Question

我正在使用POSIX运算符（〜）执行以下Redshift SQL命令以进行模式匹配（如果字符串中的任意位置连续9位，则返回true，否则返回false）

for (var row of rows) {    
  delete row['isTrue'];  // this will delete the isTrue key from the object
  delete row['isAvailable'];  // this will delete the isAvailable key from the object
}

select '123456789' ~ '\\d{9}' as val;  --TRUE

select 'abcd123456789' ~ '\\d{9}' as val;  --TRUE

如何在SparkSQL中进行一些相同的模式匹配？

Answer 1

我相信rlike应该可以解决问题：

spark.sql("""SELECT '123456789' rlike '\\\d{9}' as val""").show()

spark.sql("""SELECT 'ab123456789' rlike '\\\d{9}' as val""").show()

spark.sql("""SELECT '123456789abcd' rlike '\\\d{9}' as val""").show()

所有结果：

+----+
| val|
+----+
|true|
+----+

并且：

spark.sql("""SELECT '12345678abcd' rlike '\\\d{9}' as val""").show()

结果：

+-----+
|  val|
+-----+
|false|
+-----+

什么是SparkSQL中Redshift的posix运算符的替代品？

1 个答案: