llbzow 的摸鱼日记 (づ￣ 3￣)づ

1
import fs from "node:fs";
2
import path from "node:path";
3

4
const SCAN_DIRS = ["src", "public"];
5
const EXTENSIONS = new Set([".md", ".astro", ".html", ".json", ".ts", ".js", ".vue", ".svelte"]);
6

7
// 基础保底字符集 (ASCII 可见字符 + 常用标点符号，防止动态时间/日期漏字)
8
const BASE_CHARS = `
9
0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ
10
~`!@#$%^&*()-_=+[{]}\\|;:'",<.>/?·！￥…（）—【】、；：'“”，。、《》？
11
`;
12

13
const collectedSet = new Set(BASE_CHARS.replace(/\s+/g, ""));
14

15
function scanDirectory(dirPath) {
16
  const entries = fs.readdirSync(dirPath, { withFileTypes: true });
17
  for (const entry of entries) {
18
    const fullPath = path.join(dirPath, entry.name);
19
    if (entry.isDirectory()) {
20
      if (entry.name !== "node_modules" && entry.name !== ".git" && entry.name !== "dist") {
21
        scanDirectory(fullPath);
22
      }
23
    } else if (EXTENSIONS.has(path.extname(entry.name).toLowerCase())) {
24
      extractCharactersFromFile(fullPath);
25
    }
26
  }
27
}
28

29
function extractCharactersFromFile(filePath) {
30
  const content = fs.readFileSync(filePath, "utf8");
31
  // 剥离代码块中的大段机器代码 (可选，如果代码块不需要应用该艺术字体)
32
  const cleanContent = content.replace(/```[\s\S]*?```/g, "");
33

34
  for (const char of cleanContent) {
35
    // 过滤掉不可见控制符与多余空白
36
    const code = char.charCodeAt(0);
37
    if (code > 31 && code !== 127) {
38
      collectedSet.add(char);
39
    }
40
  }
41
}
42

43
// 执行全站扫描
44
console.log("开始扫描全站字符...");
45
for (const dir of SCAN_DIRS) {
46
  if (fs.existsSync(dir)) scanDirectory(dir);
47
}
48

49
// 排序后去重写入中间文件
50
const sortedChars = Array.from(collectedSet).sort().join("");
51
const outputPath = path.resolve("temp-site-glyphs.txt");
52
fs.writeFileSync(outputPath, sortedChars, "utf8");
53

54
console.log(`扫描完成！共提取出 ${collectedSet.size} 个唯一字符，已保存至 ${outputPath}`);