#archive #archive-format #package #file #header #file-path #file-offset

yanked klu

A small library/cli to work with a simple archive format

0.1.0 Jul 4, 2022

#5 in #archive-format

MIT license

17KB
384 lines

KLU

This is a simple archive format. It is split in two parts: - The headers - The data

Headers

Every file / directory has an header with some information about the itself.
it contains the file path (ending with a "/" for directory), the filesize, and file offset inside the archive)
the details on how it is stored is still WIP but currently it is like this: 
```
	file_size: u64,
	file_offset: u64,
	file_path_size: u16,
	file_path: PathBuf (file_path_size bytes long)
```

Data

The file's data is stored in a sequential way, where every files are concatanated together and we only choose which part of we take to grab archive's files;

API

Currently the api is VERY basic: creation of archive from a directory and reading of an archive from a Reader + reading of files.

Reading/Parsing an archive

You should Wrap the reader in a BufReader because it will do a lot of small reads (like 2-8 bytes read)

Creating an archive

For the first version, the only way to create an archive is through the file system. You don't have to wrap the writer in a BufWriter as it is wrapped inside the function.

CLI

The cli allows you to list files, pack and unpack an archive. To get the cli, do cargo install klu --features cli and the use klu help to get the list of commands

Dependencies

~0.3–7.5MB
~57K SLoC