There is a document, i.e.
{
"_id": {
"$oid": "63ee577ca5340cd594916852"
},
"id": 12345,
"price": 123,
"oldprice": 456
}
I am performing an update with
db.testupd.updateOne({'id': 12345}, [{'$set': {"id": 12345, "price": 222}}], upsert=true)
It works but "oldprice" field is still there after update and what I need is to delete the fields that no longer exist, because unfortunately the data source is not consistent.
How can I achieve this?
If the list of fields that you do not want to keep is unknown, using replace in $merge is one of the options. You can $project to keep only the fields you want.
db.collection.aggregate([
{
$match: {
"id": 12345
}
},
{
"$set": {
"id": 12345,
"price": 222
}
},
{
$project: {
id: 1,
price: 1
}
},
{
"$merge": {
"into": "collection",
"on": "_id",
"whenMatched": "replace"
}
}
])
Mongo Playgroud
Related
I have 2 collection in mongodb: Account, Information.
Account:
{
"_id": {
"$oid": "6348dc197a7b552660170d8b"
},
"username": "12345",
"password": "123dsgfdsgdfsg",
"email": "1243",
"role": "123",
"_infoid": {
"$oid": "6348dc197a7b552660170d8a"
}
}
Information:
{
"_id": {
"$oid": "6348dc197a7b552660170d8a"
},
"avatar": "hello",
"name": "Abcd",
"phonenumber": "012345678",
"address": "abcd"
}
I wanna change "phonenumber" to "123" but I just have "_id" of Account. Can I change it with aggregation pipeline?
Does this seem what you try to achieve?
// populate database with test data
use("test_db")
db.account.insertOne({
"_id": "6348dc197a7b552660170d8b",
"username": "12345",
"password": "123dsgfdsgdfsg",
"email": "1243",
"role": "123",
"_infoid": "6348dc197a7b552660170d8a"
})
db.information.insertOne({
"_id": "6348dc197a7b552660170d8a",
"avatar": "hello",
"name": "Abcd",
"phonenumber": "012345678",
"address": "abcd"
})
// define some test variables to use
let some_account_id = "6348dc197a7b552660170d8b"
let new_phone_number = "+9876543210"
// change data
db.account.aggregate([
{
$match: {_id: some_account_id}
},
{
$addFields: {phonenumber: new_phone_number}
},
{
$project: {phonenumber: 1, _id: "$_infoid"}
},
{
$merge:{
into: "information",
whenNotMatched: "fail",
}
}
])
// show final results
db.information.find()
Result:
[
{
"_id": "6348dc197a7b552660170d8a",
"avatar": "hello",
"name": "Abcd",
"phonenumber": "+9876543210",
"address": "abcd"
}
]
You don't need to duplicate _id, findOneAndUpdate() can be executed for this use case.
The definition of it is:
db.collection.findOneAndUpdate( filter, update, options )
which updates a single document based on the filter and sort criteria.
Below you can refer to the link for more details:
https://www.mongodb.com/docs/manual/reference/method/db.collection.findOneAndUpdate/
.
Performa a $lookup and perform some wrangling to your desired format. Finally do a $merge to update to the collection Information
db.Information.aggregate([
{
"$lookup": {
"from": "Account",
"localField": "_id",
"foreignField": "_infoid",
"pipeline": [
{
"$project": {
role: 1
}
}
],
"as": "AccountLookup"
}
},
{
"$unwind": "$AccountLookup"
},
{
$set: {
phonenumber: "$AccountLookup.role"
}
},
{
$unset: "AccountLookup"
},
{
"$merge": {
"into": "Information",
"on": "_id",
"whenMatched": "merge",
"whenNotMatched": "discard"
}
}
])
Here is the Mongo Playground for your reference.
I have 3 collections.
User:
{
"_id":ObjectId("60a495cdd4ba8b122899d415"),
"email":"br9#gmail.com",
"username":"borhan"
}
Panel:
{
"_id": ObjectId("60a495cdd4ba8b122899d417"),
"name": "borhan",
"users": [
{
"role": "admin",
"joined": "2021-05-19T04:35:47.474Z",
"status": "active",
"_id": ObjectId("60a495cdd4ba8b122899d418"),
"user": ObjectId("60a495cdd4ba8b122899d415")
},
{
"role": "member",
"joined": "2021-05-19T04:35:47.474Z",
"status": "active",
"_id": ObjectId("60a49600d4ba8b122899d41a"),
"user": ObjectId("60a34e167958972d7ce6f966")
}
],
}
Team:
{
"_id":ObjectId("60a495e0d4ba8b122899d419"),
"title":"New Teams",
"users":[
ObjectId("60a495cdd4ba8b122899d415")
],
"panel":ObjectId("60a495cdd4ba8b122899d417")
}
I want to receive a output from querying Panel colllection just like this:
{
"_id": ObjectId("60a495cdd4ba8b122899d417"),
"name": "borhan",
"users": [
{
"role": "admin",
"joined": "2021-05-19T04:35:47.474Z",
"status": "active",
"_id": ObjectId("60a495cdd4ba8b122899d418"),
"user": ObjectId("60a495cdd4ba8b122899d415"),
"teams":[
{
"_id":ObjectId("60a495e0d4ba8b122899d419"),
"title":"New Teams",
"users":[
ObjectId("60a495cdd4ba8b122899d415")
],
"panel":ObjectId("60a495cdd4ba8b122899d417")
}
]
},
{
"role": "member",
"joined": "2021-05-19T04:35:47.474Z",
"status": "active",
"_id": ObjectId("60a49600d4ba8b122899d41a"),
"user": ObjectId("60a34e167958972d7ce6f966")
}
],
}
I mean i want to add teams field (which is array of teams that user is existed on it) to each user in Panel collection
Here is my match query in mongoose to select specific panel:
panel_model.aggregate([
{
$match: {
users: {
$elemMatch: {user: ObjectId("60a495cdd4ba8b122899d415"), role:"admin"}
}
}
},
])
Is it possible to get my output with $lookup or $addFields aggregations?
You need to join all three collections,
$unwind to deconstruct the array
$lookup there are two kind of lookups which help to join collections. First I used Multiple-join-conditions-with--lookup, and I used standrad lookup to join Users and Teams collections.
$match to match the user's id
$expr - when you use $match inside lookup, u must use it.
$set to add new fields
$group to we already destructed using $unwind. No we need to restructure it
here is the code
db.Panel.aggregate([
{ $unwind: "$users" },
{
"$lookup": {
"from": "User",
"let": { uId: "$users.user" },
"pipeline": [
{
$match: {
$expr: {
$eq: [ "$_id", "$$uId" ]
}
}
},
{
"$lookup": {
"from": "Team",
"localField": "_id",
"foreignField": "users",
"as": "teams"
}
}
],
"as": "users.join"
}
},
{
"$set": {
"users.getFirstElem": {
"$arrayElemAt": [ "$users.join", 0 ]
}
}
},
{
$set: {
"users.teams": "$users.getFirstElem.teams",
"users.join": "$$REMOVE",
"users.getFirstElem": "$$REMOVE"
}
},
{
"$group": {
"_id": "$_id",
"name": { "$first": "name" },
"users": { $push: "$users" }
}
}
])
Working Mongo playground
Note : Hope the panel and user collections are in 1-1 relationship. Otherwise let me know
I have few referenced collections: Employees, Roles, Schools...
I need to query the database, to list all employees that are employed by a certain school. That means all Employees, that has certain School ID in their corresponding role document must be returned in the list or array of results.
So far I have tried to find it like this:
const employees = mongoose.schema("Employees");
employees.find({})
.populate({
"path": "roles"
})
.then(function(docs){
docs.filter(function(){
// write my filters here....
})
});
but this still is inefficient and I can't make it work.
It has to be a smarter way...
This is my Employee document, which references a document in Roles Collection:
{
"_id" : { "$oid" : "57027f1b9522d363243abr42" },
"assignedRoles": [
{
"type": "teacher",
"schools": [
{
"schoolId": "57027f1b9522d3632423rf32",
"classes" : ["math", "science"],
"active": true
},
{
"schoolId": "57027f1b9522d36324252fs2",
"classes" : ["science"],
"active": true
},
{
"schoolId": "57027f1b9522d36324242f35",
"classes" : ["math"],
"active": true
},
]
},
{
"type": "manager",
"schools": [
{
"schoolId": "57027f1b9522d3632423rf32",
"active": true
},
{
"schoolId": "57027f1b9522d36324252fs2",
"active": true
}
]
}
{
"type": "principal",
"schools": [
{
"schoolId": "57027f1b9522d3632423rf32",
"active": true
}
]
}
],
"rolesMeta": "Meta Info for the Roles"
}
The following is the list of Schools - irrelevant for the task, I am just adding that for the completion:
{
"_id": { "$oid" : "57027f1b9522d3632423rf32" },
"name": "G.Washington HS",
"district": "six",
"state": "New Mexico"
},
{
"_id": { "$oid" : "57027f1b9522d36324252fs2" },
"name": "A. Lincoln HS",
"district": "six",
"state": "New Mexico"
},
{
"_id": { "$oid" : "57027f1b9522d36324242f35" },
"name": "T. Roosvelt HS",
"district": "four",
"state": "New Mexico"
}
The Resolution to the problem is to use MongoDB Aggregate framework to be able to drill down and "flatten" the array structure.
Here is how I have resolved it:
db.employees.aggregate([
// I use $project to specify which field I want selected from the collection
{
$project: {_id: 1, "roles": 1}
},
// $lookup helps me get results and populate the 'role' field from the Roles collection. This returns an array
{
$lookup: {
"from": "roles",
"localField": "roles",
"foreignField": "_id",
"as": "roles"
}
},
// $unwind is self explanatory - it flattens the array and creates multiple objects, instances of the parent object for each array element.
{
$unwind: "$roles"
},
// now repeat for each array in this document
{
$unwind: "$roles.assignedRoles"
},
// repeat again, since I need to look in the schools array.
{
$unwind: "$roles.assignedRoles.schools"
},
// The following line will return all records that the school ID checks out.
{
$match: {"roles.assignedRoles.schools.schoolId": school}
},
// $group is a necessary cleanup task so we avoid repeating employee records... Nicely Done! :)
{
$group: {
"_id": "$_id",
"roles": {"$push": "$roles"},
}
}
])
.exec()
.then(function (docs) {
// here you process the resulted docs.
});
I have to sort the final results based on the lookup result. Below is my aggregate query:
{ $match : {status:'active'},
{ $limit : 10},
{ $lookup:
{
from : "metas",
localField : "_id",
foreignField: "post_id",
as : "meta"
}
}
This query produce results as:
{
"_id": "594b6adc2a8c4f294025e46e",
"title": "Test 1",
"created_at": "2017-06-22T06:59:40.809Z",
"meta": [
{
"_id": "594b6b072a8c4f294025e46f",
"post_id": "594b6adc2a8c4f294025e46e",
"views": 1,
},
{
"_id": "594b6b1c2a8c4f294025e471",
"post_id": "594b6adc2a8c4f294025e46e",
}
],
},
{
"_id": "594b6adc2a8c4f29402f465",
"title": "Test 2",
"created_at": "2017-06-22T06:59:40.809Z",
"meta": [
{
"_id": "594b6b072a8c4f294025e46f",
"post_id": "594b6adc2a8c4f29402f465",
"views": 0,
},
{
"_id": "594b6b1c2a8c4f294025e471",
"post_id": "594b6adc2a8c4f29402f465",
}
],
},
{
"_id": "594b6adc2a8c4f29856d442",
"title": "Test 3",
"created_at": "2017-06-22T06:59:40.809Z",
"meta": [
{
"_id": "594b6b072a8c4f294025e46f",
"post_id": "594b6adc2a8c4f29856d442",
"views": 3,
},
{
"_id": "594b6b1c2a8c4f294025e471",
"post_id": "594b6adc2a8c4f29856d442",
}
],
}
Now what I want here is to sort these results based on 'views' under 'meta'. Like result will be list in descending order of 'meta.views'. First result will be meta with views=3, then views=1 and then views=0
$unwind operator splits an array into seperate documents for each object contained in an array
For eg
db.collection.aggregate(
// Pipeline
[
// Stage 1
{
$unwind: {
path : "$meta"
}
},
// Stage 2
{
$sort: {
'meta.views':-1
}
},
]
);
Although $lookup does not support sorting, the easiest solution I think, and probably also the fastest, is to create a proper index on the related collection.
In this case, an index on the metas collection on the foreign field post_id and the field on which sorting is wanted views. Make sure to make the index in the correct sorting order.
Not only is the result now sorted, the query is probably also faster now it can use an index.
Assuming I have the following JSON structure I want to group by gender and want to return multiple document values on in the same field:
[
{
"id": 0,
"age": 40,
"name": "Tony Bond",
"gender": "male"
},
{
"id": 1,
"age": 30,
"name": "Nikki Douglas",
"gender": "female"
},
{
"id": 2,
"age": 23,
"name": "Kasey Cardenas",
"gender": "female"
},
{
"id": 3,
"age": 25,
"name": "Latasha Burt",
"gender": "female"
}
]
Now I know I can do something like this but I need to join both age and name into one field
.aggregate().group({ _id:'$gender', age: { $addToSet: "$age" }, name: { $addToSet: "$name"}})
Yes you can, just have a sub-document as the argument:
db.collection.aggregate([
{ "$group": {
"_id": "$gender",
"person": { "$addToSet": { "name": "$name", "age": "$age" } }
}}
])
Of course if you actually expect no duplicates here the $push does the same thing:
db.collection.aggregate([
{ "$group": {
"_id": "$gender",
"person": { "$push": { "name": "$name", "age": "$age" } }
}}
])
In addition to what Neil mentioned for the aggregation query, you have to consider the fact that the document that contains the grouped individual record cannot be beyond 16 MB (in v2.5+) or the entire aggregate result cannot be more than 16MB (on v2.4 and below).
So in case you have huge data set, you have to be aware of limitations that may impose on the model you use to aggregate data.