dynamodb
is a DynamoDB data mapper for node.js.
Features
Installation
npm install dynamodb
Getting Started
First, you need to configure the AWS SDK with your credentials.
var dynamo = require('dynamodb');
dynamo.AWS.config.loadFromPath('credentials.json');
When running on EC2 its recommended to leverage EC2 IAM roles. If you have configured your instance to use IAM roles, DynamoDB will automatically select these credentials for use in your application, and you do not need to manually provide credentials in any other format.
var dynamo = require('dynamodb');
dynamo.AWS.config.update({region: "REGION"});
You can also directly pass in your access key id, secret and region.
- Its recommend you not hard-code credentials inside an application. Use this method only for small personal scripts or for testing purposes.
var dynamo = require('dynamodb');
dynamo.AWS.config.update({accessKeyId: 'AKID', secretAccessKey: 'SECRET', region: "REGION"});
Currently the following region codes are available in Amazon:
Code | Name |
---|
ap-northeast-1 | Asia Pacific (Tokyo) |
ap-southeast-1 | Asia Pacific (Singapore) |
ap-southeast-2 | Asia Pacific (Sydney) |
eu-central-1 | EU (Frankfurt) |
eu-west-1 | EU (Ireland) |
sa-east-1 | South America (Sao Paulo) |
us-east-1 | US East (N. Virginia) |
us-west-1 | US West (N. California) |
us-west-2 | US West (Oregon) |
Define a Model
Models are defined through the toplevel define method.
var Account = dynamo.define('Account', {
hashKey : 'email',
timestamps : true,
schema : {
email : Joi.string().email(),
name : Joi.string(),
age : Joi.number(),
roles : dynamo.types.stringSet(),
settings : {
nickname : Joi.string(),
acceptedTerms : Joi.boolean().default(false)
}
}
});
Models can also be defined with hash and range keys.
var BlogPost = dynamo.define('BlogPost', {
hashKey : 'email',
rangeKey : ‘title’,
schema : {
email : Joi.string().email(),
title : Joi.string(),
content : Joi.binary(),
tags : dynamo.types.stringSet(),
}
});
Create Tables for all defined modules
dynamo.createTables(function(err) {
if (err) {
console.log('Error creating tables: ', err);
} else {
console.log('Tables has been created');
}
});
When creating tables you can pass specific throughput settings for any defined models.
dynamo.createTables({
'BlogPost': {readCapacity: 5, writeCapacity: 10},
'Account': {readCapacity: 20, writeCapacity: 4}
}, function(err) {
if (err) {
console.log('Error creating tables: ', err);
} else {
console.log('Tables has been created');
}
});
Delete Table
BlogPost.deleteTable(function(err) {
if (err) {
console.log('Error deleting table: ', err);
} else {
console.log('Table has been deleted');
}
});
Schema Types
DynamoDB provides the following schema types:
- String
- Number
- StringSet
- NumberSet
- Boolean
- Date
- UUID
- TimeUUID
UUID
UUIDs can be declared for any attributes, including hash and range keys. By
Default, the uuid will be automatically generated when attempting to create
the model in DynamoDB.
var Tweet = dynamo.define('Tweet', {
hashKey : 'TweetID',
timestamps : true,
schema : {
TweetID : dynamo.types.uuid(),
content : Joi.string(),
}
});
Configuration
You can configure dynamo to automatically add createdAt
and updatedAt
timestamp attributes when
saving and updating a model. updatedAt
will only be set when updating a record and will not be set on initial creation of the model.
var Account = dynamo.define('Account', {
hashKey : 'email',
timestamps : true,
schema : {
email : Joi.string().email(),
}
});
If you want dynamo to handle timestamps, but only want some of them, or want your
timestamps to be called something else, you can override each attribute individually:
var Account = dynamo.define('Account', {
hashKey : 'email',
timestamps : true,
createdAt: false,
updatedAt: 'updateTimestamp'
schema : {
email : Joi.string().email(),
}
});
You can override the table name the model will use.
var Event = dynamo.define('Event', {
hashKey : 'name',
schema : {
name : Joi.string(),
total : Joi.number()
},
tableName: 'deviceEvents'
});
if you set the tableName to a function, dynamo will use the result of the function as the active table to use.
Useful for storing time series data.
var Event = dynamo.define('Event', {
hashKey : 'name',
schema : {
name : Joi.string(),
total : Joi.number()
},
tableName: function () {
var d = new Date();
return ['events', d.getFullYear(), d.getMonth() + 1].join('_');
}
});
After you've defined your model you can configure the table name to use.
By default, the table name used will be the lowercased and pluralized version
of the name you provided when defining the model.
Account.config({tableName: 'AccountsTable'});
You can also pass in a custom instance of the aws-sdk DynamoDB client
var dynamodb = new AWS.DynamoDB();
Account.config({dynamodb: dynamodb});
dynamo.dynamoDriver(dynamodb);
Saving Models to DynamoDB
With your models defined, we can start saving them to DynamoDB.
Account.create({email: 'foo@example.com', name: 'Foo Bar', age: 21}, function (err, acc) {
console.log('created account in DynamoDB', acc.get('email'));
});
You can also first instantiate a model and then save it.
var acc = new Account({email: 'test@example.com', name: 'Test Example'});
acc.save(function (err) {
console.log('created account in DynamoDB', acc.get('email'));
});
Saving models that require range and hashkeys are identical to ones with only
hashkeys.
BlogPost.create({
email: 'werner@example.com',
title: 'Expanding the Cloud',
content: 'Today, we are excited to announce the limited preview...'
}, function (err, post) {
console.log('created blog post', post.get('title'));
});
Pass an array of items and they will be saved in parallel to DynamoDB.
var item1 = {email: 'foo1@example.com', name: 'Foo 1', age: 10};
var item2 = {email: 'foo2@example.com', name: 'Foo 2', age: 20};
var item3 = {email: 'foo3@example.com', name: 'Foo 3', age: 30};
Account.create([item1, item2, item3], function (err, acccounts) {
console.log('created 3 accounts in DynamoDB', accounts);
});
Use expressions api to do conditional writes
var params = {};
params.ConditionExpression = '#i <> :x';
params.ExpressionAttributeNames = {'#i' : 'id'};
params.ExpressionAttributeValues = {':x' : 123};
User.create({id : 123, name : 'Kurt Warner' }, params, function (error, acc) { ... });
Use the overwrite
option to prevent over writing of existing records.
- By default
overwrite
is set to true, allowing create operations to overwrite existing records
User.create({id : 123, name : 'Kurt Warner' }, {overwrite : false}, function (error, acc) { ... });
Updating
When updating a model the hash and range key attributes must be given, all
other attributes are optional
Account.update({email: 'foo@example.com', name: 'Bar Tester'}, function (err, acc) {
console.log('update account', acc.get('name'));
});
Model.update
accepts options to pass to DynamoDB when making the updateItem request
Account.update({email: 'foo@example.com', name: 'Bar Tester'}, {ReturnValues: 'ALL_OLD'}, function (err, acc) {
console.log('update account', acc.get('name'));
});
Account.update({email: 'foo@example.com', name: 'Bar Tester'}, {expected: {age: 22}}, function (err, acc) {
console.log('update account', acc.get('name'));
});
Account.update({email: 'foo@example.com', age: null}, function (err, acc) {
console.log('update account', acc.get('age'));
});
You can also pass what action to perform when updating a given attribute
Use $add to increment or decrement numbers and add values to sets
Account.update({email : 'foo@example.com', age : {$add : 1}}, function (err, acc) {
console.log('incremented age by 1', acc.get('age'));
});
BlogPost.update({
email : 'werner@example.com',
title : 'Expanding the Cloud',
tags : {$add : 'cloud'}
}, function (err, post) {
console.log('added single tag to blog post', post.get('tags'));
});
BlogPost.update({
email : 'werner@example.com',
title : 'Expanding the Cloud',
tags : {$add : ['cloud', 'dynamodb']}
}, function (err, post) {
console.log('added tags to blog post', post.get('tags'));
});
$del will remove values from a given set
BlogPost.update({
email : 'werner@example.com',
title : 'Expanding the Cloud',
tags : {$del : 'cloud'}
}, function (err, post) {
console.log('removed cloud tag from blog post', post.get('tags'));
});
BlogPost.update({
email : 'werner@example.com',
title : 'Expanding the Cloud',
tags : {$del : ['aws', 'node']}
}, function (err, post) {
console.log('removed multiple tags', post.get('tags'));
});
Use the expressions api to update nested documents
var params = {};
params.UpdateExpression = 'SET #year = #year + :inc, #dir.titles = list_append(#dir.titles, :title), #act[0].firstName = :firstName ADD tags :tag';
params.ConditionExpression = '#year = :current';
params.ExpressionAttributeNames = {
'#year' : 'releaseYear',
'#dir' : 'director',
'#act' : 'actors'
};
params.ExpressionAttributeValues = {
':inc' : 1,
':current' : 2001,
':title' : ['The Man'],
':firstName' : 'Rob',
':tag' : dynamo.Set(['Sports', 'Horror'], 'S')
};
Movie.update({title : 'Movie 0', description : 'This is a description'}, params, function (err, mov) {});
Deleting
You delete items in DynamoDB using the hashkey of model
If your model uses both a hash and range key, than both need to be provided
Account.destroy('foo@example.com', function (err) {
console.log('account deleted');
});
BlogPost.destroy('foo@example.com', 'Hello World!', function (err) {
console.log('post deleted')
});
BlogPost.destroy({email: 'foo@example.com', title: 'Another Post'}, function (err) {
console.log('another post deleted')
});
Model.destroy
accepts options to pass to DynamoDB when making the deleteItem request
Account.destroy('foo@example.com', {ReturnValues: true}, function (err, acc) {
console.log('account deleted');
console.log('deleted account name', acc.get('name'));
});
Account.destroy('foo@example.com', {expected: {age: 22}}, function (err) {
console.log('account deleted if the age was 22');
Use expression apis to perform conditional deletes
var params = {};
params.ConditionExpression = '#v = :x';
params.ExpressionAttributeNames = {'#v' : 'version'};
params.ExpressionAttributeValues = {':x' : '2'};
User.destroy({id : 123}, params, function (err, acc) {});
Loading models from DynamoDB
The simpliest way to get an item from DynamoDB is by hashkey.
Account.get('test@example.com', function (err, acc) {
console.log('got account', acc.get('email'));
});
Perform the same get request, but this time peform a consistent read.
Account.get('test@example.com', {ConsistentRead: true}, function (err, acc) {
console.log('got account', acc.get('email'));
});
Model.get
accepts any options that DynamoDB getItem request supports. For
example:
Account.get('test@example.com', {ConsistentRead: true, AttributesToGet : ['name','age']}, function (err, acc) {
console.log('got account', acc.get('email'))
console.log(acc.get('name'));
console.log(acc.get('age'));
console.log(acc.get('email'));
});
Get a model using hash and range key.
BlogPost.get('werner@example.com', 'dynamodb-keeps-getting-better-and-cheaper', function (err, post) {
console.log('loaded post by range and hash key', post.get('content'));
});
Model.get
also supports passing an object which contains hash and range key
attributes to load up a model
BlogPost.get({email: 'werner@example.com', title: 'Expanding the Cloud'}, function (err, post) {
console.log('loded post', post.get('content'));
});
Use expressions api to select which attributes you want returned
User.get({ id : '123456789'},{ ProjectionExpression : 'email, age, settings.nickname' }, function (err, acc) {});
Query
For models that use hash and range keys DynamoDB provides a flexible and
chainable query api
BlogPost
.query('werner@example.com')
.exec(callback);
BlogPost
.query('werner@example.com')
.loadAll()
.exec(callback);
BlogPost
.query('werner@example.com')
.limit(5)
.exec(callback);
BlogPost
.query('werner@example.com')
.where('title').beginsWith('Expanding')
.exec(callback);
BlogPost
.query('werner@example.com')
.where('title').beginsWith('Expanding')
.select('COUNT')
.exec(callback);
BlogPost
.query('werner@example.com')
.where('title').beginsWith('Expanding')
.attributes(['title', 'content'])
.limit(10)
.exec(callback);
BlogPost
.query('werner@example.com')
.ascending()
.exec(callback)
BlogPost
.query('werner@example.com')
.descending()
.exec(callback)
BlogPost
.query('werner@example.com')
.where('title').gt('Expanding')
.attributes(['title', 'content'])
.limit(10)
.ascending()
.loadAll()
.exec(callback);
DynamoDB supports all the possible KeyConditions that DynamoDB currently
supports.
BlogPost
.query('werner@example.com')
.where('title').equals('Expanding')
.exec();
BlogPost
.query('werner@example.com')
.where('title').lte('Expanding')
.exec();
BlogPost
.query('werner@example.com')
.where('title').lt('Expanding')
.exec();
BlogPost
.query('werner@example.com')
.where('title').gt('Expanding')
.exec();
BlogPost
.query('werner@example.com')
.where('title').gte('Expanding')
.exec();
BlogPost
.query('werner@example.com')
.where('title').beginsWith('Expanding')
.exec();
BlogPost
.query('werner@example.com')
.where('title').between('foo@example.com', 'test@example.com')
.exec();
Query Filters allow you to further filter results on non-key attributes.
BlogPost
.query('werner@example.com')
.where('title').equals('Expanding')
.filter('tags').contains('cloud')
.exec();
Expression Filters also allow you to further filter results on non-key attributes.
BlogPost
.query('werner@example.com')
.filterExpression('#title < :t')
.expressionAttributeValues({ ':t' : 'Expanding' })
.expressionAttributeNames({ '#title' : 'title'})
.projectionExpression('#title, tag')
.exec();
See the queryFilter.js example for more examples of using query filters
Global Indexes
First, define a model with a global secondary index.
var GameScore = dynamo.define('GameScore', {
hashKey : 'userId',
rangeKey : 'gameTitle',
schema : {
userId : Joi.string(),
gameTitle : Joi.string(),
topScore : Joi.number(),
topScoreDateTime : Joi.date(),
wins : Joi.number(),
losses : Joi.number()
},
indexes : [{
hashKey : 'gameTitle', rangeKey : 'topScore', name : 'GameTitleIndex', type : 'global'
}]
});
Now we can query against the global index
GameScore
.query('Galaxy Invaders')
.usingIndex('GameTitleIndex')
.descending()
.exec(callback);
When can also configure the attributes projected into the index.
By default all attributes will be projected when no Projection parameter is
present
var GameScore = dynamo.define('GameScore', {
hashKey : 'userId',
rangeKey : 'gameTitle',
schema : {
userId : Joi.string(),
gameTitle : Joi.string(),
topScore : Joi.number(),
topScoreDateTime : Joi.date(),
wins : Joi.number(),
losses : Joi.number()
},
indexes : [{
hashKey : 'gameTitle',
rangeKey : 'topScore',
name : 'GameTitleIndex',
type : 'global',
projection: { NonKeyAttributes: [ 'wins' ], ProjectionType: 'INCLUDE' }
}]
});
Filter items against the configured rangekey for the global index.
GameScore
.query('Galaxy Invaders')
.usingIndex('GameTitleIndex')
.where('topScore').gt(1000)
.descending()
.exec(function (err, data) {
console.log(_.map(data.Items, JSON.stringify));
});
Local Secondary Indexes
First, define a model using a local secondary index
var BlogPost = dynamo.define('Account', {
hashKey : 'email',
rangekey : 'title',
schema : {
email : Joi.string().email(),
title : Joi.string(),
content : Joi.binary(),
PublishedDateTime : Joi.date()
},
indexes : [{
hashkey : 'email', rangekey : 'PublishedDateTime', type : 'local', name : 'PublishedIndex'
}]
});
Now we can query for blog posts using the secondary index
BlogPost
.query('werner@example.com')
.usingIndex('PublishedIndex')
.descending()
.exec(callback);
Could also query for published posts, but this time return oldest first
BlogPost
.query('werner@example.com')
.usingIndex('PublishedIndex')
.ascending()
.exec(callback);
Finally lets load all published posts sorted by publish date
BlogPost
.query('werner@example.com')
.usingIndex('PublishedIndex')
.descending()
.loadAll()
.exec(callback);
Learn more about secondary indexes
Scan
DynamoDB provides a flexible and chainable api for scanning over all your items
This api is very similar to the query api.
Account.scan().exec(callback);
Account
.scan()
.loadAll()
.exec(callback);
Account
.scan()
.limit(20)
.exec();
Account
.scan()
.limit(20)
.loadAll()
.exec();
Account
.scan()
.where('email').gte('f@example.com')
.attributes(['email','created'])
.returnConsumedCapacity()
.exec();
Account
.scan()
.where('age').gte(21)
.select('COUNT')
.exec();
Account
.scan()
.where('age').notNull()
.startKey('foo@example.com')
.exec()
DynamoDB supports all the possible Scan Filters that DynamoDB currently supports.
Account
.scan()
.where('name').equals('Werner')
.exec();
Account
.scan()
.where('name').ne('Werner')
.exec();
Account
.scan()
.where('name').lte('Werner')
.exec();
Account
.scan()
.where('name').lt('Werner')
.exec();
Account
.scan()
.where('name').gte('Werner')
.exec();
Account
.scan()
.where('name').gt('Werner')
.exec();
Account
.scan()
.where('name').null()
.exec();
Account
.scan()
.where('name').notNull()
.exec();
Account
.scan()
.where('name').contains('ner')
.exec();
Account
.scan()
.where('name').notContains('ner')
.exec();
Account
.scan()
.where('name').in(['foo@example.com', 'bar@example.com'])
.exec();
Account
.scan()
.where('name').beginsWith('Werner')
.exec();
Account
.scan()
.where('name').between('Bar', 'Foo')
.exec();
Account
.scan()
.where('name').equals('Werner')
.where('age').notNull()
.exec();
You can also use the new expressions api when filtering scans
User.scan()
.filterExpression('#age BETWEEN :low AND :high AND begins_with(#email, :e)')
.expressionAttributeValues({ ':low' : 18, ':high' : 22, ':e' : 'test1'})
.expressionAttributeNames({ '#age' : 'age', '#email' : 'email'})
.projectionExpression('#age, #email')
.exec();
Parallel Scan
Parallel scans increase the throughput of your table scans.
The parallel scan operation is identical to the scan api.
The only difference is you must provide the total number of segments
Caution you can easily consume all your provisioned throughput with this api
var totalSegments = 8;
Account.parallelScan(totalSegments)
.where('age').gte(18)
.attributes('age')
.exec(callback);
Account
.parallelScan(totalSegments)
.exec()
More info on Parallel Scans
Batch Get Items
Model.getItems
allows you to load multiple models with a single request to DynamoDB.
DynamoDB limits the number of items you can get to 100 or 1MB of data for a single request.
DynamoDB automatically handles splitting up into multiple requests to load all
items.
Account.getItems(['foo@example.com','bar@example.com', 'test@example.com'], function (err, accounts) {
console.log('loaded ' + accounts.length + ' accounts');
});
var postKey1 = {email : 'test@example.com', title : 'Hello World!'};
var postKey2 = {email : 'test@example.com', title : 'Another Post'};
BlogPost.getItems([postKey1, postKey2], function (err, posts) {
console.log('loaded posts');
});
Model.getItems
accepts options which will be passed to DynamoDB when making the batchGetItem request
Account.getItems(['foo@example.com','bar@example.com'], {ConsistentRead: true}, function (err, accounts) {
console.log('loaded ' + accounts.length + ' accounts');
});
Streaming api
dynamo supports a basic streaming api in addition to the callback
api for query
, scan
, and parallelScan
operations.
var stream = Account.parallelScan(4).exec();
stream.on('readable', function () {
console.log('single parallel scan response', stream.read());
});
stream.on('end', function () {
console.log('Parallel scan of accounts finished');
});
var querystream = BlogPost.query('werner@dynamo.com').loadAll().exec();
querystream.on('readable', function () {
console.log('single query response', stream.read());
});
querystream.on('end', function () {
console.log('query for blog posts finished');
});
Dynamic Table Names
dynamo supports dynamic table names, useful for storing time series data.
var Event = dynamo.define('Event', {
hashKey : 'name',
schema : {
name : Joi.string(),
total : Joi.number()
},
tableName: function () {
var d = new Date();
return ['events', d.getFullYear(), d.getMonth() + 1].join('_');
}
});
Logging
Logging can be enabled to provide detailed information on data being sent and returned from DynamoDB.
By default logging is turned off.
dynamo.log.level('info');
Logging can also be enabled / disabled at the model level.
var Account = dynamo.define('Account', {hashKey : 'email'});
var Event = dynamo.define('Account', {hashKey : 'name'});
Account.log.level('warn');
Examples
var dynamo = require('dynamodb');
var Account = dynamo.define('Account', {
hashKey : 'email',
timestamps : true,
schema : {
email : Joi.string().email(),
name : Joi.string().required(),
age : Joi.number(),
}
});
Account.create({email: 'test@example.com', name : 'Test Account'}, function (err, acc) {
console.log('created account at', acc.get('created'));
acc.set({age: 22});
acc.update(function (err) {
console.log('updated account age');
});
});
See the examples for more working sample code.
Support
DynamoDB is provided as-is, free of charge. For support, you have a few choices:
- Ask your support question on Stackoverflow.com, and tag your question with dynamodb.
- If you believe you have found a bug in dynamodb, please submit a support ticket on the Github Issues page for dynamo. We'll get to them as soon as we can.
Maintainers
Authors
License
(The MIT License)
Copyright (c) 2016 Ryan Fitzgerald
Permission is hereby granted, free of charge, to any person obtaining
a copy of this software and associated documentation files (the
"Software"), to deal in the Software without restriction, including
without limitation the rights to use, copy, modify, merge, publish,
distribute, sublicense, and/or sell copies of the Software, and to
permit persons to whom the Software is furnished to do so, subject to
the following conditions:
The above copyright notice and this permission notice shall be
included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.