#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""compare-tree.py —— 把**本地 git 基线**（manifest：mode/type/sha\\tpath）与服务器工作树逐文件比对。

输出三类（这是"能不能整树覆盖"的唯一判据，不盲覆盖）：
  · 一致     ：服务器文件内容 == 本地基线
  · 不一致   ：两边都有但内容不同（**要逐个看是谁的改动**）
  · 服务器独有：服务器有、基线里没有（**绝不能抹**）
再加一类「仅在基线里」：本地有、服务器缺（同步时要补上）。

用法（在服务器上）：python3 scripts/compare-tree.py <manifest> <工作树根>
"""
import os
import subprocess
import sys

manifest, root = sys.argv[1], sys.argv[2]
entries = []
for line in open(manifest, encoding='utf-8'):
    line = line.rstrip('\n')
    if not line.strip():
        continue
    head, path = line.split('\t', 1)
    mode, typ, sha = head.split()
    entries.append((path, sha))

paths = [p for p, _ in entries]
present = [p for p in paths if os.path.isfile(os.path.join(root, p))]
missing = [p for p in paths if p not in set(present)]

# 批量算 sha（git hash-object --stdin-paths）
hash_out = {}
if present:
    proc = subprocess.run(['git', 'hash-object', '--stdin-paths'], cwd=root,
                          input='\n'.join(present) + '\n', capture_output=True, text=True)
    for p, h in zip(present, proc.stdout.split('\n')):
        if h.strip():
            hash_out[p] = h.strip()

same, diff = [], []
for p, sha in entries:
    if p in missing:
        continue
    if hash_out.get(p) == sha:
        same.append(p)
    else:
        diff.append((p, sha, hash_out.get(p)))

# 服务器独有：扫同一批顶层目录，取基线里没有的文件
tops = sorted({p.split('/')[0] for p, _ in entries})
known = set(paths)
extra = []
for top in tops:
    base = os.path.join(root, top)
    if not os.path.isdir(base):
        continue
    for dirpath, dirnames, filenames in os.walk(base):
        dirnames[:] = [d for d in dirnames if d not in ('node_modules', '.git', 'dist')]
        for fn in filenames:
            rel = os.path.relpath(os.path.join(dirpath, fn), root).replace(os.sep, '/')
            if rel not in known:
                extra.append(rel)

print('=== 汇总 ===')
print('  基线条目      : %d' % len(entries))
print('  一致          : %d' % len(same))
print('  不一致        : %d' % len(diff))
print('  仅基线有(缺)  : %d' % len(missing))
print('  服务器独有    : %d' % len(extra))

if diff:
    print('\n=== 不一致（要逐个判断是谁的改动）===')
    for p, want, got in diff[:60]:
        print('  %-58s 基线 %s  服务器 %s' % (p, want[:10], (got or '')[:10]))
    if len(diff) > 60:
        print('  … 还有 %d 个' % (len(diff) - 60))

if missing:
    print('\n=== 仅基线有（服务器缺）===')
    for p in missing[:40]:
        print('  ' + p)
    if len(missing) > 40:
        print('  … 还有 %d 个' % (len(missing) - 40))

if extra:
    print('\n=== 服务器独有（⛔ 覆盖前必须保留/处理）===')
    for p in extra[:60]:
        print('  ' + p)
    if len(extra) > 60:
        print('  … 还有 %d 个' % (len(extra) - 60))