Struct PpmModel 

pub struct PpmModel { /* private fields */ }

The central PPM model. Maintains up to max_order contexts and dynamically updates symbol frequencies as you encode or decode.

Higher max_order (e.g. Some(8)) means the model looks at up to 8 previous bytes for each prediction:

• Pros: Better predictions and higher compression ratio
• Cons: More memory and CPU overhead

Lower max_order (e.g. None → default order 5) is faster and lighter, but compresses less effectively.

Examples

use ppmd_core::{encode_file, decode_file, PpmResult};

fn main() -> PpmResult<()> {
    // Use default order = 5
    encode_file("input.bin", "out.ppm", None)?;

    // Use a custom order = 8 for potentially better compression
    encode_file("input.bin", "out8.ppm", Some(8))?;

    // Decode (always uses order 5)
    decode_file("out.ppm", "decoded.bin")?;
    Ok(())
}

Implementations

impl PpmModel

pub fn new(max_order: u8) -> PpmResult<Self>

Create a new PPM model with contexts up to max_order (1..=16).

Panics

Panics if max_order == 0 or max_order > 16.

pub fn encode<R: Read, W: Write>(&mut self, input: R, output: W) -> PpmResult<W>

Encode the entire contents of input into output, updating the model adaptively as you go.

pub fn decode_symbol<R: Read>( &mut self, decoder: &mut RangeDecoder<R>, history: &mut Vec<u8>, out: &mut [u8], ) -> PpmResult<()>

Decode one symbol at a time from decoder, writing to out, and update the model adaptively.

This is used by decode_file under the hood.