Install the latest production release for your system from http://www.mongodb.org/downloads. This is easy. However, if you need more details, refer to http://docs.mongodb.org/manual/installation/. Start the mongod server process (for example, from c:mongodbin on Windows) before the Dart app. To make the jobsdb database, start a mongo shell and type use jobsdb; it's that simple. Alternatively, this can also be done via mongo_dart. NoSQL databases are schemaless, so the collections (which is what tables are called here) are created when the first document (or record) is inserted.

The application starts from using_mongodb.dart, where a JobStore object is created, the database is opened, records are written to the jobs table, and then these records are retrieved. The code from using_mongodb.dart is identical to the code from using_postgresql.dart, except that we now import jobstore_mongodb.dart; so please refer to the Storing data in PostgreSQL recipe. When running this script, first comment out js.openAndRead();. In the second run, uncomment this and comment out js.openAndStore();; otherwise, the reads take place before the inserts are completed. The jobstore_mongodb.dart code contains the code to talk to the database driver:

import 'package:mongo_dart/mongo_dart.dart';
import 'job.dart';

const String DEFAULT_URI = 'mongodb://127.0.0.1/';
const String DB_NAME = 'jobsdb';
const String COLLECTION_NAME = 'jobs';

class JobStore {
  final List<Job> jobs = new List();
  Job job;
  Db db;
  DbCollection jobsColl;
  
  JobStore() {
  // 1- make a new database
  db = new Db('${DEFAULT_URI}${DB_NAME}'),
  // make a new collection
  jobsColl = db.collection(COLLECTION_NAME);
}

openAndStore() {
  db.open().then((_) {
  storeData();
  }).catchError(print);
}

storeData() {
  for (job in jobs) {
  insert(job);
  }
}

// 2- inserting a document in a collection:
insert(Job job) {
  var jobMap = job.toMap();
  jobsColl.insert(jobMap).then((_) {
  print('inserted job: ${job.type}'),
  }).catchError(print);
}

openAndRead() {
  db.open().then((_) {
  readData();
  }).catchError(print);
}

// 3- reading documents:
readData() {
  jobsColl.find().toList().then((jobList) {
  processJob(jobList);
  }).catchError(print);
}

// 4- working with document data:
processJob(jobList) {
  jobList.forEach((jobMap) {
  Job job = new Job.fromMap(jobMap);
  print('${job.dbKey} - ${job.type} - ${job.salary} - '
  '${job.company} - ${job.posted} ${job.open}'),
  });
}

// 5-  updating a document:
update(Job job) {
  var jobMap = job.toMap();
  jobsColl.update({"dbKey": jobMap["dbKey"]},jobMap).then((_) {
  print('job updated'),
  }).catchError(print);
}

// 6- deleting a document:
delete(Job job) {
  var jobMap = job.toMap();
  jobsColl.remove(jobMap).then((_) {
  print('job updated'),
  }).catchError(print);
  }
}

After running the preceding script, verify in a mongo shell with use jobsdb and db.jobs.find() that the documents have been inserted, as shown in the following screenshot:

The mongo_dart library makes it very easy to work with MongoDB. In comment 1, we created the Db and DbCollection objects in the constructor of JobStore. To manipulate documents, you have to first call the open method on the Db object.

To write a document to the collection, use the insert method on the DbCollection object, but its argument must be in the Map format. Reading documents is done with the find method. The list that is returned contains items of type Map, so to construct real objects, we have to make a new named constructor fromMap in class Job. Updating a document uses a method of the same name, but its first argument must be a selector to find the document that must be updated. Deleting a document uses the remove method.