Reputation: 91
I am new to MongoDB, I have two collections like this :
1st collection name is a
db.a.find()
{
"_id": "1234",
"versions": [{
"owner_id": ObjectId("100000"),
"versions": 1,
"type" : "info",
"items" : ["item1","item3","item7"]
},
{
"owner_id": ObjectId("100001"),
"versions": 2,
"type" : "bug",
"OS": "Ubuntu",
"Dependencies" : "Trim",
"items" : ["item1","item7"]
}
]}
2nd Collection name is b
db.b.find()
{
"_id": ObjectId("100000"),
"email": "[email protected]"
} {
"_id": ObjectId("100001"),
"email": "[email protected]"
}
Expected output is:
{
"_id": "1234",
"versions":[{
"owner_id": "[email protected]",
"versions": 1,
"type" : "info",
"items" : ["item1","item3","item7"]
},
{
"owner_id": "[email protected]",
"versions": 2,
"type" : "bug",
"OS": "Ubuntu",
"Dependencies" : "Trim",
"items" : ["item1","item7"]
}
] }
Requirement: fields inside each document of versions are not fixed, Example : versions[0] have 4 key-value pair and versions[1] have 6 key-value pair. so I am looking a query which can replace owner_id with email keeping all other filed in output.
I tried :
db.a.aggregate(
[
{$unwind:"$versions"},
{$lookup : {from : "b", "localField":"versions.owner_id", "foreignField":"_id", as :"out"}},
{$project : {"_id":1, "versions.owner_id":{$arrayElemAt:["$out.email",0]}}},
{$group:{_id:"$_id", versions : {$push : "$versions"}}}
]
).pretty()
Please help.
Thank You!!!
Upvotes: 1
Views: 2952
Reputation: 5252
Instead of $project pipeline stage use $addFields.
Example:
db.a.aggregate([
{ $unwind: "$versions" },
{
$lookup: {
from: "b",
localField: "versions.owner_id",
foreignField: "_id",
as: "out"
}
},
{
$addFields: {
"versions.owner_id": { $arrayElemAt: ["$out.email",0] }
}
},
{ $group: { _id: "$_id", versions: { $push: "$versions" } } }
]).pretty()
Upvotes: 3