Creates a subset of the table by masking the rows with the given column of Booleans.


subscript(mask: MLDataColumn<Bool>) -> MLDataTable { get }



A Boolean column indicating whether rows should be kept (true) or removed (false) in the derived table.

Return Value

A new data table.


Use this Boolean column–based subscript to create a new table by masking a subset of the table rows.

A table of book titles and genres such as "Hamlet" and "Drama" on the left side of the image. A one-column table titled "mask" in the middle of the image with true and false cells lined up with genres, where rows with "Fantasy" have a mask of false and all other rows have a mask of true. Arrows from the true cells in the mask column to all the rows in a table on the right hand side showing a subset of the rows in the table on the left side of the image. None of the genres in this table are "Fantasy".

For example, to filter the values in a data table as shown above, begin by creating a table with the original data.

let data: [String: MLDataValueConvertible] = [
    "Title": ["Alice in Wonderland", "Hamlet", "Treasure Island", "Peter Pan"],
    "Genre": ["Fantasy", "Drama", "Adventure", "Fantasy"]

let table = try? MLDataTable(dictionary: data) else {
    fatalError("Invalid dictionary data")

After you create the table, use column arithmetic operators or the map(_:) method to build a row mask. The subscript uses the Boolean values in the row mask to determine whether to keep a row.

// Retrieve the "Genre" column from the table.
guard let genreColumn = table["Genre", String.self] else {
    fatalError("Missing or invalid 'genre' column in table.")

// Create a new column of Booleans by comparing all of the values
// in `Genre` with `Fantasy` using the
// `!=(MLDataColumn<String>, String) -> MLDataColumn<Bool>` operator.
let noFantasyMask = genreColumn != "Fantasy"

Use subscript(mask: MLDataColumn<Bool>) with the Boolean column–row mask to create a filtered table.

let noFantasyTable = table[noFantasyMask]

See Also

Masking Rows

subscript(MLUntypedColumn) -> MLDataTable

Creates a subset of the table by masking the rows with the given untyped column.