Just a note I am fairly new to mongo and more notably very new to using node/js.
I'm trying to write a query to insert new documents or update already existing documents in my collection.
The proposed structure of the collection is:
{ _id: xxxxxxx, ip: "xxx.xxx.xxx.xxx:xxxxxx", date: "xx-xx-xx xxxx" }
Note that my intention is a store an fixed length int for the _id rather than the internal ObjectId (is this possible/considered bad practice?). The int is guaranteed to be unique and comes from another source.
var monk = require('monk');
var db = monk('localhost:27017/cgo_schedule');
var insertDocuments = function(db, match) {
var db = db;
var collection = db.get('cgo_schedule');
collection.findAndModify(
{
"query": { "_id": match.matchId },
"update": { "$set": {
"ip": match.ip,
"date": match.date
},
"$setOnInsert": {
"_id": match.matchId,
}},
"options": { "new": true, "upsert": true }
},
function(err,doc) {
if (err) throw err;
console.log( doc );
}
);
}
This doesn't work at all however. It doesn't insert anything to the database, but it also gives no errors, so I have no idea what I'm doing wrong.
The output (for console.log (doc)) is null.
What am I doing wrong?
The Monk docs aren't much help, but according to the source code, the options object must be provided as a separate parameter.
So your call should look like this instead:
collection.findAndModify(
{
"query": { "_id": match.matchId },
"update": {
"$set": {
"ip": match.ip,
"date": match.date
}
}
},
{ "new": true, "upsert": true },
function(err,doc) {
if (err) throw err;
console.log( doc );
}
);
Note that I removed the $setOnInsert part as the _id is always included on insert with an upsert.
Related
I am doing currently doing as
if (part.qty) {
part.status = 'In Stock';
}
const part = new partsModel(part);
return part.save();
is it possible do this in MongoDB query only without using if statement both insert and update
You can try something like:
const filterObj = {
"$or": [{
"qty": {
"$exists": true
}
}, {
"qty": {
"$ne": null
}
}, {
"qty": {
"$ne": 0 //based on what type of value you have in qty
}
}]
};
const updateObj = {
"$set": {
"status": "In Stock"
}
};
partsModel.update(filterObj, updateObj)
.then(result => {
console.log(result);
})
.catch(err => {
console.log(err);
});
This will update the first matching record with our filters. If you wanna update all matching records use updateMany() instead.
Check out the official mongoose docs for more details.
P.s: You may need some modifications as i couldn't get a chance to try the code. hope this helps :)
I want to update a array value but i am not sure about the proper method to do it ,so for i tried following method but didnt worked for me.
My model,
The children field in my model
childrens: {
type: Array,
default: ''
}
My query,
Employeehierarchy.update({ _id: employeeparent._id} ,{ $set: {"$push": { "childrens": employee._id }} })
.exec(function (err, managerparent) {});
Can anyone please provide me help.Thanks.
You can't use both $set and $push in the same update expression as nested operators.
The correct syntax for using the update operators follows:
{
<operator1>: { <field1>: <value1>, ... },
<operator2>: { <field2>: <value2>, ... },
...
}
where <operator1>, <operator2> can be from any of the update operators list specified here.
For adding a new element to the array, a single $push operator will suffice e.g. you can use the findByIdAndUpdate update method to return the modified document as
Employeehierarchy.findByIdAndUpdate(employeeparent._id,
{ "$push": { "childrens": employee._id } },
{ "new": true, "upsert": true },
function (err, managerparent) {
if (err) throw err;
console.log(managerparent);
}
);
Using your original update() method, the syntax is
Employeehierarchy.update(
{ "_id": employeeparent._id},
{ "$push": { "childrens": employee._id } },
function (err, raw) {
if (err) return handleError(err);
console.log('The raw response from Mongo was ', raw);
}
);
in which the callback function receives the arguments (err, raw) where
err is the error if any occurred
raw is the full response from Mongo
Since you want to check the modified document, I'd suggest you use the findByIdAndUpdate function since the update() method won't give you the modified document, just the full write result from mongo.
If you want to update a field in the document and add an element to an array at the same time then you can do
Employeehierarchy.findByIdAndUpdate(employeeparent._id,
{
"$set": { "name": "foo" },
"$push": { "childrens": employee._id }
}
{ "new": true, "upsert": true },
function (err, managerparent) {
if (err) throw err;
console.log(managerparent);
}
);
The above will update the name field to "foo" and add the employee id to the childrens array.
can follow this
if childrens contains string values then model can be like:
childrens: [{
type : String
}]
if childrens contains ObjectId values of another collection _id and want populate then model can be like:
childrens: [{
type : mongoose.Schema.Types.ObjectId,
ref: 'refModelName'
}]
no need to use $set just use $push to insert value in childrens array. so query can be like:
Employeehierarchy.update(
{ _id: employeeparent._id},
{"$push": { "childrens": employee._id } }
).exec(function (err, managerparent) {
//
});
This will help I guess
Employeehierarchy.findOneAndUpdate(
{ _id:employeeparent._id },
{ $set: { "childrens": employee._id }}
)
I want to update multiple docs with different values.
My Database looks something like this.
[
{
"_id": 1,
"value": 50
},
{
"_id": 2,
"value": 100
}
]
This Query return an error because i'm passing an array instead of an object in the $set.
Model.update({_id: {$in: ids}}, {$set: ids.value}, {multi: true};
I want my database to look like this
[
{
"_id": 1,
"value": 4
},
{
"_id": 2,
"value": 27
}
]
Supposing you had an array of objects that you wanted to update in your collection on matching ids like
var soldItems = [
{
"_id": 1,
"value": 4
},
{
"_id": 2,
"value": 27
}
];
then you could use the forEach() method on the array to iterate it and update your collection:
soldItems.forEach(function(item)){
Model.update({"_id": item._id}, {"$set": {"value": item.value }}, callback);
});
or use promises as
var updates = [];
soldItems.forEach(function(item)){
var updatePromise = Model.update({"_id": item._id}, {"$set": {"value": item.value }});
updates.push(updatePromise);
});
Promise.all(updates).then(function(results){
console.log(results);
});
or using map()
var updates = soldItems.map(function(item)){
return Model.update({"_id": item._id}, {"$set": {"value": item.value }});
});
Promise.all(updates).then(function(results){
console.log(results);
});
For larger arrays, you could take advantage of using a bulk write API for better performance. For Mongoose versions >=4.3.0 which support MongoDB Server 3.2.x,
you can use bulkWrite() for updates. The following example shows how you can go about this:
var bulkUpdateCallback = function(err, r){
console.log(r.matchedCount);
console.log(r.modifiedCount);
}
// Initialise the bulk operations array
var bulkOps = soldItems.map(function (item) {
return {
"updateOne": {
"filter": { "_id": item._id } ,
"update": { "$set": { "value": item.value } }
}
}
});
// Get the underlying collection via the native node.js driver collection object
Model.collection.bulkWrite(bulkOps, { "ordered": true, w: 1 }, bulkUpdateCallback);
For Mongoose versions ~3.8.8, ~3.8.22, 4.x which support MongoDB Server >=2.6.x, you could use the Bulk API as follows
var bulk = Model.collection.initializeOrderedBulkOp(),
counter = 0;
soldItems.forEach(function(item) {
bulk.find({ "_id": item._id }).updateOne({
"$set": { "value": item.value }
});
counter++;
if (counter % 500 == 0) {
bulk.execute(function(err, r) {
// do something with the result
bulk = Model.collection.initializeOrderedBulkOp();
counter = 0;
});
}
});
// Catch any docs in the queue under or over the 500's
if (counter > 0) {
bulk.execute(function(err,result) {
// do something with the result here
});
}
First of all your update() query is not ok..
See documentation for this here
Model.update(conditions, update, options, callback);
Suppose your current db model contains docs like this (as you described in question as well):
[
{
"_id": 1,
"value": 50
},
{
"_id": 2,
"value": 100
}
];
and you've below array which contains objects (i.e., docs) to be modified with current db's docs to like this:
idsArray: [
{
"_id": 1,
"value": 4
},
{
"_id": 2,
"value": 27
}
];
From my experience with mongodb & mongoose, i don't think you can update all docs with single line query (that's you're trying to do).. (P.S. I don't know about that so I am not sure to this..)
But to make your code work, you will be doing something like this:
Idea: Loop over each doc in docs i.e, idsArray and call update() over it..
So, Here's code to this:
idsArray.forEach(function(obj) {
Model.update({_id: obj._id}, {$set: {value: obj.value}});
});
In above code, I am supposing you've _id values in db docs as they 're written above (i.e, "_id": 1).. But if they're like this "_id": ObjectId('1')
[
{
"_id": ObjectId('1'),
"value": 50
},
.....
.....
]
then you'll be need to convert _id to ObjectId(obj._id) in update() query..
so for that you'll be doing like this.
var ObjectId = require('mongodb').ObjectID;
idsArray.forEach(function(obj) {
Model.update({_id: ObjectId(obj._id)}, {$set: {value: obj.value}});
});
P.S. Just confirm it (i.e., _id) before go forward..
Hope this helps.
Cheers,
Multi update can be used only for updating multiple documents to the same value(s) or updating with the same update operation for all documents.
If you want to update to different values you have to use several update statements.
I am using the below code to insert data to mongodb
router.post('/NewStory', function (req, res) {
var currentObject = { user: userId , story : story , _id:new ObjectID().toHexString() };
req.db.get('clnTemple').findAndModify({
query: { _id: req.body.postId },
update: { $addToSet: { Stories: currentObject } },
upsert: true
});
});
This code is working fine if i remove the _id:new ObjectID().toHexString()
What i want to achieve here is that for every new story i want a unique _id object to be attached to it
What am i doing wrong?
{
"_id": {
"$oid": "55ae24016fb73f6ac7c2d640"
},
"Name": "some name",
...... some other details
"Stories": [
{
"userId": "105304831528398207103",
"story": "some story"
},
{
"userId": "105304831528398207103",
"story": "some story"
}
]
}
This is the document model, the _id that i am trying to create is for the stories
You should not be calling .toHexString() on this as you would be getting a "string" and not an ObjectID. A string takes more space than the bytes of an ObjectId.
var async = require('async'),
mongo = require('mongodb'),
db = require('monk')('localhost/test'),
ObjectID = mongo.ObjectID;
var coll = db.get('junk');
var obj = { "_id": new ObjectID(), "name": "Bill" };
coll.findAndModify(
{ "_id": new ObjectID() },
{ "$addToSet": { "stories": obj } },
{
"upsert": true,
"new": true
},
function(err,doc) {
if (err) throw err;
console.log(doc);
}
)
So that works perfectly for me. Noting the "new" option there as well so the modified document is returned, rather than the original form of the document which is the default.
{ _id: 55c04b5b52d0ec940694f819,
stories: [ { _id: 55c04b5b52d0ec940694f818, name: 'Bill' } ] }
There is however a catch here, and that is that if you are using $addToSet and generating a new ObjectId for every item, then that new ObjectId makes everything "unique". So you would keep adding things into the "set". This may as well be $push if that is what you want to do.
So if userId and story in combination already make this "unique", then do this way instead:
coll.findAndModify(
{
"_id": docId,
"stories": {
"$not": { "$elemMatch": { "userId": userId, "story": story } }
}
},
{ "$push": {
"stories": {
"userId": userId, "story": story, "_id": new ObjectID()
}
}},
{
"new": true
},
function(err,doc) {
if (err) throw err;
console.log(doc);
}
)
So test for the presence of the unique elements in the array, and where they do not exist then append them to the array. Also noting there that you cannot do an "inequality match" on the array element while mixing with "upserts". Your test to "upsert" the document should be on the primary "_id" value only. Managing array entries and document "upserts" need to be in separate update operations. Do not try an mix the two, otherwise you will end up creating new documents when you did not intend to.
By the way, you can generate an ObjectID just using monk.
var db = monk(credentials.database);
var ObjectID = db.helper.id.ObjectID
console.log(ObjectID()) // generates an ObjectID
I've been successfully using $in in my node webservice when my mongo arrays only held ids. Here is sample data.
{
"_id": {
"$oid": "52b1a60ce4b0f819260bc6e5"
},
"title": "Sample",
"team": [
{
"$oid": "52995b263e20c94167000001"
},
{
"$oid": "529bfa36c81735b802000001"
}
],
"tasks": [
{
"task": {
"$oid": "52af197ae4b07526a3ee6017"
},
"status": 0
},
{
"task": {
"$oid": "52af197ae4b07526a3ee6017"
},
"status": 1
}
]
}
Notice that tasks is an array, but the id is nested in "task", while in teams it is on the top level. Here is where my question is.
In my Node route, this is how I typically deal with calling a array of IDs in my project, this works fine in the team example, but obviously not for my task example.
app.get('/api/tasks/project/:id', function (req, res) {
var the_id = req.params.id;
var query = req.params.query;
models.Projects.findById(the_id, null, function (data) {
models.Tasks.findAllByIds({
ids: data._doc.tasks,
query: query
}, function(items) {
console.log(items);
res.send(items);
});
});
});
That communicates with my model which has a method called findAllByIds
module.exports = function (config, mongoose) {
var _TasksSchema = new mongoose.Schema({});
var _Tasks = mongoose.model('tasks', _TasksSchema);
/*****************
* Public API
*****************/
return {
Tasks: _Tasks,
findAllByIds: function(data, callback){
var query = data.query;
_Tasks.find({_id: { $in: data.ids} }, query, function(err, doc){
callback(doc);
});
}
}
}
In this call I have $in: data.ids which works in the simple array like the "teams" example above. Once I nest my object, as with "task" sample, this does not work anymore, and I am not sure how to specify $in to look at data.ids array, but use the "task" value.
I'd like to avoid having to iterate through the data to create an array with only id, and then repopulate the other values once the data is returned, unless that is the only option.
Update
I had a thought of setting up my mongo document like this, although I'd still like to know how to do it the other way, in the event this isn't possible in the future.
"tasks": {
"status0": [
{
"$oid": "52995b263e20c94167000001"
},
{
"$oid": "529bfa36c81735b802000001"
}
],
"status1": [
{
"$oid": "52995b263e20c94167000001"
},
{
"$oid": "529bfa36c81735b802000001"
}
]
}
You can call map on the tasks array to project it into a new array with just the ObjectId values:
models.Tasks.findAllByIds({
ids: data.tasks.map(function(value) { return value.task; }),
query: query
}, function(items) { ...
Have you try the $elemMatch option in find conditions ? http://docs.mongodb.org/manual/reference/operator/query/elemMatch/