MongoDB: Return objects in a nested array that has the most recent date - mongodb

Question

I have the following collection:
{
"_id" : ObjectId("56f036e032ea1f27634e2f1f"),
"mockups" : [
{
"versions" : [
{
"title" : "About us 1",
"timestamp" : "2016-01-10T12:31:23.104Z",
"_id" : ObjectId("56ec65a9041c87dd6bd17922")
},
{
"title" : "About us 3",
"timestamp" : "2016-03-11T15:34:11.108Z",
"_id" : ObjectId("56ec65a9041c87dd6bd17923")
},
{
"title" : "About us 2",
"timestamp" : "2016-02-21T16:15:23.101Z",
"_id" : ObjectId("56ec65a9041c87dd6bd17924")
}
],
"_id" : ObjectId("56ec65a9041c87dd6bd17921")
},
{
"versions" : [
{
"title" : "Contact us 1",
"timestamp" : "2016-04-10T11:34:33.103Z",
"_id" : ObjectId("56ec65a9041c87dd6bd17924")
},
{
"title" : "Contact us 3",
"timestamp" : "2016-06-21T16:13:26.101Z",
"_id" : ObjectId("56ec65a9041c87dd6bd17926")
},
{
"title" : "Contact us 2",
"timestamp" : "2016-05-11T13:34:13.106Z",
"_id" : ObjectId("56ec65a9041c87dd6bd17925")
}
],
"_id" : ObjectId("56ec65a9041c87dd6bd17929")
}
]
}
I want to return all mockups with the latest version that would result in something similar to the following:
{
"_id" : ObjectId("56f036e032ea1f27634e2f1f"),
"mockups" : [
{
"versions" : [
{
"title" : "About us 3",
"timestamp" : "2016-03-11T15:34:11.108Z",
"_id" : ObjectId("56ec65a9041c87dd6bd17923")
}
],
"_id" : ObjectId("56ec65a9041c87dd6bd17921")
},
{
"versions" : [
{
"title" : "Contact us 3",
"timestamp" : "2016-06-21T16:13:26.101Z",
"_id" : ObjectId("56ec65a9041c87dd6bd17926")
}
],
"_id" : ObjectId("56ec65a9041c87dd6bd17929")
}
]
}
I have been playing with aggregate, sort, and limit, but I am extremely new to mongodb. I am currently left with having to just return everything and using something like lodash to get the version I need.
Is there a way to query this properly so that I am getting the results I need from the mongo as opposed to using something like lodash after I get the results back?

Answer 1

For a solution using aggregation, would suggest running the following pipeline to get the desired output:
db.collection.aggregate([
{ "$unwind": "$mockups" },
{ "$unwind": "$mockups.versions" },
{ "$sort": { "mockups.versions.timestamp": -1 } },
{
"$group": {
"_id": "$mockups._id",
"title": { "$first": "$mockups.versions.title" },
"timestamp": { "$first": "$mockups.versions.timestamp" },
"id" : { "$first": "$mockups.versions._id" },
"main_id": { "$first": "$_id" }
}
},
{
"$group": {
"_id": "$_id",
"versions": {
"$push": {
"title": "$title",
"timestamp": "$timestamp",
"_id": "$id"
}
},
"main_id": { "$first": "$main_id" }
}
},
{
"$group": {
"_id": "$main_id",
"mockups": {
"$push": {
"versions": "$versions",
"_id": "$_id"
}
}
}
}
])

Answer 2

I didn't actually understood what is the range for "latest" version ,however I think you can use some thing like this:
example:
db.collection.find({
versions.timestamp: {
$gte: ISODate("2010-04-29T00:00:00.000Z"),
$lt: ISODate("2010-05-01T00:00:00.000Z")
}
})
if you want to use the aggregate you can put this condition in the $match .