A drop in replacement for LevelDOWN that works with DynamoDB as its storage. Can be used as a back-end for LevelUP rather than an actual LevelDB store.
As of version 0.7, LevelUP allows you to pass a 'db'
option when you create a new instance. This will override the default LevelDOWN store with a LevelDOWN API compatible object. DynamoDOWN conforms exactly to the LevelDOWN API but performs operations against a DynamoDB database.
Caution: this should work for most cases, but some LevelDOWN tests are failing. See test/index.js
var levelup = require('levelup')
var db = levelup('table_name', {
db: require('dyanmodown'),
// required AWS config
dynamo: {
region: 'us-east-1',
secretAccessKey: 'abc',
accessKeyId: '123'
}
})
db.put('name', 'Yuri Irsenovich Kim')
db.put('dob', '16 February 1941')
db.put('spouse', 'Kim Young-sook')
db.put('occupation', 'Clown')
db.createReadStream()
.on('data', console.log)
.on('close', function () { console.log('Show\'s over folks!') })
Running our example gives:
{ key: 'dob', value: '16 February 1941' }
{ key: 'name', value: 'Yuri Irsenovich Kim' }
{ key: 'occupation', value: 'Clown' }
{ key: 'spouse', value: 'Kim Young-sook' }
Show's over folks!
DynamoDOWN is opinionated about the structure of the table and may not work with a table it did not create. This is related to how DynamoDB uses primary keys and how to get "level-like" behavior. It should work fine in most cases, but read below for more info.
If the table doesn't exist, DynamoDOWN will try auto-create a table for you. You can specify the read/write throughput (if not specified it will default to 1/1). If the table already exists, the specified throughput will have no effect (existing table throughput can be changed using the AWS Console or SDK).
See LevelUP options for info on createIfMissing
and errorIfExists
options.
var levelup = require('levelup')
var db = levelup('table_name', {
db: require('dyanmodown'),
// required AWS config
dynamo: {
// Capacity can be specified, defaults to 1/1:
ProvisionedThroughput: {
ReadCapacityUnits: 1,
WriteCapacityUnits: 1
},
region: 'us-east-1',
secretAccessKey: 'abc',
accessKeyId: '123'
})
In DynamoDB, keys are made up of two parts, a hash key
and a range key
. To achieve leveldb-like behavior, all keys in a db instance are given the same hash key
(you can't do a range query over keys with different hash keys). The default hash key
for a db instance is !
, but can be specified with the location/table name argument:
var db = levelup('tableName/yourHashKey', {
db: require('dyanmodown'),
dynamo: {
// required AWS config
}
})
db.put('some-key', 'some-value', function(err) {
// the object in DynamoDB would look like this:
// {
// hkey: 'yourHashKey',
// rkey: 'some-key',
// value: 'some-value'
// }
})
If you're fine with sharing Capacity Units across db instances/applications, you can reuse the same tableName with different hash keys -- saving you from having to create a new table.
Big thanks to @nlf and his RiakDOWN module, @rvagg for LevelUP, and everyone else in the level ecosystem.
Database actions are performed using the AWS SDK for JavaScript.
MIT