2026 年商城系统选型：别再问"能不能用"，该问"能不能掌控"

"""
商城系统架构耦合度快速扫描器
用法: python coupling_scan.py /path/to/platform/source
输出: 各核心模块之间的跨目录引用计数，数值越高耦合越严重
"""

import os
import re
import sys
from collections import defaultdict
from pathlib import Path

# 定义要扫描的核心模块目录名（按常见商城系统结构调整）
CORE_MODULES = [
    "order", "payment", "product", "inventory",
    "user", "cart", "promotion", "shipping",
    "catalog", "checkout", "customer", "marketing",
]

# 匹配跨目录引用的正则（覆盖 import/require/use/include 等常见模式)
IMPORT_PATTERNS = [
    re.compile(r'(?:import|from)\s+[\w.]*\.(\w+)', re.IGNORECASE),       # Python
    re.compile(r'use\s+[\w\\]*\\(\w+)', re.IGNORECASE),                   # PHP
    re.compile(r'import\s+[\w.]*\.(\w+)', re.IGNORECASE),                 # Java/Kotlin
    re.compile(r'require\s+[\'"][\w./]*?/(\w+)', re.IGNORECASE),          # Node
]


def detect_module(filepath: Path) -> str | None:
    """从文件路径中提取它属于哪个核心模块"""
    parts = filepath.parts
    for part in parts:
        low = part.lower()
        for mod in CORE_MODULES:
            if low.startswith(mod) or low.endswith(mod):
                return mod
    return None


def scan_cross_refs(root: str) -> dict:
    """扫描所有源文件，统计跨模块引用"""
    root_path = Path(root)
    coupling = defaultdict(lambda: defaultdict(int))

    source_exts = {".py", ".php", ".java", ".kt", ".js", ".ts", ".jsx", ".tsx"}

    for filepath in root_path.rglob("*"):
        if filepath.suffix not in source_exts:
            continue
        if any(skip in filepath.parts for skip in ["vendor", "node_modules", ".git", "test", "tests"]):
            continue

        src_module = detect_module(filepath)
        if not src_module:
            continue

        try:
            content = filepath.read_text(errors="ignore")
        except Exception:
            continue

        for pattern in IMPORT_PATTERNS:
            for match in pattern.finditer(content):
                ref_name = match.group(1).lower()
                for mod in CORE_MODULES:
                    if ref_name.startswith(mod) and mod != src_module:
                        coupling[src_module][mod] += 1

    return coupling


def print_report(coupling: dict):
    """打印耦合度报告"""
    print("\n=== 模块耦合度扫描报告 ===\n")
    total_cross = 0
    hot_pairs = []

    for src, refs in sorted(coupling.items()):
        for ref, count in sorted(refs.items(), key=lambda x: -x[1]):
            total_cross += count
            hot_pairs.append((count, src, ref))
            print(f"  {src} → {ref}: {count} 次引用")

    print(f"\n跨模块引用总数: {total_cross}")
    print("\n最危险的耦合对（改一处牵动最多）:")
    for count, src, ref in sorted(hot_pairs, reverse=True)[:5]:
        print(f"  ⚠ {src} ↔ {ref}: {count} 次")

    if total_cross > 200:
        print("\n🔴 耦合度偏高——二次开发改动面大，升级风险高")
    elif total_cross > 80:
        print("\n🟡 耦合度中等——核心模块有一定独立性，但需关注热点对")
    else:
        print("\n🟢 耦合度较低——模块边界清晰，二次开发友好度高")


if __name__ == "__main__":
    if len(sys.argv) < 2:
        print("用法: python coupling_scan.py <项目源码目录>")
        sys.exit(1)

    root_dir = sys.argv[1]
    if not os.path.isdir(root_dir):
        print(f"目录不存在: {root_dir}")
        sys.exit(1)

    result = scan_cross_refs(root_dir)
    print_report(result)