#!/usr/bin/env python3
"""
Heuristic static checks aligned with common 阿里巴巴手册 themes (not full P3C).
Outputs markdown to stdout or --out file.
"""
from __future__ import annotations

import argparse
import re
import sys
from pathlib import Path

TAB = re.compile(r"^\t+")
TRAIL = re.compile(r"[ \t]+$")
LONG_LINE = re.compile(r".{121,}")
SYSTEM_OUT = re.compile(r"\bSystem\.(out|err)\.(print|println)\s*\(")
EMPTY_CATCH = re.compile(r"catch\s*\([^)]+\)\s*\{\s*\}", re.MULTILINE)
CATCH_EXCEPTION_PASS = re.compile(
    r"catch\s*\(\s*Exception\s+\w+\s*\)\s*\{\s*\}", re.MULTILINE
)
TODO_FXXX = re.compile(r"\b(FIXME|XXX)\b", re.IGNORECASE)


def scan_file(path: Path, rel: str) -> list[tuple[str, int, str]]:
    issues: list[tuple[str, int, str]] = []
    try:
        text = path.read_text(encoding="utf-8", errors="replace")
    except OSError as e:
        return [("ERROR", 0, f"{rel}: {e}")]
    lines = text.splitlines()
    for i, line in enumerate(lines, 1):
        if TAB.search(line):
            issues.append(("FORMAT", i, "行首含 TAB，建议仅用空格缩进（手册排版/可读）"))
        if TRAIL.search(line):
            issues.append(("FORMAT", i, "行尾多余空白"))
        if LONG_LINE.search(line):
            issues.append(("FORMAT", i, "行宽超过约 120 字符，建议换行（团队规约常见上限）"))
        if SYSTEM_OUT.search(line) and "/test/" not in str(path).replace("\\", "/"):
            issues.append(("STYLE", i, "使用 System.out/err，生产代码建议改用日志框架"))
        if TODO_FXXX.search(line):
            issues.append(("STYLE", i, "含 FIXME/XXX 标记，发版前应清理或跟踪"))
    body = "\n".join(lines)
    if EMPTY_CATCH.search(body):
        issues.append(("RISK", 0, "存在空 catch 块 {}，手册建议至少记录或处理异常"))
    if CATCH_EXCEPTION_PASS.search(body):
        issues.append(
            ("RISK", 0, "存在 catch (Exception) { } 空实现，建议细化异常类型或记录日志")
        )
    return issues


def main() -> int:
    ap = argparse.ArgumentParser()
    ap.add_argument(
        "--repo",
        type=Path,
        default=Path(__file__).resolve().parent.parent,
        help="Repo root containing maven-*",
    )
    ap.add_argument("--out", type=Path, default=None, help="Write markdown report")
    args = ap.parse_args()
    repo: Path = args.repo

    roots = sorted(p for p in repo.glob("maven-*") if p.is_dir())
    lines_out: list[str] = []
    lines_out.append("# 阿里巴巴手册相关：启发式静态扫描\n")
    lines_out.append(
        "> 非 P3C 全量规则；仅覆盖 **TAB/行尾空白/超长行/System.out/空 catch/FIXME** 等易自动化项。\n"
    )

    grand = 0
    for root in roots:
        name = root.name
        lines_out.append(f"\n## `{name}`\n")
        mod_issues = 0
        for java in sorted(root.rglob("*.java")):
            sp = str(java).replace("\\", "/")
            if "/target/" in sp:
                continue
            rel = java.relative_to(repo)
            found = scan_file(java, str(rel))
            if not found:
                continue
            mod_issues += len(found)
            lines_out.append(f"\n### `{rel}`\n")
            lines_out.append("| 级别 | 行 | 说明 |\n|------|----|------|\n")
            for sev, ln, msg in found[:30]:
                lines_out.append(f"| {sev} | {ln or '-'} | {msg} |\n")
            if len(found) > 30:
                lines_out.append(f"\n… 另有 {len(found) - 30} 条（同文件省略）\n")
        if mod_issues == 0:
            lines_out.append("\n（本模块未发现上述启发式命中项）\n")
        grand += mod_issues

    lines_out.append(f"\n---\n\n**启发式命中条数（含重复行规则）**：{grand}\n")
    text = "".join(lines_out)
    if args.out:
        args.out.parent.mkdir(parents=True, exist_ok=True)
        args.out.write_text(text, encoding="utf-8")
        print(args.out, file=sys.stderr)
    else:
        sys.stdout.write(text)
    return 0


if __name__ == "__main__":
    sys.exit(main())