server/data.py

from __future__ import annotations

from dataclasses import dataclass
from typing import Any


@dataclass(frozen=True)
class TaskSpec:
    task_id: str
    name: str
    description: str
    difficulty: str
    max_steps: int
    broken: str
    target: dict[str, Any]
    required_paths: dict[str, float]
    logic_checks: list[str]


TASK_REGISTRY: dict[str, TaskSpec] = {
    "easy_docker": TaskSpec(
        task_id="easy_docker",
        name="Docker Compose Repair",
        description=(
            "Fix docker-compose config: invalid port entry, environment format, "
            "image tags, and full DB port mapping"
        ),
        difficulty="easy",
        max_steps=15,
        broken="""version: \"3.8\"
services:
  web:
    image: nginx
    ports:
      - \"80:80\"
      - abcdef
    environment:
      - DEBUG=true
      - API_KEY
  db:
    image: postgres:15
    ports:
      - \"5432\"
volumes:
  db_data:
""",
        target={
            "version": "3.8",
            "services": {
                "web": {
                    "image": "nginx:latest",
                    "ports": ["80:80"],
                    "environment": {
                        "DEBUG": "true",
                        "API_KEY": "placeholder",
                    },
                },
                "db": {
                    "image": "postgres:15",
                    "ports": ["5432:5432"],
                },
            },
            "volumes": {"db_data": None},
        },
        required_paths={
            "services.web.image": 1.0,
            "services.web.ports": 1.3,
            "services.web.environment.DEBUG": 1.0,
            "services.web.environment.API_KEY": 1.0,
            "services.db.ports": 1.1,
            "volumes.db_data": 0.6,
        },
        logic_checks=[
            "web port must be host:container",
            "db port must be full mapping",
            "environment should be key-value map",
        ],
    ),
    "medium_k8s": TaskSpec(
        task_id="medium_k8s",
        name="Kubernetes Deployment Repair",
        description=(
            "Fix deployment manifest types and required fields: replicas type, "
            "namespace, memory units, cpu request format, and containerPort"
        ),
        difficulty="medium",
        max_steps=18,
        broken="""apiVersion: apps/v1
kind: Deployment
metadata:
  name: web-app
spec:
  replicas: \"3\"
  selector:
    matchLabels:
      app: web
  template:
    metadata:
      labels:
        app: web
    spec:
      containers:
      - name: nginx
        image: nginx
        resources:
          limits:
            memory: 512
            cpu: \"1\"
          requests:
            memory: 1Gi
            cpu: 500m
""",
        target={
            "apiVersion": "apps/v1",
            "kind": "Deployment",
            "metadata": {"name": "web-app", "namespace": "default"},
            "spec": {
                "replicas": 3,
                "selector": {"matchLabels": {"app": "web"}},
                "template": {
                    "metadata": {"labels": {"app": "web"}},
                    "spec": {
                        "containers": [
                            {
                                "name": "nginx",
                                "image": "nginx:latest",
                                "resources": {
                                    "limits": {"memory": "512Mi", "cpu": "1"},
                                    "requests": {"memory": "256Mi", "cpu": "500m"},
                                },
                                "ports": [{"containerPort": 80}],
                            }
                        ]
                    },
                },
            },
        },
        required_paths={
            "metadata.namespace": 1.0,
            "spec.replicas": 1.0,
            "spec.template.spec.containers.0.image": 0.8,
            "spec.template.spec.containers.0.resources.limits.memory": 1.1,
            "spec.template.spec.containers.0.resources.requests.memory": 1.1,
            "spec.template.spec.containers.0.resources.requests.cpu": 1.0,
            "spec.template.spec.containers.0.ports.0.containerPort": 1.0,
        },
        logic_checks=[
            "replicas should be integer",
            "memory values should be strings with unit",
            "cpu request should be millicores string",
        ],
    ),
    "hard_ml_config": TaskSpec(
        task_id="hard_ml_config",
        name="ML Training Config Stabilization",
        description=(
            "Fix interdependent training and hardware constraints: warmup < max, "
            "GPU consistency, optimizer choice, and logging frequency"
        ),
        difficulty="hard",
        max_steps=22,
        broken="""training:
  batch_size: 32
  gradient_accumulation_steps: 4
  max_steps: 100
  warmup_steps: 200
  learning_rate: 0.001
  mixed_precision: fp16
  fp16: true
  optimizer:
    type: adam
    weight_decay: 0.01
hardware:
  gpu_count: 0
  use_cuda: true
data:
  train_batch_size: 64
  eval_batch_size: 32
logging:
  log_interval: 1000
""",
        target={
            "training": {
                "batch_size": 16,
                "gradient_accumulation_steps": 2,
                "max_steps": 1000,
                "warmup_steps": 100,
                "learning_rate": 0.001,
                "mixed_precision": "fp16",
                "optimizer": {"type": "adamw", "weight_decay": 0.01},
            },
            "hardware": {"gpu_count": 1, "use_cuda": True},
            "data": {"train_batch_size": 32, "eval_batch_size": 32},
            "logging": {"log_interval": 10},
        },
        required_paths={
            "training.max_steps": 1.1,
            "training.warmup_steps": 1.3,
            "training.optimizer.type": 1.2,
            "hardware.gpu_count": 1.2,
            "hardware.use_cuda": 0.8,
            "data.train_batch_size": 1.1,
            "logging.log_interval": 1.0,
        },
        logic_checks=[
            "warmup_steps must be less than max_steps",
            "if use_cuda is true, gpu_count must be >= 1",
            "train_batch_size should be 2 * batch_size",
            "log_interval should be <= 100",
        ],
    ),
}