✨ feat: 更好的元数据清理

scripts/sort-keywords.ts

@@ -0,0 +1,71 @@
+import fs from "node:fs";
+import path from "node:path";
+
+const TARGET_FILE = path.join(process.cwd(), "src/assets/data/exclude.ts");
+
+if (!fs.existsSync(TARGET_FILE)) {
+  console.error("❌ 找不到 src/assets/data/exclude.ts，确保你在项目根目录下运行此脚本");
+  process.exit(1);
+}
+
+const fileContent = fs.readFileSync(TARGET_FILE, "utf-8");
+
+// export const keywords = [ ... ];
+const keywordsMatch = fileContent.match(/(export\s+const\s+keywords\s*=\s*\[)([\s\S]*?)(\];)/);
+
+if (!keywordsMatch) {
+  console.error("❌ 找不到 `export const keywords = [...]` 结构");
+  process.exit(1);
+}
+
+const [fullMatch, prefix, rawContent, suffix] = keywordsMatch;
+
+const itemRegex = /(['"`])(.*?)\1/g;
+const rawItems: string[] = [];
+let match: RegExpExecArray | null;
+
+while ((match = itemRegex.exec(rawContent)) !== null) {
+  const content = match[2].trim();
+  if (content) {
+    rawItems.push(content);
+  }
+}
+
+console.log(`✅ 找到 ${rawItems.length} 个关键词`);
+
+const uniqueMap = new Map<string, string>();
+let duplicatesRemoved = 0;
+
+for (const item of rawItems) {
+  const fingerprint = item.toLowerCase().replace(/\s+/g, "");
+
+  if (uniqueMap.has(fingerprint)) {
+    const existing = uniqueMap.get(fingerprint)!;
+
+    if (item.length > existing.length) {
+      uniqueMap.set(fingerprint, item);
+    }
+
+    duplicatesRemoved++;
+  } else {
+    uniqueMap.set(fingerprint, item);
+  }
+}
+
+const uniqueItems = Array.from(uniqueMap.values());
+
+console.log(`🧹 去重完毕，关键词有 ${uniqueItems.length}，移除了 ${duplicatesRemoved}`);
+
+const collator = new Intl.Collator("zh-Hans-CN", { sensitivity: "accent" });
+
+uniqueItems.sort((a, b) => {
+  return collator.compare(a, b);
+});
+
+const newArrayContent = uniqueItems.map((item) => `  "${item}",`).join("\n");
+
+const newContentBlock = `\n${newArrayContent}\n`;
+
+const newFileContent = fileContent.replace(fullMatch, `${prefix}${newContentBlock}${suffix}`);
+
+fs.writeFileSync(TARGET_FILE, newFileContent, "utf-8");