct 01:50  28 Oct 22:56 28 Oct 22:58   ./test_files
17 Jun 18:19  17 Jun 18:15 17 Jun 18:16  ├──  file1.txt
10 Mar 20:29  10 Mar 20:24 10 Mar 20:29  ├──  file2.txt
25 May 06:19  25 May 06:19 25 May 06:15  ├──  file3.txt
21 Jan 05:46  21 Jan 05:43 21 Jan 05:41  ├──  file4.txt
07 Apr 07:51  07 Apr 07:51 07 Apr 07:51  └──  other_extension_file.md

After running the script:

dune exec lethe ./test_files/

The directory structure will be transformed as follows:

Date Modified Date Created Date Accessed Name
28 Oct 22:58  28 Oct 22:56 28 Oct 22:58   ./test_files
28 Oct 22:58  28 Oct 22:58 28 Oct 22:58  ├──  1eZ57lpSLB.md
28 Oct 22:58  28 Oct 22:58 28 Oct 22:58  ├──  CuXJnUtLQB.txt
28 Oct 22:58  28 Oct 22:58 28 Oct 22:58  ├──  fgmrkGjV17.txt
28 Oct 22:58  28 Oct 22:58 28 Oct 22:58  ├──  LGtETtwKJc.txt
28 Oct 22:58  28 Oct 22:58 28 Oct 22:58  └──  RmosuLFz8t.txt

Limitations

• Large Files: The current implementation reads the entire contents of files into memory, which may not be suitable for very large files. Consider implementing chunk-based reading for such cases.
• File Permissions: The new files may have different permissions compared to the originals. Adjustments might be necessary based on specific use cases.

License

This project is licensed under the MIT License. See the LICENSE file for more information.