# -*- coding: utf-8 -*-
r'''量一下仓库体积，找出推送 413 的原因。'''
from __future__ import annotations

import subprocess
import sys
from pathlib import Path

ROOT = Path(r"C:\Users\shu\Desktop\deepseekagent\LanRemote")
GIT = r"C:\Program Files\Git\bin\git.exe"


def git(*args: str) -> str:
    r = subprocess.run([GIT] + list(args), cwd=str(ROOT),
                       capture_output=True, text=True,
                       encoding="utf-8", errors="replace")
    return r.stdout or ""


def main() -> int:
    print("  一、仓库对象统计")
    print("    " + git("count-objects", "-vH").replace("\n", "\n    ").strip())

    print()
    print("  二、各提交的树体积（找有没有混进大文件）")
    # 列出所有 blob 及大小
    out = git("rev-list", "--objects", "--all")
    lines = [ln for ln in out.splitlines() if ln.strip()]
    print("    对象条目 %d 个" % len(lines))

    # 批量查类型和大小
    payload = "\n".join(ln.split(" ", 1)[0] for ln in lines)
    r = subprocess.run([GIT, "cat-file", "--batch-check=%(objecttype) %(objectsize) %(objectname)"],
                       cwd=str(ROOT), input=payload, capture_output=True,
                       text=True, encoding="utf-8", errors="replace")
    info = {}
    for line in (r.stdout or "").splitlines():
        parts = line.split()
        if len(parts) >= 3 and parts[0] == "blob":
            info[parts[2]] = int(parts[1])

    blobs = []
    for ln in lines:
        sha, _, name = ln.partition(" ")
        if sha in info:
            blobs.append((info[sha], name or "(无名)"))
    # 同一个 blob 可能出现多次，去重
    seen = set()
    uniq = []
    for size, name in blobs:
        key = (size, name)
        if key in seen:
            continue
        seen.add(key)
        uniq.append((size, name))

    uniq.sort(reverse=True)
    total = sum(s for s, _ in uniq)
    print("    去重后 %d 个 blob，合计 %.2f MB"
          % (len(uniq), total / 1024 / 1024))
    print()
    print("    最大的 15 个:")
    for size, name in uniq[:15]:
        print("      %9.1f KB  %s" % (size / 1024, name[:70]))

    print()
    print("  三、结论")
    biggest = uniq[0][0] if uniq else 0
    if total > 20 * 1024 * 1024:
        print("    仓库历史 %.1f MB，偏大 —— 推送会被 nginx 的"
              " client_max_body_size 拦（413）" % (total / 1024 / 1024))
        print("    最大单个文件 %.1f MB" / 1024 / 1024 % (biggest / 1024 / 1024)
              if False else "    最大单个文件 %.1f MB" % (biggest / 1024 / 1024))
    else:
        print("    仓库历史 %.2f MB，不算大" % (total / 1024 / 1024))

    print()
    print("  四、本机 Gitee 推得动说明不是仓库本身的问题，")
    print("      是 EzHub 那边的 nginx 上限比 Gitee 小。")
    return 0


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