Can anyone help me to group the data by key without using $unwind nor $group stage ? The reason why I don't want to use $unwind nor $group stage is because I want to be able to use the pipeline in a updateMany(filter, pipeline) operation so the stages I can use are limited to: $addFields (= $set), $project (= $unset) and $replaceRoot (= $replaceWith).
The input data looks like this:
[
{
key: "alpha",
value: {
name: "foo",
},
},
{
key: "beta",
value: {
name: "bar",
},
},
{
key: "alpha",
value: {
name: "baz",
},
},
]
The result I would like to get:
[
{
key: "alpha",
values: [
{
name: "foo",
},
{
name: "baz",
},
],
},
{
key: "beta",
values: [
{
name: "bar",
},
],
},
]
I believe that it's doable by using $reduce but I'm newbie with mongodb aggregation so I'm struggling to accumulate objects in array conditionally.
Thanks
[
{
$set: {
keys: { $setUnion: [ "$array.key" ] }
}
},
{
$set: {
newarray : {
$map : {
input : "$keys",
in : {
key : "$$this",
values : {
$map : {
input : {
$filter : {
input : "$array",
as : "elem",
cond : {
$eq : [
"$$elem.key",
"$$this"
]
}
}},
as : "vals",
in : {
name : "$$vals.value.name"
}
}
}
}
}
}
}
}
]
Try it. https://mongoplayground.net/p/m1UUvt5QNgg Or more compact version: https://mongoplayground.net/p/bFrSiyc5nSc