Multiples group aggregate in mongodb - node.js

Question

Lets say i have a collection of books like this :
{author:"john", category:"action", title:"foobar200"},
{author:"peter", category:"scifi" , title:"42test"},
{author:"peter", category:"novel", title:"whatever_t"},
{author:"jane", category:"novel", title:"the return"},
{author:"john", category:"action", title:"extreme test"},
{author:"peter", category:"scifi", title:"such title"},
{author:"jane", category:"action", title:"super book "}
I want to do a query similar to :
SELECT author,category, count(*) FROM books GROUP BY category, author
==> result :
john -> action -> 2
john -> novel -> 0
john -> scifi -> 0
jane -> action -> 1
etc...
the closest i've been to the solution is this :
db.books.aggregate(
{
$match: {category:"action"}
},
{
$group: { _id: '$author', result: { $sum: 1 } }
}
);
==> result
{ "_id" : "jane", "result" : 1 }
{ "_id" : "john", "result" : 2 }
{ "_id" : "peter", "result" : 0 }
But i can't understand how to perform the second "group by" with categories.
What is the best way to do this ?
Thanks

Answer 1

You can include multiple fields in the _id used by $group to provide multi-field grouping:
db.books.aggregate([
{$group: {
_id: {category: '$category', author: '$author'},
result: {$sum: 1}
}}
])
Result:
{
"_id" : {
"category" : "action",
"author" : "jane"
},
"result" : 1
},
{
"_id" : {
"category" : "novel",
"author" : "jane"
},
"result" : 1
},
{
"_id" : {
"category" : "novel",
"author" : "peter"
},
"result" : 1
},
{
"_id" : {
"category" : "scifi",
"author" : "peter"
},
"result" : 2
},
{
"_id" : {
"category" : "action",
"author" : "john"
},
"result" : 2
}