I am using sails JS with Mongo DB.
My model is:
module.exports = {
attributes: {
title:{type:"string",required:true},
content:{type:"string",required:true},
date:{type:"string",required:true},
filename:{type:"string",required:true},
},
};
My Controller is:
fetchposts:function(req,res){
console.log("in fetch posts")
mysort={$id:-1}
Cyberblog.find().sort(mysort).limit(5).exec(function(err, result) {
if (err || !result) {
message="no records fetched";
console.log(message);
res.redirect('/newpost');
}
else{
console.log(result)
}
I am facing an error saying that
"Warning: The sort clause in the provided criteria is specified as a dictionary (plain JS object),
meaning that it is presumably using Mongo-Esque semantics (something like { fullName: -1, rank: 1 }).
But as of Sails v1/Waterline 0.13, this is no longer the recommended usage. Instead, please use either
a string like 'fullName DESC', or an array-like [ { fullName: 'DESC' } ].
(Since I get what you mean, tolerating & remapping this usage for now...)
and I am unable to fetch any records. It is showing no records fetched.
So I have one warning on Sort and no records coming from DB. Please help me resolve the issue.
Sort clause allow send string:
var users = await User.find({ name: 'Jake'})
.sort('age ASC');
return res.json(users);
Or an array:
var users = await User.find({ name: 'Finn'})
.sort([
{ age: 'ASC' },
{ createdAt: 'ASC' },
]);
return res.json(users);
Check this out in the documentation:
https://sailsjs.com/documentation/reference/waterline-orm/queries/sort
Related
I have a NodeJS application with Mongoose ODM(Mongoose 3.3.1). I want to retrieve all fields except 1 from my collection.For Example: I have a collection Product Which have 6 fields,I want to select all except a field "Image" . I used "exclude" method, but got error..
This was my code.
var Query = models.Product.find();
Query.exclude('title Image');
if (req.params.id) {
Query.where('_id', req.params.id);
}
Query.exec(function (err, product) {
if (!err) {
return res.send({ 'statusCode': 200, 'statusText': 'OK', 'data': product });
} else {
return res.send(500);
}
});
But this returns error
Express
500 TypeError: Object #<Query> has no method 'exclude'.........
Also I tried, var Query = models.Product.find().exclude('title','Image'); and var Query = models.Product.find({}).exclude('title','Image'); But getting the same error. How to exclude one/(two) particular fields from a collection in Mongoose.
Use query.select for field selection in the current (3.x) Mongoose builds.
Prefix a field name you want to exclude with a -; so in your case:
Query.select('-Image');
Quick aside: in JavaScript, variables starting with a capital letter should be reserved for constructor functions. So consider renaming Query as query in your code.
I don't know where you read about that .exclude function, because I can't find it in any documentation.
But you can exclude fields by using the second parameter of the find method.
Here is an example from the official documentation:
db.inventory.find( { type: 'food' }, { type:0 } )
This operation returns all documents where the value of the type field is food, but does not include the type field in the output.
Model.findOne({ _id: Your Id}, { password: 0, name: 0 }, function(err, user){
// put your code
});
this code worked in my project. Thanks!! have a nice day.
You could do this
const products = await Product.find().select(['-image'])
I am use this with async await
async (req, res) => {
try {
await User.findById(req.user,'name email',(err, user) => {
if(err || !user){
return res.status(404)
} else {
return res.status(200).json({
user,
});
}
});
} catch (error) {
console.log(error);
}
In the updated version of Mongoose you can use it in this way as below to get selected fields.
user.findById({_id: req.body.id}, 'username phno address').then(response => {
res.status(200).json({
result: true,
details: response
});
}).catch(err => {
res.status(500).json({ result: false });
});
I'm working on a feature. I store a userId array name "collectedUser" than who is collected the project. And I just want to return a field "isCollected" instead of "collectedUsers". So select is not what I want. But I got this solution.
This is after I get projects from database, I add "isCollected".
for (const item of projects) {
item.set("isCollected", item.collectedUsers.includes(userId), {
strict: false,
})
}
And this is in Decorator #Schema
#Schema({
timestamps: true,
toObject: {
virtuals: true,
versionKey: false,
transform: (doc, ret, options): Partial<Project> => {
return {
...ret,
projectManagers: undefined,
projectMembers: undefined,
collectedUsers: undefined
}
}
}
})
Finally in my controller
projects = projects.map(i => i.toObject())
It's a strange tricks that set undefined, but it really work.
Btw I'm using nestjs.
You can do it like this
const products = await Product.find().select({
"image": 0
});
For anyone looking for a way to always omit a field - more like a global option rather than doing so in the query e.g. a password field, using a getter that returns undefined also works
{
password: {
type: String,
required: true,
get: () => undefined,
},
}
NB: Getters must be enabled with option { toObject: { getters:true } }
you can exclude the field from the schema definition
by adding the attribute
excludedField : {
...
select: false,
...
}
whenever you want to add it to your result,
add this to your find()
find().select('+excludedFiled')
Yet another first-timer problem here. This gets data from a database and displays it in some text fields (that part is not shown in the code below) and after the user edits it the data should be updated in the database via the findAndModify() method and I think this is where the issue lies. There are no errors, it just doesn't do anything. EDIT The following error is received: MongoError: Either an update or remove=true must be specified
server.js
MongoClient.connect("mongodb://user:secretPassword#aws-us-east-1-portal.7.dblayer.com:10712,aws-us-east-1-portal.10.dblayer.com:10316/database", function(err, db) {
if (err) throw err;
var contactList = db.collection("contactList");
app.put('/contactList/:id', function(req, res) {
var id = req.params.id;
console.log("edited: " + req.body.name); //works up until here
contactList.findAndModify({
query: {_id: mongojs.ObjectId(id)},
update: {$set: {name: req.body.name, email: req.body.email, number: req.body.number}},
new: true
}, function (err, doc) {
res.json(doc);
})
});
controller.js
$scope.update = function() {
$http.put('/contactList/' + $scope.contact._id, $scope.contact).success(function(response) {
refresh();
})
};
If this were me I would first do a couple of things:
Before your call to findAndModify just do a simple find using your query. Make sure you can actually find the object using your query. If that works you know that the 'find' part of the findAndModify is probably ok.
Do some console logging inside the callback handler of the findAndModify call. As it stands you do not do anything if an err is returned from the findAndModify call. It is possible your call is returning an error that you are just ignoring and it may provide some additional insight into your problem.
I would try these two first and see if it helps.
Update:
Example using native:
collection.findAndModify(
{ field: 'some value' },
[],
{ $set: { field2: 'some new value' } },
{ new:true },
function(err, doc) {
//handle err and doc
});
I would like to take the last n documents from my Cloudant database using a Node query. So far I have narrowed it down to the find() function, but the documentation only really explains how to retrieve all documents containing an absolute value, for example:
db.find({selector:{name:'Alice'}}, function(er, result) {
...
});
(taken from https://www.npmjs.com/package/cloudant#cloudant-query)
What I'm looking for is the equivalent of this SQL:
SELECT * FROM db WHERE name = "Alice" LIMIT 10
The code I have so far is this:
var cloudant = require('cloudant');
cloudant({account: username, password: password}, function (err, conn) {
if (err) {
callback("Could not initialize connection to Cloudant: " + err);
} else {
var db = conn.db.use('mydb');
db.find(???, function(err, data) {
if (err) {
callback("No data found: " + err);
} else {
...
}
});
}
});
If I need to make design documents, I'd do so in the Cloudant online interface, so don't worry too much about making an executable answer for that if it's necessary.
It's important to note that Cloudant Query requires you to define your index before performing the query e.g. to index the 'name' field from your documents:
db.index( {name:'nameindex', type:'json', index:{fields:['name']}}
We can now query the data using the find function as you indicated:
var query = { selector: { name: 'Alice' }};
db.find(query, function(err, data) {
});
The interesting thing about your question is the phrase 'the last n documents'. We can retrieve 'n' documents by adding a 'limit' to our query.
var query = { selector: { name: 'Alice' }, limit: 10};
db.find(query, function(err, data) {
});
but this doesn't necessarily indicate the last ten documents; it just limits the result set to ten.
If you want your query results to appear in time order, then you'll need something in your documents that indicates the time e.g.
a time string : { "name": "Alice", "datetime": "2015-11-26 10:22:24 Z" }
a timestamp : { "name": "Alice", "ts": "123456678" }
When your document contains a field which represents time, then your index can be created incorporate this into the index e.g.
db.index( {name:'nameindex', type:'json', index:{fields:['name','time']}}
and documents can be queried to appear in *reverse order * (to get the latest first):
var query = { selector: { name: 'Alice' }, sort: [ { name: "desc"}, { ts: "desc"]};
db.find(query, function(err, data) {
});
You may also want to look at type:"text" indexes too. See https://docs.cloudant.com/cloudant_query.html
var query = { selector: { name: 'Alice' }, sort: [ { name: "desc"}, { ts: "desc"]};
should be
var query = { selector: { name: 'Alice' }, sort: [ { name: "desc"}, { ts: "desc"}];
I'm trying to update an existing document by increment a counter and pushing an object to an array.
Here's the schema for User:
var UserSchema = new Schema({
[...]
posts: {
totalWords: { type: Number, min: 0, default: 0 },
_entries: [
{
words: { type: Number, min: 0 },
body: { type: String },
date: Date
}
]
},
});
And here's the update code:
var newPost = {
words: req.body.words,
body: req.body.entry,
date: new Date()
};
User.findOne(req.user._id, function (err, user) {
var previous = user.posts.totalWords;
user.posts.totalWords = previous + newPost.words;
user.posts._entries.push(newPost);
user.save(function (err) {
if (err) return res.send(400, err);
return res.json(newPost);
});
});
I get the following error:
[TypeError: Object.keys called on non-object]
Any ideas on how to solve this?
Answering my own question
I was able to solve the problem by changing:
User.findOne(req.user._id, function (err, user) { [...] });
Into this:
User.findById(req.user._id, function (err, user) { [...] });
I think, if you would like to use findOne you need follow syntax:
User.findOne({'_id': req.user._id}, {}, function (err, user) {
...
});
Not sure about findById() vs. findOne(), but I've had problems with Mongoose objects returning Object.keys called on non-object while saving or updating with mal-formed data or data corresponding to an older schema. While initializing the document, Mongoose was expecting a subdocument of some kind, but the data in the database didn't match that.
For me it usually happens when I change schemas from a simple object (String, Number, etc.) to a more complex subdocument of some kind. The schema types are mixed up and the object won't load, not even to fix the problem. I had to go into my database using the native driver, search for mal-formed documents using the $type operator, and update them individually.
I have a data set like this:
{
name : 'Doc Name',
photos: [
{
name: 'photo1',
url: 'http://.....'
},
{
name: 'photo2',
url: 'http://......'
}
],
etc ...
Using Monk https://github.com/LearnBoost/monk how do I update photo2? I can use an index as I am iterating over the fields at the moment.
My current attempt below gives me an error, and I can't use a variable for the JSON selector (as in the index).
collection.update({_id: data._id}, {photos[i].data: filename}, function(err, updatedata) {
});
Updating items at a position in an array can be done using the positional $ operator
collection.update(
{ _id: data.id, "photos.name": "photo2" },
{ $set: { "photos.$.data": "yourdata" } }
)
So I found a solution to my problem but there may be some better options and I will leave it unanswered. But for anyone else with the same issue this is what I did:
I extracted the MongoDB document as an object in Node.js, manipulated the document, and then replaced the entire array in a single update statement. For example here is some pseudo code:
collection.find({id: 1}, function(err, doc){
for(i=0; i< doc.array.length; i++) {
//do what you gotta do
doc.array[i].property = 'new value';
}
collection.update({id: 1}, {$set : {"doc.array": doc.array}}, function(err,doc){
console.log(err);
}
})