Pravah.io

Pravah is a decentralised open source data exchange platform for smart cities. It's built on top of Libp2p to support highly available and super scalable peer-to-peer network of data nodes.

Pravah connects different stakeholders in a smart city like data producers and consumers to each other to enable a data economy, improve collaboration and simplify access to real-time data. High data quality, low latency, well-defined data sharing standards and easy deployability across different platforms are some of the design objectives. This would enable anyone to build applications on top of low latency data flowing from the smart city infrastructure.

Quick Starter Guide

Being a decentralised platform, a node needs to be deployed to interact with the Pravah network but it won't be a quick starter guide if you do the hard work. So, we did it for you. Let's directly dive into how you can subscribe and consume data from the network.

Before we move forward, you need to understand two important concepts:

1. A Channel describes the kind of data you want to consume. Every Channel will have an associated data standard, defined in protobuf. This will help you know how the incoming data would look even before you start consuming it. An example is /PublicBus channel which gives you GTFS-Realtime data.
2. Next up is a Geospace which describes the geobound of your dataset. Say you want realtime public buses data for Delhi, India then the geospace will be /in/delhi. Pravah combine the channel and the geospace (/PublicBus/in/delhi) to create a unique data stream. A list of all available channels and geospaces are available here

That's it for theory, let's dive into the code.

Use this key: 1cDpdkE2Qvq26OvUAxkXswmHcXEPUBS6Wcig6ERQvPrQ

# This example demonstrate how one can get access real-time data coming out of 
# various public buses from Delhi, India.

# Setting up
# pip3 install pravah

import pravah

p = pravah.Pravah(
    '/PublicBus', 
    endpoint='rpc.pravah.io:6666',
    auth_token='<PRAVAH_API_KEY>'
)

feed = p.subscribe('/in/delhi')

for message, geospace in feed:
    # message is a `FeedMessage` instance as given in https://github.com/pravahio/protocols/blob/master/protocols/gtfs/PublicBus.proto
    # As we can subscribe to multiple geospaces at once. `geospace` will help us identify geospace of the currently received message.
    
    # Print the entire FeedMessage.
    print(message)

    # Access the first entity and print it.
    print(m.entity[0])

Prerequisite

1. git
2. go >= v1.12.*
3. make
4. gcc

Installation

Note: You need not build it before installing. Program is build and installed in a single step.

git clone https://github.com/pravahio/go-mesh.git
cd go-mesh
make install

Creating a new account

An account is a public-private key pair which will be used to sign all the messages and transactions flowing out of your node.

You can create one by using the following command

mesh account -c -p -o account.msa

Config file

mesh has many configuration options all of which are available in mesh --help.

Instead of passing them as flags we can use a json based config file (config.json) as given below:

{
  "rnz": "publicBus",
  "account": "account.msa",
  "debug": "true"
}

This config file tells the mesh client to find more peers at publicBus rendezvous point, use account.msa account file and print all debug logs ("debug": "true").

A sample config file is avaiable here

For data subscribers

You can quickly test out the platform by following the guide in this repo.

Note that by following the above guide you won't be able to obtain realtime data and it should only be used to try out the platform.

To get realtime data you need to fire up mesh as a subscriber and connect via RPC.

1. Running mesh as a subscriber node

This can be done by running mesh with en-sub flag.

mesh -c docs/config.json --en-sub

This will do multiple things. First of all it will find other peers at rnz point and connect to them. It will then start an RPC server at default port 5555.

Note: If you want to use js client for mesh-gtfsr, you need to enable web RPC using --web-rpc flag. Without this you won't be able to connect to mesh RPC through a browser. Web RPC runs on a different port (Default: 5556)

2. Once mesh client is up and running we can connect with it via RPC

Assuming we want to consume GTFS-Realtime data. Python client for mesh-gtfsr can be found here.

Install it with

pip3 install mesh-gtfsr

Now we can start consuming GTFS-Realtime data for some specific geospace. Delhi in the given example. Here you can find a list of all available geospaces.

from pravah import Pravah

def main():
  m = Pravah(
    '/PublicBus', 
    endpoint='endpoint',
    auth_token='<PRAVAH_API_KEY>')

  feed = m.subscribe([
    '/in/delhi'
  ])
  
  # You get feed in 'f'
  # and geospace in 'g
  for f, g in feed:
    print('Geospace: ' + g)
    print('Feed: ' + str(f)) 

if "__main__" == __name__:
  main()

feed is a tuple of (feed:FeedMessage, geospace:string) as defined in GTFS-Realtime specs

If you want to get the data directly in the browser use the following guide.

For data publishers

1. Running mesh as a publisher node

This can be done by running mesh with en-pub flag.

mesh -c docs/config.json --en-pub

2. Once mesh client is up and running we can connect to it via RPC and start publishing.

Assuming we want to publish GTFS-Realtime data. Python client for mesh-gtfsr can be found here.

Install it with

pip3 install mesh-gtfsr

Now we can start publishing GTFS-Realtime data.

import time
from pravah import Pravah

GEOSPACE = [
  '/your/geospace/test'
]

def main():
  m = Pravah(
    '/PublicBusTest', 
    endpoint='endpoint',
    auth_token='<PRAVAH_API_KEY>')

  rt = {
    "header": {
      "timestamp": time.time()
    },
    "entity": [{
      "id": "VEHICLE_ID",
      "vehicle": {
          "position": {
          "latitude": 77.23524,
          "longitude": 22.35343
          }
      }
    }]
  }

  m.registerToPublish(GEOSPACE)
  m.publish(GEOSPACE, rt)

if "__main__" == __name__:
  main()