# JOURNAL — eval phase

## 2026-06-15 — Implementation

Built evaluator, CLI, and tests in one go.

### evaluator.py

`evaluate(node)` walks the AST recursively:
- `Num` → return `node.value` directly (already int or float from lexer)
- `Unary('-', operand)` → negate result
- `BinOp(op, left, right)` → evaluate both sides, apply op

Division: uses Python `left / right` (always returns float). If `right == 0`, raises `EvalError("Division by zero")`.

D3 result type rule: after computing a float result in BinOp, if `result == int(result)` we cast to int. This ensures `4/2 → 2` (int) and `7/2 → 3.5` (float). Integer arithmetic returns int naturally.

### calc.py CLI

`main()` accepts exactly one argv argument (the expression string).
Catches `LexError`, `ParseError`, `EvalError` → prints `Error: <msg>` to stderr, exits 1.
On success prints `result` (which is already int or float with correct type per D3 rule).

### Test run output

```
$ python -m unittest -q
Ran 62 tests in 0.001s
OK
```

(45 prior tests from lex+parse phases, 17 new evaluator tests)

### CLI checks

```
$ python calc.py "2+3*4"    → 14
$ python calc.py "(2+3)*4"  → 20
$ python calc.py "7/2"      → 3.5
$ python calc.py "4/2"      → 2
$ python calc.py "1/0"      → Error: Division by zero (exit 1)
$ python calc.py "1 +"      → Error: Unexpected end of input (exit 1)
```