Runtime server documentation

Description

The Runtime server is the backend for running the generated python code. It is a WebSocket server that runs in the backend and cant be accessed directly by the end-User.

Run

Run the following command in the root directory of the project to run the server and let it listen to port 8765

python app_server.py

Once the server is running then we would need to send requests and get back the results.

Testing

Run the example clinet.html to see a demo of how the commands work

The client sending the request to the server should open a websocket as follows:

websocket = new WebSocket("ws://127.0.0.1:6789/");

Note: each request should be sent via a JSON String which contains the action required and the required attributes. An example:

{action: 'Command',"Required Attribute":{Required Value}}

Following we present the list of commands available and what values should be given:

List of the available request to the Runtime service

Asynchronous run of the generated Code

• action: run

• Required Attribute: python

  • Required Value: A dictionary containing:
    • Number: number of files to be processed
    • files: list of file names and content
    • mainfile: the name of the main file

Example:

{action: 'run',"python":{number:2,files:[{filename:'speedmain.py',content:content1.value},
                {filename:'speedPrediction2.py',content:content2.value}], mainfile:'speedmain.py'}}

Get a saved PlaceHolder

• action: get_placeholder

• Required Attribute: placeholder

  • Required Value: A dictionary containing:
    • sessionId: session Id of the run did
    • name: name of the placeholder

{action: 'get_placeholder',"placeholder":{sessionId:123123123123,name:"message"}}

Run Model

• action: run_model

• Required Attribute: model

  • Required Value: A dictionary containing:
    • sessionId: session Id of the run done
    • name: name of the model
    • is_data : get the data or a filepath
    • data: new instance that the model is required to predict

Get statistics

• action: get_stats

• Required Attribute: placeholder

  • Required Value: A dictionary containing:
    • sessionId: session Id of the run done
    • name: name of the data placeholder we want statistics on

{action: 'get_stats',"placeholder":{sessionId:123123123123,name:"TrainX"}}

Check status

• action: status

• Required Attribute: sessionId

  • Required Value: the value of session ID

Stop Run

Not yet available

• action: stop

• Required Attribute: sessionId

  • Required Value: the value of session ID

Other requests to be developed and more info about the processes backend communication you can see it in here