Skip to content

Latest commit

 

History

History
171 lines (110 loc) · 5.73 KB

README.md

File metadata and controls

171 lines (110 loc) · 5.73 KB
Raw

AlgoSearch logo image

License

AlgoSearch (live deployment)

AlgoSearch enables you to explore and search the Algorand blockchain for transactions, addresses, blocks, assets, statistics, and more, in real-time. It's a simple, easy-to-deploy, and open-source block explorer to be used alongside an Algorand archival node.

Dependencies

  • Node.js 8+ for use with server and front-end.
  • go-algorand for Algorand goal node (must support archival indexing).
  • Algorand Indexer for reading committed blocks from the Algorand blockchain and maintains a database of transactions and accounts that are searchable and indexed.
  • CouchDB as database solution.

Work on AlgoSearch is funded by the Algorand Foundation through a grant to Anish Agnihotri. The scope of work includes the development of an open-source block explorer (AlgoSearch) and a WIP analytics platform.

The Fastest Approach

The fastest approach to set everything up for development is using the Sandbox and docker-compose. To do that, just setup the Sandbox and do the following:

# Start the services
bash docker-run.sh

Run locally

Linux / OSX

The go-algorand node currently aims to support only Linux and OSX environments for development.

Step 1: Environment setup

This section explains how to set up everything locally.

The Native Approach

Algorand's Node

First you'll need to install Algorand's Node locally. Follow the instructions through the hyperlink.

Make sure node is running on the preferred network and that algod details are correct in service/global.js.

Indexer

Then you'll need to install the Indexer locally. Follow the instructions through the hyperlink.

CouchDB

Finally you'll need to install CouchDB locally.

You can also run CouchDB using Docker easily:

# Create a folder called db-data
mkdir db-data

docker run -e COUCHDB_USER=admin -e COUCHDB_PASSWORD=password -p 5984:5984 --name my-couchdb -v $(pwd)/db-data:/opt/couchdb/data -d couchdb

If you are using docker compose to start the services, you can skip this step.

The Sandbox Approach

You can also set up Algorand's Node and the Indexer using Algorand's Sandbox. Follow the instructions through the hyperlink.

Note that you will still have to set up CouchDB if you are not using the docker-compose.yml offered here.

Step 2: Configuration

  1. Enter your site name in src/constants.js.
    • it's set to be http://localhost:8000 as default, but if you are changing the port, remember to update PORT in server.js and server.local.js
  2. Enter the API endpoint of the Algorand's Node in src/constants.js.
  3. Copy service/global.sample.js to service/global.js and enter your node and DB details.
    • If you are using the Sandbox approach, copy service/global.sandbox.js instead and update the CouchDB details if needed.

Step 3: Running AlgoSearch

The Native Approach

Install the dependencies

# Run in folder root directory
npm install

Build the code

# Run in folder root directory
npm run build

Run it

Make sure the configurations in your src/global.js is correct, then you'll have to do three things.

One, execute the following to create tables in CouchDB:

node service/sync/initSync.js

Second, execute the following to start syncing the tables:

node service/sync/syncAll.js

Note that this step takes time to sync and should stay running as long as the server is running.

Finally, start the server:

nodemon server.js

Docker Approach

You can skip the native approach entirely and simply start the application using Docker (remember to make sure your src/global.js is having the correct details):

# Build the image
docker build -t algosearch .

# Run the container
docker run algosearch

If you are using Linux and your container needs to access the host machine, for instance, the CouchDB you set up on your machine, run the following the start the container:

docker run --network="host" algosearch

Docker Compose Approach

To start the server using docker-compose, you only need the Node and Indexer, and use DB details in src/global.sandbox.js, that is, make sure

dbhost = 'couchdb.server:5984', // Database URL
dbuser = 'admin', // Database user
dbpass = 'password', // Database password

Then start the services:

# Create the folder for CouchDB
mkdir db-data

# Start the services
docker-compose up

If your Node and Indexer are on the host machine, your containers will have to access localhost, instead of using docker-compose up, run the startup script instead:

bash docker-run.sh

This script will find the IPs of localhost and to be accessed through dockerhost. In other words, in your src/global.js, use dockerhost instead of localhost.

Documentation

The Wiki is currently under construction.

License

License

Copyright (C) 2020, Anish Agnihotri.