DynamoDB Time Series

This module allows one to use a dynamodb table as a time-series database, allowing queries for time ranges, and allows keeping data from different sources in the same table.

Usage

1. Add this package and if necessary, a peer dependency of the AWS SDK to your package.json file.
   Note that AWS Lambda does not require the inclusion of the AWS SDK, it does that for you.

    "dependencies": {
        ...
        "dynamodb-timeseries": "^0.1.00",
        "aws-sdk: "^2.x",
        ...
    }
   

2. Add a require statement to your code:

    const DynamoDbTimeSeries = require('dynamodb-timeseries');
   

3. Create an instance of the API that attaches to a DynamoDB table:

    const tableName           = 'Your existing DynamoDB table name';
    const dbTimeSeriesOptions = {
        tableName: tableName,
        awsOptions: {
            credentials: {
                accessKeyId:     'An access key ID, typically created by STS',
                secretAccessKey: 'A secret access key, typically created by STS',
                sessionToken:    'A session token, typically created by STS',
                expiration:      'The credentials expiration, typically created by STS'
            }
        }
    };
    const dbApiInstance = Object.create(DynamoDbTimeSeries).setOptions(dbTimeSeriesOptions);
   

4. Create an event that will be written to the DynamoDB table:

    const event = {
        foo: 'bar'
    };
   

5. Write to the DynamoDB table:

    const userId        = 'Your user ID';
    const eventType     = 'Your event type';
    const epochTime     = new Date().getTime();
    await dbApiInstance.putEvent(userId, eventType, epochTime, event);
   

See files test/*.js for specific examples.

output format

[
  {
   // outer attributes are common no matter the manufacturer
   epochTimeMilliSec: ... per Javascript standard.  56 bit int or a 64 bit float ... ,
   mfgrId: '<ID of the manufacturer>'
   event: { ... event specific to the manufacturer, see their API ... },
  }
]

Prerequisites

• You must have access to AWS, including any credentials, IAM permissions, and region as required by the AWS SDK

Unit Tests

Prerequisites

• Set environment variable AWS_DEFAULT_REGION
• set environment variable export NPM_TOKEN=$(cat ~/.npm/token)
• Make your AWS credentials available in file ~/.aws

Interactive Invocation

1. Run Docker Compose:

    docker-compose run app  /bin/bash
   

2. Navigate to the directory where the source code is located:

    cd /home/code       
   

3. Run NPM install

    npm install
   

4. Run the unit tests

    npm test test/*.js
   

Automated Invocation

1. Set environment variable AWS_DEFAULT_REGION to your desired region:

    export AWS_DEFAULT_REGION=us-east-1
   

2. Run script test.sh

    ./test.sh
   

You should see a transcipt that looks similar to the following:

$ ./test.sh
audited 19 packages in 0.644s
found 0 vulnerabilities

up to date in 10.89s

> dynamodb-timeseries-test@0.0.1 test /home/code/test
> node_modules/tape/bin/tape "dynamodb-time-series.js"

TAP version 13
# create dynamoDB table for all tests. Use random keys to make tests independent
ok 1 table name is the random string created before the tests start
ok 2 table is active
# put and get a time series event
ok 3 put result is an empty object?... okay aws
ok 4 one item put, one item queried
ok 5 user ID and type match what was queried
ok 6 event contents queried is same as was put
# delete test table
ok 7 Table was deleted because we got correct exception trying to wait for it to delete

1..7
# tests 7
# pass  7

# ok