private lazy var textField: UITextView = {
        let fontSize: CGFloat = 16

        let tv = UITextView()
        tv.font = UIFont.systemFont(ofSize: fontSize)
        tv.textAlignment = .center

        var text = NSMutableAttributedString(string: "一二三四五上山打老虎")
        for _ in 1...100 {
            let att = NSTextAttachment()
            att.image = UIImage(named: "1f601")
            att.bounds = CGRect(x: 0, y: tv.font?.descender ?? 0, width: fontSize , height: fontSize)
            text.append(NSAttributedString(attachment: att))
        }

        tv.attributedText = text
        return tv
    }()

关于对齐的问题tv.font?.descender是关键，具体可参考：http://stackoverflow.com/questions/26105803/center-nstextattachment-image-next-to-single-line-uilabel

这个场景实际就是聊天中的文字+表情混排。

另外要说明下，UITextField（单行输入框）和NSTextAttachment有兼容问题，暂时 没法用这种方法实现。