修复:批量添加禁用词 UTF-8 编码错误

- 新增 sanitizeUtf8() 私有方法:
  去除 BOM、零宽字符、非法 UTF-8 字节、控制字符
  防止从聊天/文档复制时混入隐藏字符导致 json_encode 失败
- batchStore() 先净化 words 和 reason 输入
- preg_split 加 /u 修饰符(完整 Unicode 支持)
- 响应加 JSON_UNESCAPED_UNICODE 防中文被转义
This commit is contained in:
2026-03-01 14:10:13 +08:00
parent f114c6b168
commit 477bba3003

View File

@@ -97,11 +97,14 @@ class ForbiddenUsernameController extends Controller
'words.required' => '请输入至少一个词语。',
]);
// 支持换行、逗号、中文逗号、空格分隔
$rawWords = preg_split('/[\r\n,\s]+/', $validated['words']);
$reason = trim($validated['reason'] ?? '');
// 净化输入:去除非法 UTF-8 字节零宽字符、BOM、控制字符等防止 json_encode 失败
$rawInput = $this->sanitizeUtf8($validated['words']);
$reason = $this->sanitizeUtf8(trim($validated['reason'] ?? ''));
// 过滤空串、截断过长、去重
// 支持换行、逗号、中文逗号、空格分隔
$rawWords = preg_split('/[\r\n,\s]+/u', $rawInput);
// 过滤空串、超长词、去重
$words = collect($rawWords)
->map(fn ($w) => trim($w))
->filter(fn ($w) => $w !== '' && mb_strlen($w) <= 50)
@@ -112,11 +115,11 @@ class ForbiddenUsernameController extends Controller
return response()->json(['status' => 'error', 'message' => '没有有效的词语,请检查输入。'], 422);
}
// 查询已存在的(批量一次查询)
// 批量查询已存在的(一次查询)
$existing = UsernameBlacklist::permanent()
->whereIn('username', $words->all())
->pluck('username')
->flip(); // 转为键名方便 has() 查询
->flip();
$now = Carbon::now();
$added = 0;
@@ -124,7 +127,7 @@ class ForbiddenUsernameController extends Controller
foreach ($words as $word) {
if ($existing->has($word)) {
continue; // 跳过已存在
continue;
}
$rows[] = [
'username' => $word,
@@ -143,7 +146,27 @@ class ForbiddenUsernameController extends Controller
$skipped = $words->count() - $added;
$msg = "成功添加 {$added} 个词语".($skipped > 0 ? ",跳过 {$skipped} 个(已存在)" : '').'。';
return response()->json(['status' => 'success', 'message' => $msg, 'added' => $added]);
return response()->json(['status' => 'success', 'message' => $msg, 'added' => $added], 200, [], JSON_UNESCAPED_UNICODE);
}
/**
* 净化字符串,移除非法 UTF-8 字节及常见控制/零宽字符。
*
* @param string $str 待净化字符串
* @return string 合法的 UTF-8 字符串
*/
private function sanitizeUtf8(string $str): string
{
// 去除 BOM
$str = str_replace("\xEF\xBB\xBF", '', $str);
// 去除零宽字符(零宽空格、零宽不连字等)
$str = preg_replace('/[\x{200B}-\x{200D}\x{FEFF}\x{00AD}]/u', '', $str);
// 转换为合法 UTF-8忽略非法字节
$str = mb_convert_encoding($str, 'UTF-8', 'UTF-8');
// 保底:去除控制字符(保留换行 \r\n
$str = preg_replace('/[^\P{C}\r\n]+/u', '', $str);
return $str;
}
/**