from __future__ import annotations

import json
from dataclasses import dataclass
from datetime import datetime, timezone
from pathlib import Path
from typing import Dict, List


@dataclass(frozen=True)
class Signal:
    name: str
    weight: int


SIGNALS: List[Signal] = [
    Signal("has_tests", 10),
    Signal("has_validation_script", 10),
    Signal("has_greedy_baseline", 10),
    Signal("has_training_pipeline", 15),
    Signal("has_submission_bundle_script", 15),
    Signal("has_judge_brief", 15),
    Signal("has_replay_endpoint_docs", 10),
    Signal("has_judge_pack_endpoint_docs", 10),
    Signal("has_tuned_metrics_artifact", 15),
]

PROJECT_DIR_CANDIDATES = [
    "gdpr_eraseops",
    "incident-commander",
    "llmfleet-sre-main",
    "quantam_main",
    "finops_arb_env",
]


def _read_text_if_exists(path: Path) -> str:
    if not path.exists():
        return ""
    return path.read_text(encoding="utf-8", errors="ignore")


def _detect_signals(project_root: Path) -> Dict[str, bool]:
    readme = _read_text_if_exists(project_root / "README.md")
    scripts_dir = project_root / "scripts"
    artifacts_dir = project_root / "artifacts"

    return {
        "has_tests": (project_root / "tests").exists(),
        "has_validation_script": (scripts_dir / "validate_project.py").exists()
        or (scripts_dir / "validate-submission.sh").exists()
        or (scripts_dir / "validate-submission.ps1").exists(),
        "has_greedy_baseline": (project_root / "greedy_baseline.py").exists(),
        "has_training_pipeline": (scripts_dir / "train_sb3_agent.py").exists(),
        "has_submission_bundle_script": (scripts_dir / "generate_submission_bundle.py").exists(),
        "has_judge_brief": (project_root / "JUDGE_BRIEF.md").exists(),
        "has_replay_endpoint_docs": "/replay" in readme,
        "has_judge_pack_endpoint_docs": "/judge_pack" in readme,
        "has_tuned_metrics_artifact": (artifacts_dir / "ppo_tuned_v3" / "evaluation_summary.json").exists(),
    }


def _score(signals: Dict[str, bool]) -> int:
    total = 0
    for signal in SIGNALS:
        if signals.get(signal.name, False):
            total += signal.weight
    return total


def _project_kind(project_name: str) -> str:
    if project_name == "gdpr_eraseops":
        return "target"
    return "peer"


def main() -> None:
    script_path = Path(__file__).resolve()
    gdpr_root = script_path.parents[1]
    workspace_root = gdpr_root.parent

    results = []
    for candidate in PROJECT_DIR_CANDIDATES:
        project_root = workspace_root / candidate
        if not project_root.exists() or not project_root.is_dir():
            continue

        signals = _detect_signals(project_root)
        score = _score(signals)
        results.append(
            {
                "project": candidate,
                "kind": _project_kind(candidate),
                "score": score,
                "signals": signals,
            }
        )

    results.sort(key=lambda item: item["score"], reverse=True)

    output_dir = gdpr_root / "artifacts" / "readiness"
    output_dir.mkdir(parents=True, exist_ok=True)

    report = {
        "generated_at": datetime.now(timezone.utc).isoformat(),
        "rubric": [{"name": signal.name, "weight": signal.weight} for signal in SIGNALS],
        "results": results,
    }

    json_path = output_dir / "cohort_comparison.json"
    json_path.write_text(json.dumps(report, indent=2), encoding="utf-8")

    md_lines = [
        "# Cohort Readiness Comparison",
        "",
        f"Generated at: {report['generated_at']}",
        "",
        "## Leaderboard",
        "",
        "| Rank | Project | Score |",
        "|---:|---|---:|",
    ]

    for idx, row in enumerate(results, start=1):
        md_lines.append(f"| {idx} | {row['project']} | {row['score']} |")

    md_lines.extend(
        [
            "",
            "## Signals",
            "",
            "Legend: Y = signal present, N = signal absent",
            "",
        ]
    )

    header = "| Project | " + " | ".join(signal.name for signal in SIGNALS) + " |"
    separator = "|---|" + "|".join(["---"] * len(SIGNALS)) + "|"
    md_lines.append(header)
    md_lines.append(separator)

    for row in results:
        cells = ["Y" if row["signals"].get(signal.name, False) else "N" for signal in SIGNALS]
        md_lines.append("| " + row["project"] + " | " + " | ".join(cells) + " |")

    md_lines.extend(
        [
            "",
            "## Recommendation",
            "",
            "Use this report alongside JUDGE_BRIEF.md and submission_report.md during review.",
        ]
    )

    md_path = output_dir / "cohort_comparison.md"
    md_path.write_text("\n".join(md_lines) + "\n", encoding="utf-8")

    print(json.dumps({"json": str(json_path), "markdown": str(md_path), "leader": results[0] if results else None}, indent=2))


if __name__ == "__main__":
    main()