#!/usr/bin/env python3
"""数据一致性校验脚本。

校验 notebooks/pilot_analysis.md 中的关键数据与 plots/*.csv 数据文件一致，
并验证论文 reports/paper_draft.md 中引用的核心数值与数据源吻合。

用法：
    python3 processing/validate_data.py
"""

import csv
import os
import re
import sys

ROOT = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
PLOTS = os.path.join(ROOT, "plots")
NOTEBOOK = os.path.join(ROOT, "notebooks", "pilot_analysis.md")
PAPER = os.path.join(ROOT, "reports", "paper_draft.md")

errors = []
checks = 0


def ok(msg):
    global checks
    checks += 1
    print(f"  PASS: {msg}")


def fail(msg):
    global errors
    errors.append(msg)
    print(f"  FAIL: {msg}")


def load_csv(name):
    path = os.path.join(PLOTS, name)
    with open(path, encoding="utf-8") as f:
        return list(csv.DictReader(f))


def check_run_data():
    print("[1] 校验运行数据 CSV")
    rows = load_csv("pilot_run_data.csv")
    cities = [r["city"] for r in rows]
    if len(cities) == 8:
        ok(f"地市数 = 8: {', '.join(cities)}")
    else:
        fail(f"地市数 = {len(cities)}, 期望 8")

    producing = [r for r in rows if r["stage"] == "实际产出"]
    if len(producing) == 2:
        ok(f"实际产出地市 = 2: {', '.join(r['city'] for r in producing)}")
    else:
        fail(f"实际产出地市 = {len(producing)}, 期望 2")

    huizhou = next(r for r in rows if r["city"] == "惠州")
    if int(huizhou["alert_count"]) == 19800:
        ok("惠州预警数据 = 19,800")
    else:
        fail(f"惠州预警数据 = {huizhou['alert_count']}, 期望 19800")

    foshan = next(r for r in rows if r["city"] == "佛山")
    if int(foshan["alert_count"]) == 3923:
        ok("佛山预警数据 = 3,923")
    else:
        fail(f"佛山预警数据 = {foshan['alert_count']}, 期望 3923")


def check_vendor_effort():
    print("[2] 校验供应商人天投入 CSV")
    rows = load_csv("vendor_effort.csv")
    total = sum(int(r["person_days"]) for r in rows)
    if total == 100:
        ok(f"人天合计 = {total}")
    else:
        fail(f"人天合计 = {total}, 期望 100")

    expected = {
        "需求沟通与前期对接": 15,
        "现场培训与推广": 20,
        "技术接入与配置": 18,
        "问题响应与日常运营": 20,
        "个性化适配与优化": 12,
        "差旅在途": 10,
        "文档编写": 5,
    }
    for row in rows:
        cat = row["category"]
        days = int(row["person_days"])
        if cat in expected and days == expected[cat]:
            ok(f"{cat} = {days} 人天")
        else:
            fail(f"{cat} = {days}, 期望 {expected.get(cat, '?')}")


def check_access_cost():
    print("[3] 校验接入耗时 CSV")
    rows = load_csv("access_cost.csv")
    steps = {r["step"]: r for r in rows}

    ideal_sum = sum(int(steps[s]["ideal_days"]) for s in
                    ["账号开通", "权限配置", "组织数据匹配", "应用入口配置", "网络策略审批"])
    if ideal_sum == 10:
        ok(f"理想端到端 = {ideal_sum} 工作日")
    else:
        fail(f"理想端到端 = {ideal_sum}, 期望 10")

    actual_sum = sum(int(steps[s]["actual_days"]) for s in
                     ["账号开通", "权限配置", "组织数据匹配", "应用入口配置", "网络策略审批"])
    if actual_sum == 18:
        ok(f"各环节实际合计 = {actual_sum} 工作日（端到端实际约 20，含排队余量）")
    else:
        fail(f"各环节实际合计 = {actual_sum}, 期望 18")


def check_paper_numbers():
    print("[4] 校验论文核心数值")
    with open(PAPER, encoding="utf-8") as f:
        text = f.read()

    patterns = [
        (r"144[,]?853", "累计命中异常数据 144,853"),
        (r"476\s*次", "累计预警推送 476 次"),
        (r"10\s*条.*规则", "监测规则 10 条"),
        (r"19[,]?800", "惠州预警 19,800"),
        (r"3[,]?923", "佛山预警 3,923"),
        (r"8\s*个", "推广地市 8 个"),
        (r"7\s*场.*培训", "推广培训 7 场"),
        (r"100\s*人天", "供应商投入 100 人天"),
        (r"约\s*4\s*周", "端到端实际约 4 周"),
        (r"约\s*30\s*%", "标准化降本约 30%"),
        (r"表 6", "表 6 地市间对比"),
        (r"表 7", "表 7 实施路线图"),
        (r"表 8", "表 8 风险缓解矩阵"),
        (r"5\.0\s*倍", "惠州/佛山产出量差 5.0 倍"),
        (r"三阶段", "三阶段路线图"),
        (r"阶段门", "阶段门审核"),
    ]
    for pat, desc in patterns:
        if re.search(pat, text):
            ok(f"论文含 {desc}")
        else:
            fail(f"论文未找到 {desc}")


def check_plots_exist():
    print("[5] 校验图表文件存在")
    for name in ["pilot_run_data.png", "vendor_effort.png", "access_cost.png",
                 "city_comparison.png", "city_comparison.csv", "roadmap_data.csv"]:
        path = os.path.join(PLOTS, name)
        if os.path.isfile(path) and os.path.getsize(path) > 0:
            ok(f"{name} 存在 ({os.path.getsize(path)} bytes)")
        else:
            fail(f"{name} 缺失或为空")


def check_v4_content():
    print("[6] 校验 v4.0 新增内容")
    with open(PAPER, encoding="utf-8") as f:
        text = f.read()

    checks_v4 = [
        (r"3\.7\s*地市间对比", "3.7 地市间对比分析"),
        (r"5\.4\s*实施路线图", "5.4 实施路线图与风险缓解"),
        (r"稳定运行梯队", "稳定运行梯队"),
        (r"实际产出梯队", "实际产出梯队"),
        (r"推广接入梯队", "推广接入梯队"),
        (r"标准化筑基", "第一阶段：标准化筑基"),
        (r"评价闭环", "第二阶段：评价闭环"),
        (r"扩围深化", "第三阶段：扩围深化"),
    ]
    for pat, desc in checks_v4:
        if re.search(pat, text):
            ok(f"论文含 {desc}")
        else:
            fail(f"论文未找到 {desc}")

    nb_path = os.path.join(ROOT, "notebooks", "pilot_analysis.md")
    with open(nb_path, encoding="utf-8") as f:
        nb = f.read()
    for section in ["## 9. 地市间对比分析", "## 10. 实施路线图与风险缓解"]:
        if section in nb:
            ok(f"notebooks 含 {section}")
        else:
            fail(f"notebooks 缺少 {section}")


def main():
    print("=" * 60)
    print("数据一致性校验 — AI时代市场化运作机制研究")
    print("=" * 60)
    print()

    check_run_data()
    print()
    check_vendor_effort()
    print()
    check_access_cost()
    print()
    check_paper_numbers()
    print()
    check_plots_exist()
    print()
    check_v4_content()

    print()
    print("=" * 60)
    if errors:
        print(f"校验完成：{checks} 通过，{len(errors)} 失败")
        for e in errors:
            print(f"  - {e}")
        sys.exit(1)
    else:
        print(f"校验完成：全部 {checks} 项通过，0 失败")
        sys.exit(0)


if __name__ == "__main__":
    main()