Rest API

You may find in Postman Collection the collection of our Rest API. You may find instructions as to how to import the postman collection here. There are two endpoints exposed: Finetune and Predict The code to be used for training or to highlight must be given base64encoded.

Wiki

We heave also created a Github Wiki with important documents such as the documentation of testing and project organisation. You can find it here: https://github.com/fayepsr/ase/wiki The wiki also contains:

• Project Organisation: https://github.com/fayepsr/ase/wiki/Project-Organization
• Testing: https://github.com/fayepsr/ase/wiki/Testing
• Microservice Overview: https://github.com/fayepsr/ase/wiki/Microservice-Overview

Example api calls

This comment would not exist in a real application. The secret to use is: hsdiwu8&%$$

Finetune

curl --location --request POST 'http://localhost:8089/api/v1/finetune'
--form 'lang="python"'
--form 'code="cHJpbnQoIjEyMyIp"'
--form 'secret="XXXXXXXX"'

cHJpbnQoIjEyMyIp is the python code base64encoded

Predict

curl --location --request POST 'http://localhost:8089/api/v1/highlight'
--form 'lang="python"'
--form 'code="cHJpbnQoIjEyMyIp"'
--form 'secret="XXXXXXXX"'
--form 'mode="json"'

• cHJpbnQoIjEyMyIp is the python code base64encoded
• mode: "json" or "html" The html mode will return a full HTML document with agiven HTML class for each type of token. The json mode will return a json array as shown in the example below. Each element of the array is a token.
  • startIndex, endIndex: refer to the position of the token in the code
  • token: is the token itself
  • type: ["GENERAL", "KEYWORD", "LITERAL", "CHAR_STRING_LITERAL", "COMMENT", "CLASS_DECLARATOR", "FUNCTION_DECLARATOR", "VARIABLE_DECLARATOR", "TYPE_IDENTIFIER", "FUNCTION_IDENTIFIER", "FIELD_IDENTIFIER", "ANNOTATION_DECLARATOR", "UNKONOWN"]
  • Example [ { "startIndex": 0, "endIndex": 5, "type": "KEYWORD", "token": "import" }, { "startIndex": 7, "endIndex": 10, "type": "GENERAL", "token": "java" }, .... ]

Docker

To compose the containers navigate inside src and run .\build_and_run_all.bat (for Windows Environments).

If not in a Windows Environment, you may run the following commands:

cd src 
docker-compose down
docker image rm src_php:latest
docker image rm src_learner:latest
docker image rm src_react:latest
cd ./src_react
docker build -f ./dockerfile -t src_react:latest . 
cd ../src_php
docker build -f ./dockerfile -t src_php:latest .
cd ../src_learner
docker build -f ./dockerfile -t src_learner:latest .
cd ../
docker-compose up

Our demo (docker-compose)

You can start the demo by navigating with the command line to the "src" folder in this repo and then type "docker-compose up". The yaml file is called "docker-compose.yml".

Our private domain for highlighting logic learner: http://learner:9007 Our public domain: http://localhost:8089/api/v1 Our demo frontend: http://localhost:3007

Our demo (AWS lightsail)

Our public domain: https://ase-service-1.iugkfeabdb168.eu-central-1.cs.amazonlightsail.com/
Our demo frontend: https://container-service-2.iugkfeabdb168.eu-central-1.cs.amazonlightsail.com/

We have scheduled an event using AWS eventbridge and the finetune function is called every 30 minutes.

Our Containers

• React
• PHP
• python_JAVA