fix: align engine models with task spec

engine/models.py

@@ -1,115 +1,120 @@
+from __future__ import annotations
+
 from dataclasses import dataclass, field
 from pathlib import Path
-from typing import Literal
+from typing import Any
 
 
 @dataclass(frozen=True)
 class ArtifactSpec:
-    include: list[str] = field(default_factory=list)
-    exclude: list[str] = field(default_factory=list)
-    max_files_per_iteration: int = 0
+    include: list[str]
+    exclude: list[str]
+    max_files_per_iteration: int
 
 
 @dataclass(frozen=True)
 class MutationSpec:
-    mode: str = ""
-    allowed_file_types: list[str] = field(default_factory=list)
-    max_changed_lines: int = 0
+    mode: str
+    allowed_file_types: list[str]
+    max_changed_lines: int
 
 
 @dataclass(frozen=True)
 class RunnerSpec:
-    command: str = ""
-    cwd: Path = Path(".")
-    timeout_seconds: int = 0
+    command: str
+    cwd: str
+    timeout_seconds: int
 
 
 @dataclass(frozen=True)
 class ScorerParseSpec:
-    format: str = ""
-    score_field: str = ""
-    metrics_field: str = ""
+    format: str
+    score_field: str
+    metrics_field: str
 
 
 @dataclass(frozen=True)
 class ScorerSpec:
-    type: str = ""
-    command: str = ""
-    parse: ScorerParseSpec = field(default_factory=ScorerParseSpec)
+    type: str
+    command: str
+    parse: ScorerParseSpec
 
 
 @dataclass(frozen=True)
 class ObjectiveSpec:
-    primary_metric: str = ""
-    direction: Literal["maximize", "minimize"] = "maximize"
+    primary_metric: str
+    direction: str
 
 
 @dataclass(frozen=True)
 class ConstraintSpec:
-    metric: str = ""
-    op: str = ""
-    value: int | float = 0
+    metric: str
+    op: str
+    value: Any
 
 
 @dataclass(frozen=True)
 class PolicySpec:
-    keep_if: str = ""
-    tie_breakers: list[str] = field(default_factory=list)
-    on_failure: str = ""
+    keep_if: str
+    tie_breakers: list[dict[str, str]]
+    on_failure: str
 
 
 @dataclass(frozen=True)
 class BudgetSpec:
-    max_iterations: int = 0
-    max_failures: int = 0
+    max_iterations: int
+    max_failures: int
 
 
 @dataclass(frozen=True)
 class LoggingSpec:
-    results_file: Path = Path()
-    candidate_dir: Path = Path()
+    results_file: str
+    candidate_dir: str
 
 
 @dataclass(frozen=True)
 class TaskSpec:
-    id: str = ""
-    description: str = ""
-    root_dir: Path = Path(".")
-    artifacts: ArtifactSpec = field(default_factory=ArtifactSpec)
-    mutation: MutationSpec = field(default_factory=MutationSpec)
-    runner: RunnerSpec = field(default_factory=RunnerSpec)
-    scorer: ScorerSpec = field(default_factory=ScorerSpec)
-    objective: ObjectiveSpec = field(default_factory=ObjectiveSpec)
-    constraints: list[ConstraintSpec] = field(default_factory=list)
-    policy: PolicySpec = field(default_factory=PolicySpec)
-    budget: BudgetSpec = field(default_factory=BudgetSpec)
-    logging: LoggingSpec = field(default_factory=LoggingSpec)
+    id: str
+    description: str
+    artifacts: ArtifactSpec
+    mutation: MutationSpec
+    runner: RunnerSpec
+    scorer: ScorerSpec
+    objective: ObjectiveSpec
+    constraints: list[ConstraintSpec]
+    policy: PolicySpec
+    budget: BudgetSpec
+    logging: LoggingSpec
+    root_dir: Path
 
 
 @dataclass(frozen=True)
 class BaselineSnapshot:
-    root_dir: Path = Path(".")
-    files: dict[str, str] = field(default_factory=dict)
-
-
-@dataclass(frozen=True)
-class ScoreResult:
-    score: float = 0.0
-    metrics: dict[str, float | int] = field(default_factory=dict)
-
-
-@dataclass(frozen=True)
-class DecisionResult:
-    keep: bool = False
-    reason: str = ""
-    primary_metric: str = ""
-    primary_value: float = 0.0
+    file_contents: dict[Path, str]
+    file_hashes: dict[Path, str]
 
 
 @dataclass(frozen=True)
 class RunResult:
-    task_id: str = ""
-    iteration: int = 0
-    score: ScoreResult = field(default_factory=ScoreResult)
-    decision: DecisionResult = field(default_factory=DecisionResult)
-    baseline: BaselineSnapshot = field(default_factory=BaselineSnapshot)
+    command: str
+    cwd: Path
+    exit_code: int
+    runtime_seconds: float
+    stdout: str
+    stderr: str
+
+
+@dataclass(frozen=True)
+class ScoreResult:
+    primary_score: float
+    metrics: dict[str, Any]
+    raw_output: dict[str, Any]
+
+
+@dataclass(frozen=True)
+class DecisionResult:
+    status: str
+    reason: str
+    baseline_score: float | None
+    candidate_score: float | None
+    constraint_failures: list[str] = field(default_factory=list)