构建Dataframe不需要for循环。可以将数组转换为 Seq 使用 toDF a方法 Seq 构造生成的Dataframe。

def getUnicodeOfEmoji (emojiArray : Array[String]) : DataFrame = {
    val existingSparkSession = SparkSession.builder().getOrCreate()
    import existingSparkSession.implicits._
    val df = emojiArray.toSeq.toDF("emoji")
    val result = df.selectExpr(
        "emoji",
        "'U+' || trim('0' , string(hex(encode(emoji, 'utf-32')))) as result"
    )
    result.show(false)
    return result
  }

val emojis = "???"
val input = raw"\p{block=Emoticons}".r.findAllIn(emojis).toArray
val converted = getUnicodeOfEmoji(input)
+-----+-------+
|emoji|result |
+-----+-------+
|?   |U+1F603|
|?   |U+1F61C|
|?   |U+1F60D|
+-----+-------+

一个小小的改进是将你的表情串转换成 Seq[String] 直接输入函数之前，例如。

def getUnicodeOfEmoji (emojiArray : Seq[String]) : DataFrame = {
    val existingSparkSession = SparkSession.builder().getOrCreate()
    import existingSparkSession.implicits._
    val df = emojiArray.toDF("emoji")
    val result = df.selectExpr(
        "emoji",
        "'U+' || trim('0' , string(hex(encode(emoji, 'utf-32')))) as result"
    )
    result.show(false)
    return result
  }

val emojis = "???"
val input = raw"\p{block=Emoticons}".r.findAllIn(emojis).toSeq
val converted = getUnicodeOfEmoji(input)