Liviz

Liviz Framework for plotting during a debugging session. The plotter is written in VanillaJS and is provided by a Rest API using NodeJs and ExpressJS.

• This framework allows to display 2D curves in GNU Plot format (https://people.duke.edu/~hpgavin/gnuplot.html) incrementally, using the d3js library. The communication of the curve is done via a websocket chatting with a RabbitMQ server. For now, this plotter is used to follow the evolution of the value of variables resulting from the interpretation of a nablab program. The interpreter communicates the values which will be plotted in real time by the framework.

• The data exchanged between the model and the plotter goes through a Rest API accessible in Sources/ written in NodeJs. It is possible to send a plot configuration, and append data to the graphs. You can find exemples of api calls in Scripts/API/

URL	Method	Description	Error handling
/	GET	Returns a plotter as an html page
/config	PUT	Define or replace the configuration and generate queues UUIds	Returns 400 (Bad Request) if the configuration is malformated
/config	GET	Returns the configuration of the instance (json)	Returns 404 (Not found) If no configuration has been specified
/graphs/{name}	GET	Returns the UId of the queue according to the name of the graph.	Returns 404 (Not found) If the graph does not exist
/graphs/{name}	PUT	Append graph data to the specified graph.	Returns 404 (Not found) If the graph does not exist

Content

This repository contains:

• Scripts : Contains linux scripts (sudo required) allowing to launch rabbitmq and the plugins necessary for the proper functioning of StompJs
• Sources : The sources of the rest API
• Sources/html : The sources of the plotter
• Misc : Some random files related to the project (images, rawdata etc)
• org.gemoc.liviz.ui : A plugin for eclipse (gemoc) to configure liviz.
• org.gemoc.liviz.render : A plugin for eclipse (gemoc) to render liviz.

Dependencies

Name	Version	Install
RabbitMQ (Docker setup recommanded)	3.8.19	https://www.rabbitmq.com/download.html
node	14.17.1	https://nodejs.org/en/download/
Npm	6.14.13	Installed with nodejs
AMQP	*	npm install amqplib
ExpressJs	*	npm install express --save
body-parser	*	npm i body-parser
uuid	*	npm i uuid
StompJs	6.1.0	Included
D3Js	4.0	Included
Bootstrap	5.0.1	Included

Configuration

In order to plot your graphs. It is necessary to send a configuration file to the web api. Here is the diagram of the configuration to send:

{
  "$schema": "http://json-schema.org/draft-04/schema#",
  "type": "object",
  "properties": {
    "graphs": {
      "type": "array",
      "items": [
        {
          "type": "object",
          "properties": {
            "variables": {
              "type": "array",
              "items": [
                {
                  "type": "string"
                },
                {
                  "type": "string"
                }
              ]
            },
            "window": {
              "type": "string"
            },
            "name": {
              "type": "string"
            },
            "x": {
              "type": "string"
            },
            "type": {
              "type": "string"
            }
          },
          "required": [
            "variables",
            "window",
            "name",
            "x",
            "type"
          ]
        }
      ]
    }
  },
  "required": [
    "graphs"
  ]
}

Run

1. Install docker (https://docs.docker.com/engine/install/)

2. Install docker compose using this command :

sudo curl -L "https://github.com/docker/compose/releases/download/1.29.2/docker-compose-$(uname -s)-$(uname -m)" -o /usr/

3. Run the command sudo chmod +x /usr/local/bin/docker-compose

4. Run the command sudo docker-compose up at the repository root.

Build

• Run the command sudo docker-compose up --build to rebuild liviz api & plotter (server part).

• You can build the eclipse plugins using Gemoc Studio (https://gemoc.org/download.html)

Exemple of usage

1. Run Liviz using docker compose
2. Run Scripts/API/conf.sh
3. Run Script/API/data.sh
4. The plotter result is accessible at http://localhost:3000/