I'm trying to group items by tag. But I do not get what exactly I want to do.
There is schema Item (_id, name, tags) with tags is array of tags_id ( was referenced from Tag schema )
Item = new Schema({
id: Schema.ObjectId,
name: String,
tags: [ { type: Schema.ObjectId, ref: 'Tag' }]
});
Tag = new Schema({
id: Schema.ObjectId,
name: String
});
This is what I have done so far:
Item.aggregate([
{
"$lookup": {
"from": Tag.collection.name,
"let": { "tags": "$tags" },
"pipeline": [
{
"$match": {
"$expr": { "$in": ['$_id', '$$tags'] }
}
},
{
"$project": {
"_id": 1,
"name": 1
}
}
],
"as": 'tags'
}
},
{
"$group": {
"_id": "$tags._id",
"items": {
"$push": {
"id": "$_id",
"name": "$name",
}
}
}
}
]);
This is what I get
{
"data": [
{
"_id": [
"5eb95e8dcae79713f1de0a27"
],
"items": [
{
"id": "5eb95e9fcae79713f1de0a28",
"name": "My Item 1"
}
]
},
{
"_id": [
"5eb9564dc4317411fe79e1bf"
],
"items": [
{
"id": "5eb95b1430f138131ed90f4f",
"name": "My Item 2"
},
{
"id": "5eb95ed0cae79713f1de0a29",
"name": "My Item 3"
}
]
}
]
}
I would like to get the name of each tag, and not only the _id possibly not within an array. Below the result that I would like to receive:
{
"data": [
{
"_id": "5eb95e8dcae79713f1de0a27",
"name": "My Tag name 1",
"items": [
{
"id": "5eb95e9fcae79713f1de0a28",
"name": "My Item 1"
}
]
},
{
"_id": "5eb9564dc4317411fe79e1bf",
"name": "My Tag name 2",
"items": [
{
"id": "5eb95b1430f138131ed90f4f",
"name": "My Item 2"
},
{
"id": "5eb95ed0cae79713f1de0a29",
"name": "My Item 3"
}
]
}
]
}
What am I doing wrong?
One way to to this is to add the tag-name to your $group operator:
...
{
"$group": {
"_id": "$tags._id",
"tagName": {$first: "$tags.name"},
"items": {
"$push": {
"id": "$_id",
"name": "$name",
}
}
}
}
...
Note that this will take the first matching entry for each group. Another option would be to use $replaceRoot and the $mergeObjects operator, like it was done here.
Related
My mongo collection name tests and whose having the following documents in it.
[
{
"title": "One",
"uid": "1",
"_metadata": {
"references": [
{
"uid": "2"
},
{
"asssetuid": 10
}
]
}
},
{
"title": "Two",
"uid": "2",
"_metadata": {
"references": [
{
"uid": "3"
},
{
"asssetuid": 11
}
]
}
},
{
"title": "Three",
"uid": "3",
"_metadata": {
"references": []
}
}
]
And I want the result in the following format (for uid:1)
[
{
"title": "One",
"uid": 1,
"_metadata": {
"references": [
{
"asssetuid": 10
},
{
"asssetuid": 11
},
{
"title": "Two",
"uid": "2",
"_metadata": {
"references": [
{
"title": "Three",
"uid": "3"
}
]
}
}
]
}
}
]
for uid:2 I want the following result
[
{
"title": "Two",
"uid": 2,
"_metadata": {
"references": [
{
"asssetuid": 11
},
{
"title": "Three",
"uid": "3"
}
]
}
}
]
Which query I used here to get a respected result. according to its uid. here I want the result in the parent-child relationship. is this possible using MongoDB graph lookup query or any other query that we can use to get the result. Please help me with this.
New Type Output
[{
"title": "One",
"uid": 1,
"_metadata": {
"assets": [{
"asssetuid": 10,
"parent": 1
}, {
"asssetuid": 11,
"parent": 2
}],
"entries": [{
"title": "Two",
"uid": "2",
"parent": 1
}, {
"title": "Three",
"uid": "3",
"parent": 2
}]
}
}]
Mongo supports the automatic reference resolution using $ref but for that, you need to change your schema a little and resolve resolution is only supported by some drivers.
You need to store your data in this format:
[
...
{
"_id": ObjectId("5a934e000102030405000000"),
"_metadata": {
"references": [
{
"$ref": "collection",
"$id": ObjectId("5a934e000102030405000001"),
"$db": "database"
},
{
"asssetuid": 10
}
]
},
"title": "One",
"uid": "1"
},
....
]
For more details on $ref refer to official documentation: label-document-references
OR
you can resolve the reference using the $graphLookup but the only problem with the $graphlookup is that you will lose the assetuid. Here is the query and it will resolve references and give output in flat map
db.collection.aggregate([
{
$match: {
uid: "1"
}
},
{
$graphLookup: {
from: "collection",
startWith: "$_metadata.references.uid",
connectFromField: "_metadata.references.uid",
connectToField: "uid",
depthField: "depth",
as: "resolved"
}
},
{
"$addFields": {
"references": "$resolved",
"metadata": [
{
"_metadata": "$_metadata"
}
]
}
},
{
"$project": {
"references._metadata": 0,
}
},
{
"$project": {
"references": "$references",
"merged": {
"$concatArrays": [
"$metadata",
"$resolved"
]
}
}
},
{
"$project": {
results: [
{
merged: "$merged"
},
{
references: "$references"
}
]
}
},
{
"$unwind": "$results"
},
{
"$facet": {
"assest": [
{
"$match": {
"results.merged": {
"$exists": true
}
}
},
{
"$unwind": "$results.merged"
},
{
"$unwind": "$results.merged._metadata.references"
},
{
"$match": {
"results.merged._metadata.references.asssetuid": {
"$exists": true
}
}
},
{
"$project": {
_id: 0,
"asssetuid": "$results.merged._metadata.references.asssetuid"
}
}
],
"uid": [
{
"$match": {
"results.references": {
"$exists": true
}
}
},
{
"$unwind": "$results.references"
},
{
$replaceRoot: {
newRoot: "$results.references"
}
}
]
}
},
{
"$project": {
"references": {
"$concatArrays": [
"$assest",
"$uid"
]
}
}
}
])
Here is the link to the playground to test it: Mongo Playground
I'm trying to group my participants array in to such a way that for a single participant I should get all the meeting under that participant in an array.
[
{
"_id": "5fc73e7131e6a20f6c492178",
"participants": [
{
"_id": "5fc74bc5e7c54d0ea8f133ca",
"rsvp": "Yes",
"name": "Participant 1",
"email": "participant1#gmail.com"
},
{
"_id": "5fc74e1254b8d337b4ae36d2",
"rsvp": "Not Answered",
"name": "Participant 2",
"email": "participant2#gmail.com"
}
],
"title": "Meeting 1",
"startTime": "2020-11-01T18:30:00.000Z",
"endTime": "2020-11-03T18:30:00.000Z"
},
{
"_id": "5fc73f1cdfc45d3ca0c84654",
"participants": [
{
"_id": "5fc74bc5e7c54d0ea8f133ca",
"rsvp": "Yes",
"name": "Participant 2",
"email": "participant2#gmail.com"
}
],
"title": "Meeting 2",
"startTime": "2020-11-01T18:30:00.000Z",
"endTime": "2020-11-03T18:30:00.000Z"
}
]
my expected result should be like
[{
"participant": {
"_id": "5fc74bc5e7c54d0ea8f133ca",
"rsvp": "Yes",
"name": "Participant 1",
"email": "participant1#gmail.com"
},
meetings: [{meeting1, meeting2, ...and so on}]
},
{
"participant": {
"_id": "5fc74bc5e7c54d0ea8f133ca",
"rsvp": "Yes",
"name": "Participant 2",
"email": "participant2#gmail.com"
},
meetings: [{meeting2, meeting3, ...and so on}]
}
]
I'm kinda stuck for hours to figure it out. I tried the approach by using $group and $unwind but I was getting participants as an array consisting of a single participant(object). and on that I was unable to run $match to match according to the participant's email because participants field was an array.
I tried this
const docs = await Meeting.aggregate([
{ $unwind: '$participants' },
{
$lookup: {
from: 'participants',
localField: 'participants',
foreignField: '_id',
as: 'participants'
}
},
{ $match },
{ $group: { _id: "$participants", meetings: { $push: "$$ROOT" } } },
]);
but this is not matching the expected result which I want it to be.
You can unwind to deconstruct the array and use group to get your desired output
db.collection.aggregate([
{
"$unwind": "$participants"
},
{
$group: {
_id: "$participants._id",
participants: {
$first: "$participants"
},
meetings: {
"$addToSet": "$title"
}
}
}
])
Working Mongo playground
{
"_id": "5e28b029a0c8263a8a56980a",
"name": "Recruiter",
"data": [
{
"_id": "5e28b0980f89ba3c0782828f",
"targetLink": "https://www.linkedin.com/in/dan-kelsall-7aa0926b/",
"name": "Dan Kelsall",
"headline": "Content Marketing & Copywriting",
"actions": [
{
"result": 1,
"name": "VISIT"
},
{
"result": 1,
"name": "FOLLOW"
}
]
},
{
"_id": "5e28b0980f89ba3c078283426f",
"targetLink": "https://www.linkedin.com/in/56wergwer/",
"name": "56wergwer",
"headline": "asdgawehethre",
"actions": [
{
"result": 1,
"name": "VISIT"
}
]
}
]
}
Here is one of my mongodb document. I'd like to update data->actions->result
So this is what I've done
Campaign.updateOne({
'data.targetLink': "https://www.linkedin.com/in/dan-kelsall-7aa0926b/",
'data.actions.name': "Follow"
}, {$set: {'data.$.actions.result': 0}})
But it seems not updating anything and even it can't find the document by this 'data.actions.name'
You need the positional filtered operator since the regular positional operator ($) can only be used for one level of nested arrays:
Campaign.updateOne(
{ "_id": "5e28b029a0c8263a8a56980a", "data.targetLink": "https://www.linkedin.com/in/dan-kelsall-7aa0926b/" },
{ $set: { "data.$.actions.$[action].result": 0 } },
{ arrayFilters: [ { "action.name": "Follow" } ] }
)
I have searched online but could not find any match my case. Here is the situation.
I am using aggregate to combine one collection and one document which is from another collection together
restaurants.aggregate([
{
$match: {
_id: {
$in: idList
}
}
},
{
$lookup: {
from: "tags",
localField: "details.restaurantType",
foreignField: "details.restaurantType._id",
as: "types"
}
},
{
$project: {
restaurantName: "$details.restaurantName",
restaurantType: "$details.restaurantType",
type: {
$filter: {
input: "$types",
as: "type",
cond: {
$eq: ["$$type._id", "$details.restaurantType"]
}
}
},
currency: "$details.currency",
costPerPax: "$details.costPerPax"
}
}
]);
current result
The 'type' field in my current result is [], I need a matched value instead
[
{
"id": "5c20c7a0036dda80a8baabcc",
"restaurantName": "Villagio Restaurant Sutera Mall",
"type": [],
"currency": "RM",
"costPerPax": 22,
},
{
"id": "5c20ceb07715216d3c217b7a",
"restaurantName": "Thai Food Thai Now Sutera Mall",
"type": [],
"currency": "RM",
"costPerPax": 16,
}
]
expected result
I need the 'type' fields has match tag name from another collection like this
[
{
"id": "5c20c7a0036dda80a8baabcc",
"restaurantName": "Villagio Restaurant Sutera Mall",
"type": "Western",
"currency": "RM",
"costPerPax": 22,
},
{
"id": "5c20ceb07715216d3c217b7a",
"restaurantName": "Thai Food Thai Now Sutera Mall",
"type": "Thai",
"currency": "RM",
"costPerPax": 16,
}
]
Extra Information
two document from restaurants collection
{
"details": {
"restaurantName": "Villagio Restaurant Sutera Mall",
"restaurantType": "5c01fb57497a896d50f498a8"
},
"_id": "5c20c7a0036dda80a8baabcc",
"status": "OP",
"__v": 0
},
{
"details": {
"restaurantName": "Kingshahi Japanese Shop",
"restaurantType": "5c01fb57497a896d50f49879"
},
"_id": "5c20cb4fb7e75180480690c2",
"status": "OP",
"__v": 0
}
One document from tag collection
{
"_id": "5c01fb57497a896d50f49876",
"details": {
"restaurantTypeId": "5c01fb57497a896d50f49877",
"restaurantTypes": [
{
"_id": "5c01fb57497a896d50f49879",
"name": "Asian",
"counter": 1
},
{
"_id": "5c01fb57497a896d50f4987a",
"name": "Bakery",
"counter": 0
}
]
}
}
You can use below optimised aggregation pipeline
db.restaurants.aggregate([
{ "$lookup": {
"from": "tags",
"let": { "restaurantType": "$details.restaurantType" },
"pipeline": [
{ "$match": {
"$expr": { "$in": ["$$restaurantType", "$details.restaurantTypes._id"] }
}},
{ "$unwind": "$details.restaurantTypes" },
{ "$match": {
"$expr": { "$eq": ["$details.restaurantTypes._id", "$$restaurantType"] }
}}
],
"as": "types"
}},
{ "$project": {
"restaurantName": "$details.restaurantName",
"restaurantType": "$details.restaurantType",
"type": { "$arrayElemAt": ["$types.details.restaurantTypes.name", 0] },
"currency": "$details.currency",
"costPerPax": "$details.costPerPax"
}}
])
I have a simple datastructure in mongodb:
{
_id: ObjectID,
name: 'Name',
birthday: '25.05.2001'
items: [
{
_id: ObjectID,
name: 'ItemName',
info: 'ItemInfo',
},
{
_id: ObjectID,
name: 'ItemName',
info: 'ItemInfo',
}
]
}
Now i want a query, that takes a ObjectID (_id) of an item as criteria and gives me back the object with all items in the array AND projects a new field "selected" with value true or false into a field in the result of each array item:
I tried that with this query:
{ $unwind: '$items' },
{
$project: {
selected: {
$cond: { if: { 'items._id': itemObjectID }, then: true, else: false },
},
},
},
but MongoDB gives me back an error:
MongoError: FieldPath field names may not contain '.'.
Have no clue why its not working, any help or ideas? Thank you very much!
What you are missing here is $eq aggregation operator which checks the condition for the equality.
You can try below aggregation here if you want to check for ObjectId then you need to put mongoose.Types.ObjectId(_id)
db.collection.aggregate([
{ "$unwind": "$items" },
{ "$addFields": {
"items.selected": {
"$eq": [
1111,
"$items._id"
]
}
}},
{ "$group": {
"_id": "$_id",
"name": { "$first": "$name" },
"items": {
"$push": {
"_id": "$items._id",
"selected": "$items.selected"
}
}
}}
])
Will give following output
[
{
"_id": ObjectId("5a934e000102030405000000"),
"items": [
{
"_id": 1111,
"selected": true
},
{
"_id": 2222,
"selected": false
}
],
"name": "Name"
}
]
You can check it here
#Ashish: Thank you very much for your help! Your answer helped me to build the right query for me:
db.collection.aggregate([
{
$unwind: "$items"
},
{
$project: {
"items.name": 0,
"birthday": 0
}
},
{
"$addFields": {
"items.selected": {
"$eq": [
1111,
"$items._id"
]
}
}
},
{
$group: {
_id: "$_id",
"name": {
"$first": "$name"
},
items: {
$push: "$items"
}
}
},
{
$match: {
"items._id": {
$eq: 1111
}
}
},
])
and leads to a result that looks like:
[
{
"_id": ObjectId("5a934e000102030405000000"),
"items": [
{
"_id": 1111,
"selected": true
},
{
"_id": 2222,
"selected": false
}
],
"name": "Name"
}
]