When Use cursor in Node.js

Question

When Use cursor in Node.js

When And why use Cursor in your project?

If I have this Mock code

var mongoClient = require('mongodb').MongoClient;


mongoClient.connect('mongodb://localhost:27017/tut3',function(err,db){
    if(err)throw err;

    var collection = db.collection('messages');


    collection.find({},{_id:true}).each(function(err,doc){
        console.log("---- CLASSIC ----")
        console.dir(doc);
            //stuff//
    });

    var cursor = collection.find({},{_id:true});

    cursor.each(function(err,doc){
        console.log("---- CURSOR ----")
        console.dir(doc);
        //stuff2
    });

})

The collection messages is huge for example.

Which different between //stuffand //stuff2

I know that if I do

var cursor = collection.find({},{_id:true});

I Know that when cursor return I Have all the documents (synchronous) and it has many methods, but also inside stuff, the query is complete and I have all Documents...

Where is the difference? and when use var cursor instead the "classic" find?

javascript
node.js
mongodb

Answer 1

The difference between

 cursor = collection.find();

and:

collection.find().each(function(err,doc) {

Is basically what is called "method chaining". It's really just an opinion of how you want to write your code. So what is being acted on with methods such as .each() is still just a cursor object that could have optionally be returned on the left hand side.

Also the "cursor" has not yet been "executed" until such a method is called. This means that "modifiers" can be applied without executing yet, as in:

cursor = cursor.skip(10);
cursor = cursor.limit(100);

As all modifiers will also return to the left a cursor.

This is essentially the principle applied in "method chaining", where whatever "type" is returned from the left can be "chained" on the right:

collection.find().skip(10).limit(100).each(function(err,doc) {

If you are dealing with "small" sets of results you can just call the .toArray() method on the cursor:

collection.find({},{_id:true}).toArray(function(err,array) {
   console.log( array ); // everything
});

But if have value 1000's or millions of results, you probably don't want to load all of that in memory. This is where you process with an iterator:

collection.find({},{_id:true}).each(function(err,doc) {
    // do something with the current doc
});

Iteration is why cursors exist.