discord-chunker

Got a large file you wanna share, but don't have Discord Nitro? discord-chunker is a simple python script that splits uploads into 25mb chunks, and then reassembles them on the other side.

---

Usage

Installation

1. Install python 3, if not already installed
2. Place this script somewhere, like ~/.local/bin on Linux or C:\Users\...\Documents on Windows
3. Done!

File Standard

This program operates with the following files:

header: Defines the name, size, length, and byte index of the decoded files

chunk-NN: The packaged data itself

For encoding, these files will be generated. For decoding, place these files in a new empty folder, and call decode mode on that folder.

Chunking

Call the script in encode mode:

python3 chunker.py encode <input_file_or_dir> <output_dir>

This will chunk together the given input_file, or, if given a directory, all the files within input_dir. The output files will be placed in output_dir.

Uploading/Downloading

On the discord side of things, upload all the chunk files that were produced, as well as the header file.

To reproduce the files, download all the chunk files and the header file and place those into a new, empty folder.

De-chunking

Call the script in decode mode:

python3 chunker.py decode <chunk_dir>

This will de-chunk the chunk and header files found in chunk_dir.

To specify the output folder of the new files, use the -o flag:

python3 chunker.py decode <chunk_dir> -o <output_dir>

This will place the decoded files in output_dir, instead of the current working directory.

---

How?

First, the files are specified to the program. An example is shown with the files below.

_A__B_____________C_____D_________E________F____
|  |             |     |         |        |    |
|  |             |     |         |        |    |
|  |             |     |         |        |    |
|  |             |     |         |        |    |
|__|_____________|_____|_________|________|____|

These files are then glued end-to-end, with a header file keeping track of the original file lengths:

_0__________1__________2__________3__________4__
|          |          |          |          |  |
|  |       |     |    ||         |        | |  |
|  A       |     B    |C         D        E |  F
|  |       |     |    ||         |        | |  |
|__________|__________|__________|__________|__|

These chunks are all 25mb in length (or, if specified, some other predetermined length) for uploading to Discord.

Then, on the other side, download all the files. The program will then glue them all back together, and cut appropriately, based on the generated header file:

_A__B_____________C_____D_________E________F____
|  |             |     |         |        |    |
|  |             |     |         |        |    |
|  |             |     |         |        |    |
|  |             |     |         |        |    |
|__|_____________|_____|_________|________|____|

Voilà.