# Available aggregation functions

Note that most functions ignore empty values. This said, functions working on
numbers will yield an error when encountering a string that cannot be safely
parsed as a suitable number.

You can always cast values around and force aggregation functions to
consider empty values or make them avoid non-numerical values altogether.

For instance, considering null values when computing a mean is as easy
as `mean(number || 0)`.

Finally, note that expressions returning lists will be understood as multiplexed rows.
This means that computing `cardinality([source, target])`, for instance, will return
the number of nodes in a graph represented by a CSV edge list.
