table< T >

A tabular representation of the type T. The tabular result will contain columns of scalar types similar to a relational table.

Given an example such as:

entity Person key( Id : long ) {
    Name : string
    Age : int
    Phone : list< string >
}

A example value of type table< Person > is represented as:

Id :long	Name :string	Age :int	Phone_Id__ :int	Phone :string
001	Bob	23	0	02930918290
001	Bob	23	1	02949084509
002	Joe	25	0	07940938092

The flatten rules are:

1. A scalar becomes a column.
2. Some flattening will result in auxiliary columns being added. Auxiliary column is prefixed with string in form of ‘_usage__’ where usage is replaced with an appropriate name given the context, such as index. Naming conflicts are possible, for this reason, ALFA compiler reports a warning if fields are declared with 2 consecutive underscores, which may result in conflict with internally generated names.
3. A map< K, V> becomes a minimum of 2 columns, more based on how K or V flatten.
4. The key and value fields of a map will name such that Key and Value are appended to the field name.
5. The default Map column names can be overridden by specifying them in the declaration - map< KeyName : K, ValueName : V>
6. A set< T > becomes a minimum of 2 columns, containing an _identity__ column and flattening for T
7. A list< T > becomes a minimum of 2 columns, containing an _index__ column and flattening for T
8. When T is a trait, the set of columns are derived from the trait, however there can be more columns based on implementations of the trait. Therefore the columns is a union of all trait implementations.
9. An optional T? value is represented by a minimum of 2 columns. An _isSet__ prefixed field and flattening for T.

Exhaustive list of all ALFA flattening for table - Additional content will be published soon.

Optional Parameters

• table< T >( ... ) Optional named parameters can be specified to table to manipulate the table type. Currently the usage and types for the parameters are undefined.