Distribution Functions
Overview
Reference
random
random() double
Returns a uniformly distributed double number, greater than or equal to 0.0 and less than 1.0.
The mean value is 0.5. The range of values is [0, 1).
To generate a sample for a customized mean and value range, use the following formula:
customMean + (random()-0.5)*customRange
# Generate a random value from a uniform distribution with a mean of 10 in the [9, 11) range.
10 + (random()-0.5)*2
randomNormal
randomNormal() double
Returns a normally distributed double number, with a mean value of 0.0 and standard deviation 1.0.
To generate a sample for a customized mean and deviation, use the following formula:
customMean + randomNormal()*customStdDev
# Generate a random value from a normal distribution with a mean of 10 and standard deviation of 2.
# 68.2% of values will be in the [8, 12] range.
10 + randomNormal()*2
randomItem
randomItem([object] c | map m) string
Returns a random element from a collection or map using the uniform distribution.
The probability of each element to be selected is 1/c.size() or 1/m.size() respectively.
Returns the selected element converted to string format. For maps, the returned object is a key-value object.
Input collection can contain elements of any type, such as strings or numbers, and can be specified as follows:
String Collection
randomItem(['a', 'b', 'c'])
Number Collection
randomItem([1, 2, 3])
Note that although the input collection contains numbers, the returned element is a string which has to be parsed to a number if necessary.
randomItem([1, 2, 3]) = '2'
Double.parseDouble(randomItem([1, 2, 3])) >= 2
Named Collection
Named collections are listed on the Data > Named Collections page.
Assuming the collection contains the following records and the 2nd entry is randomly selected:
John Doe
Mary Jones
Jack Smith
// returns Mary Jones
randomItem(collection('oncall-person'))
Key-Value Map
Key-value maps can be loaded with the replacementTable function.
The replacement tables are listed on the Data > Replacement Tables page.
Assuming the replacement table contains the following rows and the second entry is randomly selected:
John Doe =(800) 555-0100
Mary Jones=(800) 555-0200
Jack Smith=(800) 555-0300
// returns Mary Jones
randomItem(replacementTable('oncall-person').keySet())
// returns (800) 555-0200
randomItem(replacementTable('oncall-person').values())
/* Returns a key-value object consisting of string key and string value
The object fields can be accessed with .key and .value methods. */
randomItem(replacementTable('oncall-person'))
selItem = randomItem(replacementTable('oncall-person'))
selKey = selItem.key
randomKey
randomKey(map m) string
Returns a random element from the specified map m of objects using uniform distribution.
The keys in the map can be of any type, whereas the values must be numeric and represent probabilities of the given key to be selected.
The function returns the randomly selected key converted to string.
The sum of probabilities does not have to equal 1.0 as the inputs are weighted to total 1.0.
An input map can be obtained using the replacementTable() lookup function.
randomKey(replacementTable('oncall-person'))
Replacement tables are listed on the Data > Replacement Tables page.
Assuming the table contains the following records, the second element has a 20% chance of being selected:
John Doe=0.5
Mary Jones=0.2
Jack Smith=0.3
The excludeKeys function can be used to remove some elements from the input map prior to invoking the randomKey function.
randomKey(excludeKeys(replacementTable('oncall-person'),['Jack Smith']))