Hawkynt

BWT (Burrows-Wheeler Transform)

Reversible data transformation that rearranges string characters to improve performance of other compression techniques. Used as preprocessing step in bzip2 and other advanced compressors.

Properties

Property Value
Category Compression Algorithms
Sub-category Transform
Security status Not classified
Complexity Advanced
Inventor Michael Burrows, David Wheeler
Year 1994
Origin πŸ‡ΊπŸ‡Έ United States
Source algorithms/compression/bwt.js

Security

Status: not classified β€” 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 β€” Empty data test - still emits the 4-byte primary-index header

Source: Edge case test

Field Value
input (empty)
expected 00000000

Vector 2 β€” Single byte test

Source: Minimal transformation test

Field Value
input 41
expected 0000000041

Vector 3 β€” Regression: all 256 byte values

Source: Regression test for sentinel-free cyclic rotation sort

Field Value
input 000102030405060708090a0b0c0d0e0f 101112131415161718191a1b1c1d1e1f 202122232425262728292a2b2c2d2e2f 303132333435363738393a3b3c3d3e3f 404142434445464748494a4b4c4d4e4f 505152535455565758595a5b5c5d5e5f 606162636465666768696a6b6c6d6e6f 707172737475767778797a7b7c7d7e7f 808182838485868788898a8b8c8d8e8f 909192939495969798999a9b9c9d9e9f a0a1a2a3a4a5a6a7a8a9aaabacadaeaf b0b1b2b3b4b5b6b7b8b9babbbcbdbebf c0c1c2c3c4c5c6c7c8c9cacbcccdcecf d0d1d2d3d4d5d6d7d8d9dadbdcdddedf e0e1e2e3e4e5e6e7e8e9eaebecedeeef f0f1f2f3f4f5f6f7f8f9fafbfcfdfeff
expected 00000000ff000102030405060708090a 0b0c0d0e0f101112131415161718191a 1b1c1d1e1f202122232425262728292a 2b2c2d2e2f303132333435363738393a … (260 bytes; the full value is in the source)

Vector 4 β€” Regression: pseudo-random data, length 91 - exercises the gap-doubling tie-break passes

Source: Regression test - non-repeating pseudo-random input

Field Value
input 000040004000400040003980c0000000 40800040004000000040000000004000 00400000400000408000003980000000 00400000004000000040808000004000 4000000040000000000000004080b880 c000408000000040000040
expected 1a000000400000804000404000400080 c0400080000000000000404040804040 00000000400000404000000000000000 400080404000c0000000000000000000 00000000000000000000000000000000 0000003940408040404039b8808080

Vector 5 β€” Regression: alternating pattern, length 83 - heavily tied rotations, exercises the ascending-position tie-break

Source: Regression test - repetitive alternating input

Field Value
input 61626162616261626162616261626162 61626162616261626162616261626162 61626162616261626162616261626162 61626162616261626162616261626162 61626162616261626162616261626162 616261
expected 29000000626262626262626262626262 62626262626262626262626262626262 62626262626262626262626262616161 61616161616161616161616161616161 61616161616161616161616161616161 61616161616161

Vector 6 β€” Regression: period-4 input, length 64 - four classes of 16 identical rotations

Source: Regression test - fully tied rotation classes

Field Value
input 61626364616263646162636461626364 61626364616263646162636461626364 61626364616263646162636461626364 61626364616263646162636461626364
expected 00000000646464646464646464646464 64646464616161616161616161616161 61616161626262626262626262626262 62626262636363636363636363636363 63636363

← All algorithms