Cannot resolve column (numeric column name) in Spark Dataframe

Question

This is my data:

scala> data.printSchema
root
 |-- 1.0: string (nullable = true)
 |-- 2.0: string (nullable = true)
 |-- 3.0: string (nullable = true)


        

Answer 1

You can use backticks to escape the dot, which is reserved for accessing columns for struct type:

data.select("`2.0`").show
+---+
|2.0|
+---+
| , |
+---+

Answer 2

The problem is you can not add dot character in the column name while selecting from dataframe. You can have a look at this question, kind of similar.

val data = spark.createDataFrame(Seq(
  ("Hello", ", ", "World!")
)).toDF("1.0", "2.0", "3.0")
data.select(sanitize("2.0")).show

def sanitize(input: String): String = s"`$input`"