data/processed/RQ1/BookWriter-MCP/evaluate_success.py

#!/usr/bin/env python3
"""
Compute the success rate of each model on the BookWriter-MCP task.

Success criteria:
1. A chapters folder exists
2. The chapters folder contains at least one .md file
3. The session root contains a main book .md file
"""

import json
import os
from pathlib import Path
from collections import defaultdict

# Base path and model list
BASE_DIR = Path("/Users/wzr/TOSEM-2025/RESULTS")
MODELS = [
    "DeepSeek-R1",
    "DeepSeek-V3-1",
    "GPT-4o-mini",
    "GPT-5",
    "Gemini-2.5-flash",
    "Gemini-2.5-flash-nothinking",
    "Qwen3-235b",
]
PROJECT_NAME = "BookWriter-MCP"


def check_session_success(session_dir: Path) -> dict:
    """
    Check whether a single session is successful.

    Returns:
        {
            "success": bool,
            "has_chapters_folder": bool,
            "has_main_md": bool,
            "chapter_count": int,
            "chapter_files": list,
            "main_md_file": str or None,
        }
    """
    result = {
        "success": False,
        "has_chapters_folder": False,
        "has_main_md": False,
        "chapter_count": 0,
        "chapter_files": [],
        "main_md_file": None,
    }

    # Check chapters folder
    chapters_dir = session_dir / "chapters"
    if chapters_dir.exists() and chapters_dir.is_dir():
        result["has_chapters_folder"] = True

        # Count .md files in chapters
        chapter_files = sorted(chapters_dir.glob("*.md"))
        result["chapter_count"] = len(chapter_files)
        result["chapter_files"] = [f.name for f in chapter_files]

    # Check main book .md file in session root (exclude execution_path.md)
    md_files = [f for f in session_dir.glob("*.md") if f.name != "execution_path.md"]
    if md_files:
        result["has_main_md"] = True
        result["main_md_file"] = md_files[0].name

    # Success criteria: chapters folder exists AND at least one chapter file
    result["success"] = result["has_chapters_folder"] and result["chapter_count"] > 0

    return result


def analyze_model_results():
    """Analyze execution results for all models."""
    results = {}

    for model in MODELS:
        test_results_dir = BASE_DIR / model / PROJECT_NAME / "test_results"

        if not test_results_dir.exists():
            print(
                f"⚠️  test_results directory not found for model {model}: {test_results_dir}"
            )
            continue

        model_stats = {
            "success_count": 0,
            "failure_count": 0,
            "total_count": 0,
            "avg_chapters": 0,
            "sessions": [],
        }

        total_chapters = 0

        # Iterate over all session subdirectories
        for session_dir in sorted(test_results_dir.iterdir()):
            if not session_dir.is_dir():
                continue

            # Check success status
            check_result = check_session_success(session_dir)

            model_stats["total_count"] += 1
            if check_result["success"]:
                model_stats["success_count"] += 1
                total_chapters += check_result["chapter_count"]
            else:
                model_stats["failure_count"] += 1

            model_stats["sessions"].append(
                {
                    "session": session_dir.name,
                    "success": check_result["success"],
                    "has_chapters_folder": check_result["has_chapters_folder"],
                    "has_main_md": check_result["has_main_md"],
                    "chapter_count": check_result["chapter_count"],
                    "chapter_files": check_result["chapter_files"],
                    "main_md_file": check_result["main_md_file"],
                }
            )

        # Compute average chapter count among successful sessions
        if model_stats["success_count"] > 0:
            model_stats["avg_chapters"] = total_chapters / model_stats["success_count"]

        results[model] = model_stats

    return results


def print_summary(results):
    """Print summary statistics."""
    print("\n" + "=" * 80)
    print(f"Model execution summary - {PROJECT_NAME}")
    print("=" * 80 + "\n")
    print(
        "Success criteria: chapters folder exists AND at least one chapter .md file\n"
    )

    for model, stats in results.items():
        total = stats["total_count"]
        success = stats["success_count"]
        failure = stats["failure_count"]
        avg_chapters = stats["avg_chapters"]

        if total > 0:
            success_rate = (success / total) * 100
            print(f"📊 {model}")
            print(f"   Total: {total}")
            print(f"   Success: {success} ({success_rate:.1f}%)")
            print(f"   Failure: {failure} ({100-success_rate:.1f}%)")
            if success > 0:
                print(f"   Avg chapters (success only): {avg_chapters:.1f}")
            print()
        else:
            print(f"📊 {model}")
            print("   No data")
            print()

    print("=" * 80)


def print_failure_details(results):
    """Print details for failed sessions."""
    print("\n" + "=" * 80)
    print("Failure details")
    print("=" * 80 + "\n")

    for model, stats in results.items():
        failures = [s for s in stats["sessions"] if not s["success"]]

        if failures:
            print(f"📊 {model} - {len(failures)} failed sessions:")
            for session in failures[:5]:  # show only the first 5
                print(f"\n   🔴 {session['session']}")
                print(
                    f"      - chapters folder: {'✅' if session['has_chapters_folder'] else '❌'}"
                )
                print(
                    f"      - main .md file: {'✅' if session['has_main_md'] else '❌'} {session['main_md_file'] or ''}"
                )
                print(f"      - chapter count: {session['chapter_count']}")
                if session["chapter_files"]:
                    print(
                        f"      - chapter files: {', '.join(session['chapter_files'][:3])}"
                    )

            if len(failures) > 5:
                print(f"\n   ... and {len(failures) - 5} more failures")
            print()


def save_detailed_results(results, output_file="success_detailed_results.json"):
    """Save detailed results to a JSON file."""
    output_path = Path(__file__).parent / output_file
    with open(output_path, "w", encoding="utf-8") as f:
        json.dump(results, f, indent=2, ensure_ascii=False)
    print(f"\n✅ Detailed results saved to: {output_path}")


def save_csv_summary(results, output_file="success_rate.csv"):
    """Save summary to a CSV file."""
    import csv

    output_path = Path(__file__).parent / output_file
    with open(output_path, "w", newline="", encoding="utf-8") as f:
        writer = csv.writer(f)
        writer.writerow(
            ["Model", "Total", "Success", "Failure", "Success_Rate(%)", "Avg_Chapters"]
        )

        for model, stats in results.items():
            total = stats["total_count"]
            success = stats["success_count"]
            failure = stats["failure_count"]
            success_rate = (success / total * 100) if total > 0 else 0
            avg_chapters = stats["avg_chapters"]

            writer.writerow(
                [
                    model,
                    total,
                    success,
                    failure,
                    f"{success_rate:.2f}",
                    f"{avg_chapters:.2f}",
                ]
            )

    print(f"✅ CSV summary saved to: {output_path}")


if __name__ == "__main__":
    print(f"Starting success-rate evaluation - {PROJECT_NAME}")
    print(
        "Success criteria: chapters folder exists AND at least one chapter .md file\n"
    )

    results = analyze_model_results()
    print_summary(results)
    print_failure_details(results)
    save_detailed_results(results)
    save_csv_summary(results)

    print("\n✅ Done")