使用MongoDB的NodeJS-实例化“ db”变量的最佳实践

Question

I'm currently trying to get a grip on NodeJS / Express / MongoDB and read several tutorials about it. In this tutorial by Christopher Buecheler , consider step 5 and 6:

In app.js , the following code is added:

// ...
var mongo = require('mongodb');
var monk = require('monk');
var db = monk('localhost:27017/nodetest1');

app.get('/userlist', routes.userlist(db));
// ...

And in routes/index.js you can find the handler which belongs to the /userlist route:

exports.userlist = function(db) {
    return function(req, res) {
        // do stuff
    };
};

I was confused by passing the db object to every handler, since, in my opinion, it creates a lot of redundancy, because you need to pass it to every handler and wrap an additional function around it. So I tried to add the initialisation of the db object into the routes/index.js file which seems to work fine and saves a lot of lines when dealing with more urls:

var mongodb = require('mongodb');
var monk = require('monk');

var db = monk('localhost:27017/nodetest1');

exports.userlist = function(req, res) {
  // do stuff
};

Since I'm new to MongoDB, are there any downsides doing it this way? What advantages would it have to pass the db object to every handler instead?

Answer 1

Personally I do it like you did, I instantiate it in the place that uses it.

But I have seen other people do it like in this tutorial, passing it over every time and I think it does make sense if you need to use that db in multiple places, think if you had many different files that all need that database... now instantiating it every time, that would be redundant.

Well, that's my 2 cents... not sure there is any "good" answer to this... It might be a good practice to keep it in app.js in case you need it all over the place, but I don't think there is any disadvantage of doing otherwise if only used at a specific place.