Skip to content

Latest commit

 

History

History
166 lines (133 loc) · 6.39 KB

README.md

File metadata and controls

166 lines (133 loc) · 6.39 KB

Grpc-Json Transcoder project

Price version

This is a filter that allows a RESTful JSON API client (Ocelot Gateway) to send requests to .NET Web API (Aggregator Service) over HTTP and get proxied to a gRPC service (on the downstream).

This project is inspired by grpc-gateway which is totally for golang, grpc-dynamic-gateway is for nodejs. And especially, Envoy gRPC-JSON transcoder is the best of transcoding in this area, but it is only on the infrastructure level. You also can use it just like my project used at coolstore-microservices. We might use the approach from Microsoft at GrpcHttpApi, but it is not publish in the official release by Microsoft in the meantime.

gRPC parser borrows the idea from Ocelot.GrpcHttpGateway code-based.

Give a Star!

If you liked GrpcJsonTranscoder project or if it helped you, please give a star ⭐ for this repository. That will not only help strengthen our .NET community but also improve cloud-native apps development skills for .NET developers in around the world. Thank you very much 👍

Check out my blog or say hi on Twitter!

How to run it!

$ docker-compose up

or

$ bash
$ start.sh # I haven't done it yet :p

In the mean time, we open up the visual studio, run multiple projects included OcelotGateway, AggregationRestApi, ProductCatalogGrpcServer and GreatGrpcServer

Test it as below:

# gRPC
$ curl -X GET -H 'content-type: application/grpc' -k http://localhost:5000/say/Bob
$ {"Message":"Hello Bob"}
# gRPC
$ curl -X GET -H 'content-type: application/grpc' -k http://localhost:5000/products
$ {"Products":[{"Id":1,"Name":"product 1","Quantity":52,"Description":"description of product 1"},...]}
# gRPC
$ curl -X POST -H 'content-type: application/grpc' -d '{ "name": "product 1", "quantity": 1, "description": "this is product 1" }' -k http://localhost:5000/products
$ {"Product":{"Id":915,"Name":"product 1 created","Quantity":1,"Description":"this is product 1 created"}}
# REST Api
$ curl -X GET -H 'content-type: application/json' -k http://localhost:5000/weather 
$ [{"date":"2019-08-17T18:34:41.1090164+07:00","temperatureC":-6,"temperatureF":22,"summary":"Sweltering"},{"date":"2019-08-18T18:34:41.1090371+07:00","temperatureC":27,"temperatureF":80,"summary":"Hot"},{"date":"2019-08-19T18:34:41.1090499+07:00","temperatureC":33,"temperatureF":91,"summary":"Balmy"},{"date":"2019-08-20T18:34:41.1090617+07:00","temperatureC":-14,"temperatureF":7,"summary":"Chilly"},{"date":"2019-08-21T18:34:41.1090743+07:00","temperatureC":22,"temperatureF":71,"summary":"Hot"}]

Notes:

  • REST method: content-type: application/json is for REST method on the downstream services.
  • gRPC method: content-type: application/grpc is really important if you call to gRPC endpoint on the downstream services.

How to understand it!

The project aims to .NET community and its ecosystem which leverage the power of Ocelot Gateway which is very powerful in the gateway components were used by various of companies and sample source code when we try to adopt the microservices architecture project.

That's quite simple with only a few steps to make it work :) Create the .NET Core project with Ocelot in place, then put the configuration as below

{
  "ReRoutes": [
    {
      "UpstreamPathTemplate": "/say/{name}",
      "UpstreamHttpMethod": [ "Get" ],
      "DownstreamPathTemplate": "/Greet.Greeter/SayHello",
      "DownstreamScheme": "http",
      "DownstreamHostAndPorts": [
        {
          "Host": "127.0.0.1",
          "Port": 5003
        }
      ]
    },
    {
      "UpstreamPathTemplate": "/products",
      "UpstreamHttpMethod": [ "Get" ],
      "DownstreamPathTemplate": "/ProductCatalog.Product/GetProducts",
      "DownstreamScheme": "http",
      "DownstreamHostAndPorts": [
        {
          "Host": "127.0.0.1",
          "Port": 5002
        }
      ]
    },
    {
      "UpstreamPathTemplate": "/products",
      "UpstreamHttpMethod": [ "Post" ],
      "DownstreamPathTemplate": "/ProductCatalog.Product/CreateProduct",
      "DownstreamScheme": "http",
      "DownstreamHostAndPorts": [
        {
          "Host": "127.0.0.1",
          "Port": 5002
        }
      ]
    },
    {
      "UpstreamPathTemplate": "/weather",
      "UpstreamHttpMethod": [ "Get" ],
      "DownstreamPathTemplate": "/weatherforecast",
      "DownstreamScheme": "http",
      "DownstreamHostAndPorts": [
        {
          "Host": "localhost",
          "Port": 5001
        }
      ]
    }
  ],
  "GlobalConfiguration": {
  }
}

Then in code Program.cs, you only put a few line

var configuration = new OcelotPipelineConfiguration
{
    PreQueryStringBuilderMiddleware = async (ctx, next) =>
    {
        await ctx.HandleGrpcRequestAsync(next);
    }
};

app.UseOcelot(configuration).Wait();

Don't believe what I said. Try it!

We haven't tested it with stream and full-duplex transport protocols yet. So we feel free to contribute by the .NET community.

Contributing

  1. Fork it!
  2. Create your feature branch: git checkout -b my-new-feature
  3. Commit your changes: git commit -am 'Add some feature'
  4. Push to the branch: git push origin my-new-feature
  5. Submit a pull request :p