Node MongoDB Wrapper
This package greatly simplifies working with MongoDB and Node MongoDB Native.
It removes a lot of the callback madness and provides a simple shorthand for common operations. It also localizes your db configuration within itself so you only specify a name rather than a multitude of properties and has its own very simple connection pooling and caching.
All of these features can be removed or replaced easily.
This is in use in production at Playtomic as part of the high-volume api server.
Requires
- MongoDB has to be running somewhere.
- Node MongoDB Native and NodeJS
How to use
- git clone https://github.com/benlowry/node-mongodb-wrapper
- cd node-mongodb-wrapper
- node test.js
or just npm install node-mongodb-wrapper
Methods
Node MongoDB Wrapper provides methods for:
get
performs a find() with optional cachinggetAndCount
performs a find() + count() with optional cachinggetOrInsert
performs a find() and inserts if not existscount
performs a count() with optional cachinginsert
performs a save()update
performs an update()move
performs a save(doc) on new collection then remove(doc) on oldremove
performs a remove()
Examples
A complete suite of examples is available in the included test.js file.
db.get("test", "stuff", {filter: {x: 1, y: 2, z: 3}, cache: true, cachetime: 60}, function(error, items) {
console.log("huzzah!");
});
or (see shorthand note below)
db.test.stuff.get({filter: {x: 1, y: 2, z: 3}, cache: true, cachetime: 60}, function(error, items) {
console.log("huzzah!");
});
In that short example "test" is one of our configured database's names:
var databases = {
test: {
address: "127.0.0.1",
port: 27017,
name: "test", // your db and this object's name must match
//username: "optional",
//password: "optional"
}
}
We're passing an object that contains a nested filter object which is the query criteria and is exactly as you would use directly, it also supports limit, sort and skip in the outer object. The query is marked as cacheable and will store the results for 60 seconds.
Shorthand
I saw this on mongode and thought it looked super cool, so I copied the idea.
You can use traditional db.databasename.collectionname.method as well now to save on the parameter overload. This also has the benefit of making sure your collection names are strict.
The only bad bit is you have to predefine the collection names because JavaScript has no 'catch all' property which is unfortunate, but you can do it in 3 ways and if a collection is already defined it will just skip doing it again.
db.databasename.collection("acollection");
db.databasename.collections(["an", "array", "of", "collections"]);
db.databasename.collections(callback);
The final example will query your database and create the shorthand path for any collection names without dots (eg no system.indexes).
The callback has only an error parameter so you know if it worked or not, this is an async operation and you cannot use the shorthand until it is complete.
Why
Because without this you end up with too much boilerplate and nesting:
var db = new Db("local", new Server("127.0.0.1", 27017));
db.open(function(error, connection) {
if(error) {
console.log("error: " + error);
return;
}
connection.authenticate(username, password, function(error) {
if(error) {
console.log("error2: " + error);
return;
}
var collection = new mongodb.Collection(connection, "stuff");
collection.find({x: 1, y: 2, z: 3}, function(error, items) {
if(error) {
console.log("error3: " + error);
return;
}
console.log("huzzah!");
});
});
});
What's missing
There's one feature that would be great to have that I haven't built in and that is sending a batch of operations in to be processed in the single request. This could remove code complexity even further although with the connection re-use there may not be much performance gain.
License
Copyright Playtomic Inc, 2012. Licensed under the MIT license. Certain portions may come from 3rd parties and carry their own licensing terms and are referenced where applicable.