Start by generating the query using the first aggregation function (SUM in the example) in the measure's list of aggregations to aggregate the template expression.
先使用度量的聚合列表中的第一个聚合函数(在这个示例中是SUM)对表达式模板进行聚合，产生一个查询。
2
we can aggregate subtrees of the "plain" type pairwise. But I would prefer if Spark allowed a more EBNF-style grammar expression like this
我们可以成对地聚集“plain“类型的子树。
3
When you move to a cell in this column and select an aggregate function to apply to the associated data column, by default the resulting expression is added as an output column for the result set.