I have a query like this one :
.findOne( {
_id: myObjectId,
}, {
fields: {
smallSingleValue: 1,
HUDGE_OBJECT: 1,
}
} );
I do not need the entire HUDGE_OBJECT (it is roughly 1 Mo), I need HUDGE_OBJECT[ smallSingleValue ] ( less than a Ko ).
Right now, I can make that request and get the entire HUDGE_OBJECT, or make two requests ; one the guet the smallSingleValue and the other one to get the HUDGE_OBJECT[ smallSingleValue ].
Both solution are crapy.
Is there a way to do something like that :
fields: {
smallSingleValue: 1,
`HUDGE_OBJECT.${ $smallSingleValue }`: 1,
}
Obviously not with that syntax, but you get the idea.
I tried aggregation, but it’s probably not the solution.
Does the projection operator allows it ?
https://docs.mongodb.com/manual/reference/operator/projection/positional/
EDIT FOR THE COMMENTS :
Example of data :
{
_id: xxx,
firstName: xxx,
lastName: xxx,
currentLevel: bar, // <- that one is important
progress: {
foo: { big object },
bar: { big object },
baz: { big object },
...
}
}
What I need :
{
firstName: xxx,
lastName: xxx,
currentLevel: bar,
progress: {
bar: { big object },
}
}
But the question is : performance wise, is it better to just get the entire object (easy query) or to get a truncated object (more complex query but passing a bigger object) ? There are 50 «levels».
I don’ remember what I tried with the aggregation :x but it seemed to be a bad idea.
Related
I have a situation where a model changed at some point in time and I am faced with (for argument sake) half my data liks like this
{
_id: OID,
things: [{
_id:OID,
arm: string,
body: string
}],
other: string
}
and the other half of my data look like this
{
_id: OID,
things: [{
_id:OID,
upper_appendage: string,
body: string
}],
other: string
}
I would like to 'correct' half of the data - so that I DON'T have to accommodate both names for 'arm' in my application code.
I have tried a couple different things:
The first errors
db.getCollection('x')
.find({things:{$exists:true}})
.forEach(function (record) {
record.things.arm = record.things.upper_appendage;
db.users.save(record);
});
and this - which destroys all the other data in
db.getCollection('x')
.find({things:{$exists:true}})
.forEach(function (record) {
record.things = {
upper_appendage.arm = record.things.upper_appendage
};
db.users.save(record);
});
Keeping in mind that there is other data I want to maintain...
How can I do this???
the $rename operator should have worked for this job but unfortunately it doesn't seem to support nested array fields (as of mongodb server 4.2). instead you'd need a forEach like the following:
db.items.find({
things: {
$elemMatch: {
arm: {
$exists: true
}
}
}
}).forEach(function(item) {
for (i = 0; i != item.things.length; ++i)
{
item.things[i].upper_appendage = item.things[i].arm;
delete item.things[i].arm; ;
}
db.items.update({
_id: item._id
}, item);
})
note: i've assumed you want to make all records have upper_appendageand get rid of 'arm' field. if it's the other way you want, just switch things around.
As part of my studies, we recently started learning MongoDB, and so I am very young in the field.
In one of the tasks we have been assigned to, MapReduce is used, but I can not do it properly.
Giving a collection of students of the following form:
{
_id: ObjectId("4ffmdjd8cfd99k03"),
tz: "11111",
FirstName: "Anton",
LastName: "Gill",
Department: "Computer Science"
Year: 1
Courses: [ { name: "Linear Algebra 1", grade: 70}, {name: "OS", grade: 88}]
}
The task is to write a mapReduce function, so that it returns a list of the names of the students in each department and each year, all whose grades are greater than 90. Example of normal output:
[{Department: "Computer Science", Year: 1, students: ["Anton", "John"]},
{Department: "Physics", Year: 2, students: ["Dean", "David"]}]
Can I write more than one map function? That is, the mapReduce structure will look like this:
db.students.mapReduce(
map1,
map2,
map3,
..
reduce,
..
I try unsuccessfully to create the desired structure.
This is the best I've been able to get to, and I'm still not sure how to write the reduce function.
var map = function(){
var value = 0, n = this.Courses.length;
for(var i = 0; i < n; i++){
value += this.Courses[i].grade;
}
var avgGrade = value/n;
if(avgGrade > 90){
var key = {
tz:this.tz,
Department:this.Department,
Year:this.Year,
};
value = {students:[this.FirstName]};
}
emit(key, value);
};
var reduce = function(keysObj, valuesObj){
return 1000; //Ignore this function, I've no idea how to deal with it.
};
db.students.mapReduce(
map,
reduce,
{
out: "output",
}
)
I would highly appreciate any assistance :)
So I have found quite few related posts on SO on how to update a field in a sub array, such as this one here
What I want to achieve is basically the same thing, but updating a field in a subarray dynamically, instead of just calling the field name in the query.
Now I also found how to do that straight in the main object, but cant seem to do it in the sub array.
Code to insert dynamically in sub-object:
_.each(data.data, function(val, key) {
var obj = {};
obj['general.'+key] = val;
insert = 0 || (Documents.update(
{ _id: data._id },
{ $set: obj}
));
});
Here is the tree of what I am trying to do:
Documents: {
_id: '123123'
...
smallRoom:
[
_id: '456456'
name: 'name1'
description: 'description1'
],
[
...
]
}
Here is my code:
// insert a new object in smallRoom, with only the _id so far
var newID = new Mongo.ObjectID;
var createId = {_id: newID._str};
Documents.update({_id: data._id},{$push:{smallRooms: createId}})
And the part to insert the other fields:
_.each(data.data, function(val, key) {
var obj = {};
obj['simpleRoom.$'+key] = val;
console.log(Documents.update(
{
_id: data._id, <<== the document id that I want to update
smallRoom: {
$elemMatch:{
_id : newID._str, <<== the smallRoom id that I want to update
}
}
},
{
$set: obj
}
));
});
Ok, having said that, I understand I can insert the whole object straight away, not having to push every single field.
But I guess this question is more like, how does it work if smallRoom had 50 fields, and I want to update 3 random fields ? (So I would NEED to use the _each loop as I wouldnt know in advance which field to update, and would not want to replace the whole object)
I'm not sure I 100% understand your question, but I think the answer to what you are asking is to use the $ symbol.
Example:
Documents.update(
{
_id: data._id, smallRoom._id: newID._str
},
{
$set: { smallroom.$.name: 'new name' }
}
);
You are finding the document that matches the _id: data._id, then finding the object in the array smallRoom that has an _id equal to newId._str. Then you are using the $ sign to tell Mongo to update that object's name key.
Hope that helps
I have a Collection that has documents with an array of nested objects.
Here is fixture code to populate the database:
if (Parents.find().count() == 0) {
var parentId = Parents.insert({
name: "Parent One"
});
Children.insert({
parent: parentId,
fields: [
{
_id: new Meteor.Collection.ObjectID(),
position: 3,
name: "three"
},
{
_id: new Meteor.Collection.ObjectID(),
position: 1,
name: "one"
},
{
_id: new Meteor.Collection.ObjectID(),
position: 2,
name: "two"
},
]
});
}
You might be asking yourself, why do I even need an ObjectID when I can just update based off of the names. This is a simplified example to a much more complex schema that I'm currently working on and the the nested object are going to be created dynamically, the ObjectID's are definitely going to be necessary to make this work.
Basically, I need a way to save those nested objects with a unique ID and be able to update the fields by their _id.
Here is my Method, and the call I'm making from the browser console:
Meteor.methods({
upChild: function( options ) {
console.log(new Meteor.Collection.ObjectID());
Children.update({_id: options._id, "fields._id": options.fieldId }, {$set: {"fields.$.position": options.position}}, function(error){
if(error) {
console.log(error);
} else {
console.log("success");
}
});
}
});
My call from the console:
Meteor.call('upChild', {
_id: "5NuiSNQdNcZwau92M",
fieldId: "9b93aa1ef3868d762b84d2f2",
position: 1
});
And here is a screenshot of the html where I'm rendering all of the data for the Parents and Children collections:
Just an observation, as I was looking how generate unique IDs client side for a similar reason. I found calling new Meteor.Collection.ObjectID() was returning a object in the form 'ObjectID("abc...")'. By assigning Meteor.Collection.ObjectID()._str to _id, I got string as 'abc...' instead, which is what I wanted.
I hope this helps, and I'd be curious to know if anyone has a better way of handling this?
Jason
Avoid using the _str because it can change in the future. Use this:
new Meteor.Collection.ObjectID().toHexString() or
new Meteor.Collection.ObjectID().valueOf()
You can also use the official random package:
Random.id()
I am trying to insert nested documents in to a MongoDB using C#. I have a collection called categories. In that collection there must exist documents with 2 array, one named categories and one named standards. Inside those arrays must exist new documents with their own ID's that also contain arrays of the same names listed above. Below is what I have so far but I am unsure how to proceed. If you look at the code what I want to do is add the "namingConventions" document nested under the categories array in the categories document however namingConventions must have a unique ID also.
At this point I am not sure I have done any of this the best way possible so I am open to any and all advice on this entire thing.
namespace ClassLibrary1
{
using MongoDB.Bson;
using MongoDB.Driver;
public class Class1
{
public void test()
{
string connectionString = "mongodb://localhost";
MongoServer server = MongoServer.Create(connectionString);
MongoDatabase standards = server.GetDatabase("Standards");
MongoCollection<BsonDocument> categories = standards.GetCollection<BsonDocument>("catagories");
BsonDocument[] batch = {
new BsonDocument { { "categories", new BsonArray {} },
{ "standards", new BsonArray { } } },
new BsonDocument { { "catagories", new BsonArray { } },
{ "standards", new BsonArray { } } },
};
categories.InsertBatch(batch);
((BsonArray)batch[0]["categories"]).Add(batch[1]);
categories.Save(batch[0]);
}
}
}
For clarity this is what I need:
What I am doing is building a coding standards site. The company wants all the standards stored in MongoDB in a tree. Everything must have a unique ID so that on top of being queried as a tree it can be queried by itself also. An example could be:
/* 0 */
{
"_id" : ObjectId("4fb39795b74861183c713807"),
"catagories" : [],
"standards" : []
}
/* 1 */
{
"_id" : ObjectId("4fb39795b74861183c713806"),
"categories" : [{
"_id" : ObjectId("4fb39795b74861183c713807"),
"catagories" : [],
"standards" : []
}],
"standards" : []
}
Now I have written code to make this happen but the issue seems to be that when I add object "0" to the categories array in object "1" it is not making a reference but instead copying it. This will not due because if changes are made they will be made to the original object "0" so they will not be pushed to the copy being made in the categories array, at least that is what is happening to me. I hope this clears up what I am looking for.
So, based on your latest comment, it seems as though this is the actual structure you are looking for:
{
_id: ObjectId(),
name: "NamingConventions",
categories: [
{
id: ObjectId(),
name: "Namespaces",
standards: [
{
id: ObjectId(),
name: "TitleCased",
description: "Namespaces must be Title Cased."
},
{
id: ObjectId().
name: "NoAbbreviations",
description: "Namespaces must not use abbreviations."
}
]
},
{
id: ObjectId(),
name: "Variables",
standards: [
{
id: ObjectId(),
name: "CamelCased",
description: "variables must be camel cased."
}
]
}
]
}
Assuming this is correct, then the below is how you would insert one of these:
var collection = db.GetCollection("some collection name");
var root = new BsonDocument();
root.Add("name", "NamingConventions");
var rootCategories = new BsonArray();
rootCategories.Add(new BsonDocument
{
{ "id": ObjectId.GenerateNewId() },
{ "name", "Namespaces" },
{ "standards", new BsonArray() }
});
root.Add("categories", rootCategories);
//etc...
collection.Save(root);
Hope that helps, if not, I give up :).
So, I guess I'm confused by what you are asking. If you just want to store the namingConventions documents inside the array, you don't need a collection for them. Instead, just add them to the bson array and store them.
var categoriesCollection = db.GetCollection<BsonDocument>("categories");
var category = new BsonDocument();
var namingConventions = new BsonArray();
namingConventions.Add(new BsonDocument("convention1", "value"));
category.Add("naming_conventions", namingConventions);
categoriesCollection.Insert(category);
This will create a new document for a category, create an array in it called naming_conventions with a single document in it with an element called "convention1" and a value of "value".
I also am not quite sure what you are trying to accomplish. Perhaps if you posted some sample documents in JSON format we could show you the C# code to write documents that match that.
Alternatively, if you wish to discuss your schema, that could also be better done in the context of JSON rather than C#, and once a schema has been settled on then we can discuss how to write documents to that schema in C#.
One thing that didn't sound right in your original description was the statement "in that collection must exist 2 arrays". A collection can only contain documents, not arrays. The documents themselves can contain arrays if you want.
var filter = Builders<CollectionDefination>.Filter.Where(r => r._id== id);
var data = Builders<CollectionDefination>.Update.Push(f=>
f.categories,categoriesObject);
await _dbService.collection.UpdateOneAsync( filter,data);
Note: Make sure embedded document type [Categories] is array.