agent-orchestrator-benchmark/calculators/builder-adversary-stateless/run-03/machine-docs/REVIEW-parse.md

REVIEW — Phase parse (Adversary)

Status

All gates D1–D6: PASS. Phase complete — no defects found.

Gate verdicts

D1 — Precedence: PASS @2026-06-15T04:08Z

Cold run: parse(tokenize('1+2*3')) → BinOp('+', Num(1), BinOp('*', Num(2), Num(3))) ✓
Matches expected. Also verified 2*3+1, 10-4/2, 1+2*3-4/2 all produce correct precedence trees.

D2 — Left Associativity: PASS @2026-06-15T04:08Z

Cold run:

• parse(tokenize('8-3-2')) → BinOp('-', BinOp('-', Num(8), Num(3)), Num(2)) ✓
• parse(tokenize('8/4/2')) → BinOp('/', BinOp('/', Num(8), Num(4)), Num(2)) ✓
  Also verified 1+2+3 and 2*3*4 fold left correctly.

D3 — Parentheses: PASS @2026-06-15T04:08Z

Cold run: parse(tokenize('(1+2)*3')) → BinOp('*', BinOp('+', Num(1), Num(2)), Num(3)) ✓
Also verified nested parens ((2+3)), 8-(3-2), and (42).

D4 — Unary Minus: PASS @2026-06-15T04:08Z

Cold run:

• parse(tokenize('-5')) → Unary('-', Num(5)) ✓
• parse(tokenize('-(1+2)')) → Unary('-', BinOp('+', Num(1), Num(2))) ✓
• parse(tokenize('3 * -2')) → BinOp('*', Num(3), Unary('-', Num(2))) ✓
  Also verified --5 → Unary('-', Unary('-', Num(5))) and 1 + -(2 * 3).

D5 — Errors: PASS @2026-06-15T04:08Z

All five plan-specified malformed inputs raise ParseError (not any other exception):

• "1 +" → ParseError: unexpected token 'EOF' ✓
• "(1" → ParseError: expected RPAREN, got 'EOF' ✓
• "1 2" → ParseError: unexpected token 'NUMBER' ✓
• ")(" → ParseError: unexpected token 'RPAREN' ✓
• "" → ParseError: empty input ✓
  Extra adversarial cases also raise ParseError: *5, /5, (), 1++2, * alone.

D6 — Tests Green: PASS @2026-06-15T04:08Z

python -m unittest -q → Ran 44 tests in 0.001s OK ✓
23 parser tests (TestPrecedence×4, TestLeftAssociativity×4, TestParentheses×4, TestUnaryMinus×5, TestErrors×6) + 21 lexer tests.
All assertions are structural (equality via __eq__ on Num/BinOp/Unary nodes), not evaluation.

Adversary findings

None. No defects found.