Varint-rs

Varint is an alternative way of storing integer numbers.

Varints allow for the storage of larger integer types in a smaller amount of space. It does this by storing an integer using the 7 lower bits and a flag in the most-significant bit. This flag is set to 1 when more bytes should be read. The groups of 7 bits are then added from the least-significant group first.

Features

• signed (default): allows for signed integers to be encoded and decoded using zigzag encoding
• std (default): implements the VarintReader and VarintWriter traits respectively on:
  • all std::io::Read implementors
  • all std::io::Write implementors

Note: Disabling the std feature (which is enabled by default) allows for the crate to be used in a #![no_std] environment.

Example

// to allow the use of the `VarintWriter::write_*_varint` functions
use varint_rs::VarintWriter;
// to allow the use of the `VarintReader::read_*_varint` functions
use varint_rs::VarintReader;

// an example to use for the buffer
use std::io::Cursor;

// create an i32 set to `300`
let number: i32 = 300;
// create a buffer for the varint to be writen to
// an i32 can be `4` bytes maximum, so we pre-allocate the capacity
let mut buffer: Cursor<Vec<u8>> = Cursor::new(Vec::with_capacity(4));

// now we can write the varint into the buffer
// `300` should only use `2` bytes instead of all `4`
// the `write_*_varint` functions may return an `std::io::Error`
buffer.write_i32_varint(number).unwrap();

// we reset the cursor pos back to `0`, this isn't varint stuff
buffer.set_position(0);

// now we can read the varint from the buffer
// we should read `300` which was the number we stored
// the `read_*_varint` functions may return an `std::io::Error`
let number: i32 = buffer.read_i32_varint().unwrap();

Note: This example assumes that the default features are in use.

Credit

Much of this code is ported from the varint crate by Cruz Bishop. A massive thanks to them for the awesome alogrithms!