add

Returns DataFrame which contains all columns from original DataFrame followed by newly added columns. Original DataFrame is not modified.

Create a new column and add it to DataFrame

add(columnName: String) { rowExpression }

rowExpression: DataRow.(DataRow) -> Value

df.add("year of birth") { 2021 - age }

val age by column<Int>()
val yearOfBirth by column<Int>("year of birth")

df.add(yearOfBirth) { 2021 - age }

df.add("year of birth") { 2021 - "age"<Int>() }

See row expressions

You can use the newValue() function to access value that was already calculated for the preceding row. It is helpful for recurrent computations:

df.add("fibonacci") {
    if (index() < 2) 1
    else prev()!!.newValue<Int>() + prev()!!.prev()!!.newValue<Int>()
}

Create and add several columns to DataFrame

add { 
    columnMapping
    columnMapping
    ...
}

columnMapping = column into columnName 
    | columnName from column 
    | columnName from { rowExpression }
    | columnGroupName { 
        columnMapping
        columnMapping
        ...
    }

df.add {
    "year of birth" from 2021 - age
    age gt 18 into "is adult"
    "details" {
        name.lastName.length() into "last name length"
        "full name" from { name.firstName + " " + name.lastName }
    }
}

val yob = column<Int>("year of birth")
val lastNameLength = column<Int>("last name length")
val age by column<Int>()
val isAdult = column<Boolean>("is adult")
val fullName = column<String>("full name")
val name by columnGroup()
val details by columnGroup()
val firstName by name.column<String>()
val lastName by name.column<String>()

df.add {
    yob from 2021 - age
    age gt 18 into isAdult
    details from {
        lastName.length() into lastNameLength
        fullName from { firstName() + " " + lastName() }
    }
}

df.add {
    "year of birth" from 2021 - "age"<Int>()
    "age"<Int>() gt 18 into "is adult"
    "details" {
        "name"["lastName"]<String>().length() into "last name length"
        "full name" from { "name"["firstName"]<String>() + " " + "name"["lastName"]<String>() }
    }
}

Create columns using intermediate result

Consider this API:

class CityInfo(val city: String?, val population: Int, val location: String)

fun queryCityInfo(city: String?): CityInfo {
    return CityInfo(city, city?.length ?: 0, "35.5 32.2")
}

Use the following approach to add multiple columns by calling the given API only once per row:

val personWithCityInfo = df.add {
    val cityInfo = city.map { queryCityInfo(it) }
    "cityInfo" {
        cityInfo.map { it.location } into CityInfo::location
        cityInfo.map { it.population } into "population"
    }
}

val city by column<String?>()
val personWithCityInfo = df.add {
    val cityInfo = city().map { queryCityInfo(it) }
    "cityInfo" {
        cityInfo.map { it.location } into CityInfo::location
        cityInfo.map { it.population } into "population"
    }
}

val personWithCityInfo = df.add {
    val cityInfo = "city"<String?>().map { queryCityInfo(it) }
    "cityInfo" {
        cityInfo.map { it.location } into CityInfo::location
        cityInfo.map { it.population } into "population"
    }
}

Add existing column to DataFrame

val score by columnOf(4, 3, 5, 2, 1, 3, 5)

df.add(score)
df + score

Add all columns from another DataFrame

df.add(df1, df2)

addId

Adds a column with sequential values 0, 1, 2,... The new column will be added in the beginning of the column list and will become the first column in DataFrame.

addId(name: String = "id")

Parameters:

name: String = "id" - name of the new column.

Last modified: 29 March 2024