def ageCalculate(self, column, dateFormat, nameColAge):
"""
This method compute the age of clients based on their born dates.
:param column Name of the column born dates column.
:param dateFormat String format date of the column provided.
:param nameColAge Name of the new column, the new columns is the resulting column of ages.
"""
# Check if column argument a string datatype:
self.__assertTypeStr(column, "column")
# Check if dateFormat argument a string datatype:
self.__assertTypeStr(dateFormat, "dateFormat")
# Asserting if column if in dataFrame:
assert column in self.__df.columns, "Error: Column assigned in column argument does not exist in dataFrame"
# Output format date
Format = "yyyy-MM-dd" # Some SimpleDateFormat string
exprs = format_number(
mag(
months_between(
date_format(
unix_timestamp(column, dateFormat).cast("timestamp"),
Format), current_date()) / 12), 4).alias(nameColAge)
self.__df = self.__df.withColumn(nameColAge, exprs)
self.__addTransformation() # checkpoint in case
return self
def ageCalculate(self, column, dateFormat, nameColAge):
"""
This method compute the age of clients based on their born dates.
:param column Name of the column born dates column.
:param dateFormat String format date of the column provided.
:param nameColAge Name of the new column, the new columns is the resulting column of ages.
"""
# Check if column argument a string datatype:
self.__assertTypeStr(column, "column")
# Check if dateFormat argument a string datatype:
self.__assertTypeStr(dateFormat, "dateFormat")
# Asserting if column if in dataFrame:
assert column in self.__df.columns, "Error: Column assigned in column argument does not exist in dataFrame"
# Output format date
Format = "yyyy-MM-dd" # Some SimpleDateFormat string
exprs = format_number(
mag(
months_between(date_format(
unix_timestamp(column, dateFormat).cast("timestamp"), Format), current_date()) / 12), 4).alias(
nameColAge)
self.__df = self.__df.withColumn(nameColAge, exprs)
self.__addTransformation() # checkpoint in case
return self