投影汇总

Question

Been searching through many methods of aggregation although I cant understand how to achieve my goal. The documents contained within the AggregateIterable in the following code only contain the 2 fields _id and copies as mentioned in Aggregates.group(...) . How would I include the fields field1 and field2 aswell?

AggregateIterable<Document> items = itemCollection.aggregate(Arrays.asList(
    Aggregates.project(include("_id","field1","field2")),
    Aggregates.group("$_id",Accumulators.sum("copies",1)),
    Aggregates.sort(descending("field1")))
);

I'm at a loss since I can't seem to find (or at least understand) any example. Any help would be greatly appreciate, especially an example of how to do as I wish.

Answer 1

the $group agg function in mongo requires you explicitly include any output columns, using a group-friendly transform as well. If you want ghetto passthru for a few fields, i believe there is a $first operator that can come into play.

It is NOT the same as the c# linq group() function where the obects are included regardless. If you need that functionality (various unmatched objects in the value collection for the key), you will have to instantiate the results earlier, then use the standard linq group function.

Answer 2

Here is the example of what I got working, using the first() operator.

AggregateIterable<Document> items = Main.itemCollection.aggregate(Arrays.asList(
    Aggregates.project(fields(include("field1","field2"),excludeId())),
    Aggregates.group(
        "$field1",
        first("field2","$field2r")
        Accumulators.sum("Copies",1)
    ),
    Aggregates.sort(descending("_id"))
));