Hasura API Code Generator

npm module to generate code for Hasura APIs in different languages/frameworks

Languages Covered

• Python
• JavaScript
• Node.js
• Swift
• Java

Usage

npm install api-codegen

import {CodeGenerator} from 'api-codegen';

const moduleName = 'Node Fetch'; // replace with module of choice
let generator = new CodeGenerator(moduleName);
let generatedCode;

try {
  generatedCode = generator.generate({
    url: '', // replace with request URL
    method: '', // replace with request Method (GET, POST, PUT, DELETE)
    headers: {}, // add header key-value pairs
    param: {}, // add request body
    type: '' // specify 'file' for generating filestore API or leave it empty
  });
} catch (e) {
  generatedCode = e.message;
}
console.log(generatedCode); // Generated Code Snippet

If you want to add support for a language or a library within a language, follow the steps below.

Getting started

1. Build your library

• Run yarn install (recommended) or npm install to get the project's dependencies
• Run yarn build or npm run build to produce minified version of your library.

2. Development mode

• Having all the dependencies installed run yarn dev or npm run dev. This command will generate an non-minified version of your library and will run a watcher so you get the compilation on file change.

Scripts

• yarn build or npm run build - produces production version of your library under the lib folder
• yarn dev or npm run dev - produces development version of your library and runs a watcher
• yarn test or npm run test - well ... it runs the tests :)
• yarn test:watch or npm run test:watch - same as above but in a watch mode

Testing

• Once the code generator for the new module has been written, it can be tested by: sh test.sh '<MODULE_NAME>' <PATH_TO_DOCKERFILE>

Here the

• '<MODULE_NAME>' comes from the CodeGenerator.TYPES
• '<PATH_TO_DOCKERFILE>' should give the path of the Dockerfile written for this new module execution.

This shell script takes the module name, generates the Code Snippet based on the written function and saves the snippet file named as test in the given path. A docker image is built with the test file added. Docker container is run with the built image and is expected to only output the response of the sample http request.

Contributing

We love community contributions. If you would like to contribute, raise a PR after testing your code generation logic.