v0.2.0 — Multi-Language Ports

Full Changelog: v0.1.0...v0.2.0

v0.1.0 — First Stable Release ★

Production-ready

• Portable mutation harness
• Published files narrowed to dist + LICENSE + README
• CI: test, mutation, packaging (@arethetypeswrong/cli), benchmark
• 
  pm install lombokecc\

Test results

Suite	Result
test-rs (5 integration)	✅ PASS
test-encode-only (6 cases)	✅ PASS
test-gf-poly (exhaustive)	✅ PASS
test-capacity (1800 randomised)	✅ 1800/1800
test-overcapacity (1400 randomised)	✅ 0% miscorrection
mutate.sh (12 mutations)	✅ 12/12 expected

v0.0.7 — Mutation Tested

Changes from v0.0.6

• All test suites use \process.exit(1)\ on failure (real assertions)
• Mutation testing harness (\mutate.sh) — 12/12 sesuai ekspektasi
• \ est-gf-poly.ts\ rewritten with exhaustive field property checks
• \prepublishOnly\ script added

v0.0.6 — 3200+ Stress Tests + CI

Changes from v0.0.5

• 1,800 randomised within-capacity trials (0–8 errors) — 100% recovery
• 1,400 randomised overcapacity trials (9–64 errors) — 0% miscorrection
• Throughput benchmark
• GitHub Actions CI (Node 20.x + 22.x)
• fix(solver): rank-deficient system handling

v0.0.5 — Full Decode Pipeline Working ★

Milestone: encoder + decoder both verified

• fix: systemic RS message comparison corrected
• fix: codeword validation after correction (fail-closed)
• All encode → corrupt → decode round-trip tests PASS
• Corrects up to 8 byte errors per 255-byte block

v0.0.4 — Peterson-GZ Replaces BMA

Breaking changes from v0.0.3

• Berlekamp-Massey replaced by Peterson-GZ (Gaussian elimination)
• Forney formula replaced by direct error-value solver
• 296 diff lines — major decoder rewrite
• Error values correct, message extraction offset bug remains

v0.0.3 — BMA Root Cause Confirmed

Changes from v0.0.2

• Decoder internals refactored (111 diff lines)
• BMA L-update condition confirmed as root cause
• Decision: full algorithm replacement needed

v0.0.2 — BMA Bug Identified

Changes from v0.0.1

• Multiple Forney formula variants tested in decoder
• BMA L-computation bug identified: computes L=8 for single-error case
• Encoder remains stable

v0.0.1 — GF(256) + Encoder

First release

• GF(256) field arithmetic (exp/log tables, mul, div, inv, pow)
• Systematic RS(255, 239) encoder
• Decoder skeleton (BMA/Forney) — known incorrect, available via
  pm run test:decode\
• 
  pm run test:full\ runs encoder + GF tests only (all PASS)