Conditional Operations

Ternary If Statements

Conditional operations allow users to check for certain conditions and then assign different values depending on the outcome. This is commonly implemented as if-then-else statements in programming. Another common form, but with fewer characters/keystrokes, is the inline if, or ternary if.

The ternary operator is a conditional expression that includes three operands:

a condition that evaluates to true or false
outcome 1 (if true)
outcome 2 (if false)

The syntax for a ternary if statement with only one condition to evaluate follows:

condition ? value if condition is true : value if condition is false

The question mark (?) separates the condition from the values, and the colon (:) is used to separate the values to be used after evaluating the condition.

For example, the expression x ? y : z evaluates to y if the value of x is true. If x is false, the operation evaluates to z.

Nesting Multiple Ternary If Statements

Ternary if statements can also be nested to evaluate more than one condition. Here is the syntax when evaluating two conditions:

condition1 ? condition2 ? value1 : value2 : value3

As shown below, adding parentheses around the nested condition and value pair makes the query more readable, but they do not alter the results.

condition1 ? (condition2 ? value1 : value2) : value3

The table below shows how the multiple conditions are evaluated logically:

Condition1	Condition2	Value
True	True	value1
True	False	value2
False	[skipped]	value3

Let's look at the following example of a nested ternary if operation:

x ? (y ? 1 : 2) : 3

If both x and y are true, the expression evaluates to 1.
If x is true, and y is false, the expression evaluates to 2.
If x is false, the expression evaluates to 3.

Example

The example below illustrates how to use a conditional operation (ternary if) with the update() method in a Deephaven query.

The table below has two columns of data. We want to check each existing row for null values. We then want to create a new column in the table and insert different values in that new column based on whether null values are present in either or both columns for that given row.

t

ColA	ColB

	20
30
40	55

The following query can be used to process that logic:

t=t.update("ColC = isNull(ColA) ? (isNull(ColB) ? 0 : ColB) : ColA")

The query begins with the .update method, which defines a new column, ColC.

The value to be used for ColC will be determined by the conditions contained in the embedded ternary if statement, highlighted below:

t=t.update("ColC = isNull(ColA) ? (isNull(ColB) ? 0 : ColB) : ColA")

The possible outcomes of each condition are represented in the table below:

Condition 1 Value in ColA is null	Condition 2 Value in ColB is null	Value to insert in ColC
True	True	0
True	False	ColB
False	[skipped]	ColA

Here's the table generated after running the query:

t

ColA	ColB	ColC	Explanation
		0	The values in `ColA` and `ColB` are both null, so `ColC` is set to 0.
	20	20	The value in `ColA` is null, but the value in `ColB` is not null, so, `ColC` is set to the same value as that in `ColB`.
30		30	The value in `ColA` is not null, so `ColC` will be set to the same value as `ColA`.
40	55	30	The value in `ColA` is not null, so `ColC` will be set to the same value as `ColA`. The value present in `ColB` does not matter.

Last Updated: 16 February 2021 18:07 -04:00 UTC Deephaven v.1.20200928 (See other versions)