⚡ nzfat (WIP)

A very generic and configurable FAT12/16/32 implementation library with VFAT support written in zig. Almost all the features are configurable, from long filename support (and the max number of characters it supports) to the maximum supported FAT type (12/16/32).

❓ Usage

The library is not mature enough for production usage, but if you still want to use this, here's some little info:

The FatFilesystem type only needs two arguments, the block device context and its configuration. Some methods change their signature depending on the configuration and all methods require the block context to be passed to it. This library never stores the passed block device anywhere.

For example, mounting a FAT filesystem might be done like this:

const nzfat = @import("nzfat");
const Fat = nzfat.FatFilesystem(...);
// ...
var fat_ctx = try Fat.mount(&blk);
// See zfat.MountError

To traverse the directory entries you have directoryIterator and directoryIteratorContext at your disposal:

// TODO: Explain why directoryIteratorContext is sometimes needed for long filenames.
var dir_it = fat_ctx.directoryIterator(fat_ctx.getRoot());
defer dir_it.deinit();

while (try dir_it.next(&blk)) |entry| : (entries += 1) {
    // Do something with entry 
}

As searching for a specific entry in a directory is very common, searchEntry and searchEntryContext abstract away the case-insensitive nature of the filesystem:

// TODO: Explain that for long filenames a case-insensitive comparison function must be provided as its out of scope for this project
if(try fat_ctx.searchEntry(&blk, handle, name)) |entry| {
    // Do something with the entry we found...
}

📝 TODO

Small TODO's

Searching for 1 free entry in a directory (a.k.a: Short filename only)
Searching for a free cluster linearly
Deletion of files and directories
Short filename alternatives when using a VFAT FatFilesystem
Searching for N free clusters for file and directory creation
Allocate new directory entries if no entries found and not in root (FAT12/16 only)
Creation of directory entries with LFN entries if needed
Searching for N free entries in a directory
API to allocate clusters for files
API to modify dates and times and attributes in entries

Big TODO's

Implement some sort of I/O cache? Or leave it to the BlockDevice implementation?
Some sort of cache strategy for FAT entries if requested.
Reorganize/rename and API Freeze