API processor Docker Image

This action does:

generates Go stubs from protobuf specifications
generates OpenAPI specification from protobuf specifications
generate JS stubs from OpenAPI specification

Inputs

`proto_spec_dir_location`

Optional
Description:
Default value: the dir with protobuf/Open API specifications.

NB: the processor is looking for api.proto / openapi.yml files ONLY.

Example usage

name: api-generate

on: [ push ]

jobs:
  build-test:
    runs-on: ubuntu-18.04
    steps:
      - name: src-checkout  # here we check out our source dir
        uses: actions/checkout@v2
      - name: run-api-processor # here we use the processor
        uses: fedor-malyshkin/api-processor@master
        with:
          github_token: ${{ secrets.GITHUB_TOKEN }}

Workflow inside

An important thing here is a VERSION file that lies inside API project's file. It keeps a string of form 1.2.3 or something similar for master, and 1.2.3-beta15 for develop.

When you set up the processor as shown in the snippet you should stick to the usage of several branches:

develop - you merge here all your personal branches, when you do it the prerelease mark will be updated in the git: 1.2.3-beta15 -> 1.2.3-beta16 and the new tag 1.2.3-beta16 will be created.
master/main - the main branch where source of truth lies, you should mergge here all you changes from develop. In this case VERSION will be updated to release stat with patch pat increased: 1.2.3-beta15 -> 1.2.4. The tag with this value will be created and the same update will be done in develop brunch.
any other branches - any of your personal branch

Image preparation

You can look at base image and how it was created here: base docker image

fedor-malyshkin/api-processor