Hawkynt

LZMAT

Real-time compression using match tables instead of hash chains. Developed by Vitaly Evseenko, LZMAT balances fast compression/decompression speed with good compression ratios. Uses efficient match table lookups for pattern finding.

Properties

Property Value
Category Compression Algorithms
Sub-category Dictionary
Security status Not classified
Complexity Intermediate
Inventor Vitaly Evseenko
Year 2007
Origin 🇷🇺 Russia
Source algorithms/compression/lzmat.js

Security

Status: not classified — treat as unverified.

No vulnerabilities are recorded for this implementation.

Documentation

References

Test vectors

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

Vector 1 — No repetition - all literals

Field Value
input 41424344
expected 0041004200430044

Vector 2 — Single character repetition

Field Value
input 41414141
expected 004101000103

Vector 3 — Pattern repetition - ABCABC

Field Value
input 414243414243
expected 00410042004301000303

Vector 4 — Overlapping pattern - ABABABAB

Field Value
input 4142414241424142
expected 0041004201000206

Vector 5 — English sentence compression

Field Value
input 54686520717569636b2062726f776e20 666f78206a756d7073206f7665722074 6865206c617a7920646f67
expected (empty)

Vector 6 — Repetitive text long enough to exercise the maximum match length

Field Value
input 74686520717569636b2062726f776e20 666f78206a756d7073206f7665722074 6865206c617a7920646f672e20746865 20717569636b2062726f776e20666f78 … (1000 bytes; the full value is in the source)
expected (empty)

← All algorithms