These are aggregate functions as available on stream views and pipes. Aggregate functions serve the purpose of aggregating data and giving it more eaningfull meaning. They are usually a common feature in any data based environment. In database management, for instance aggregate functions are used where multiple rows are grouped together to form a singe value or to find out more significant meaning, represented by a single value on a set of values.
Aggregate functions are also available on pipes and views in PhixFlow. The following composes a set of available functions:
| Function | Description | ||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| average | This function returns the arithmetic average of the set of numbers. It takes one argument (a name of a field) of numeric type: Long, Float or Integer. It is only available if field on which it is applied, is of such type. | ||||||||||||||||||||||
| maximum | This function returns the maximum numeric value out of the set of numeric values. It takes one argument (a name of a field) of type, either: Long, Float, or Integer and it can be applied only on field of such type and only then it is available. | ||||||||||||||||||||||
| minimum | This function returns the minimum numeric value out of the set of numeric values. It takes one argument (a name of a field) of type, either: Long, Float, or Integer and it can be applied only on field of such type and only then it is available. | ||||||||||||||||||||||
| sum | This function returns the sum calculated by suming all the values in a given set of values. It is applicable only (and only then it is available) if the name of the field passed to this function represents numeric field of type: Long, Float or Integer. | ||||||||||||||||||||||
| percent | This function expects a name of numeric field to calculate percentage of sum of numerical values of each group in relation to a whole set of grouped values. It is therefore necessary to use this function in conjunction with grouping attribute, as otherwise only default grouping would be applied and this would present no meaningful result, as it would always default to 100% (each record would have been in its own group). If however we assume having a file with 2 records, like so:
and we also assume grouping by "Label", our percent function would return (if applied o "Number" attribute), the following result in a view (assuming we also selected Label attribute to be shown in our view):
| ||||||||||||||||||||||
| distinct | |||||||||||||||||||||||
| count |