Search code examples
pythonapache-spark-sqlpyspark

Convert column of binary string to int in spark dataframe python

So I have a dataframe with one column like this:

+----------+
|some_colum|
+----------+
|        10|
|        00|
|        00|
|        10|
|        10|
|        00|
|        10|
|        00|
|        00|
|        10|
+----------+

where the column some_colum are binary strings.

I want to convert this column to decimal.

I've tried doing

data = data.withColumn("some_colum", int(col("some_colum"), 2))

But this doesn't seem to work. as I get the error:

int() can't convert non-string with explicit base

I think cast() might be able to do the job but I'm unable to figure it out. Any ideas?

Solution

  • I think the int cannot be applied directly to a column. You can use in a udf:

    from org.apache.spark.sql import functions
binary_to_int = functions.udf(lambda x: int(x, 2), IntegerType())
data = data.withColumn("some_colum", binary_to_int("some_colum").alias('some_column_int'))