Hawkynt

ROLZ (Reduced Offset LZ)

Context-aware dictionary compression using reduced offset sets. Combines LZ77 dictionary matching with context modeling to reduce active offsets and improve compression efficiency.

Properties

Property Value
Category Compression Algorithms
Sub-category Dictionary
Security status 🎓 Educational Only
Complexity Advanced
Inventor Malcolm Taylor
Year 1999
Origin 🇬🇧 United Kingdom
Source algorithms/compression/rolz.js

Security

Status: 🎓 Educational Only

No vulnerabilities are recorded for this implementation.

Documentation

References

Test vectors

7 vectors ship with this algorithm and run in the test suite. Byte values are hexadecimal.

Vector 1 — Empty input test

Field Value
input (empty)
expected 00000000

Vector 2 — Single character - no context established

Field Value
input 41
expected 010000002080

Vector 3 — Two characters - building context

Field Value
input 4142
expected 02000000209080

Vector 4 — Alternating pattern - context-aware matching

Field Value
input 41424142
expected 040000002090882420

Vector 5 — Repeating sequence - reduced offset advantage

Field Value
input 414243414243
expected 060000002090886412110c

Vector 6 — Natural text with character repetition

Field Value
input 48656c6c6f20576f726c64
expected 0b00000024194d86c37880ae6f391b0c80

Vector 7 — Structured runs with repetition - optimal case

Field Value
input 616161626262636363616161
expected 0c00000030984c26231188c66331984c2610

← All algorithms