backstopmedia/grpc-book-example

gRPC-book-example

Based loosely on models from swapi.

Getting Started

• Install Golang
• Install protobuf compiler
• Clone this repo into your $GOPATH
• make setup

Lint proto with:

make lint

grpc-book-example's Issues

nouns should be plural

We should use a single noun-name (plural form) for http annotations.

now:

/people and /person/{ID}

should be:

/people/ and /people/{ID}

node stuff

Should we move /tools to a separate node-implementation? In addition to the tools for massaging the data from swapi, I've already got the full service implemented, including a RESTish HTTP1 gateway. I'm also fine with removing it, but it might be nice to keep it around for people implementing js-based gRPC services.

Streaming endpoint

We still need a streaming endpoint in this example repository.

Update to proto3

Can you please update the code? :)

Codecs and Compressors

Do you find it relevant to talk about codecs and compressors?
I've never used them, but some example about gzip compressing messages could be interesting, specially for an api as swapi, where there is a lot of plain text (film names and film descriptions) spending bandwidth on the net.
https://github.com/grpc/grpc-go/blob/master/Documentation/encoding.md

Well known types Timestamp for dates

we should be using Timestamp for fields that represent a date.

Getting Started Guide

New to Go. Couldn't follow through the Guide past make setup. Missing steps in the Guide.

make setup has the following error:

Synching dev tools and dependencies...
make: retool: No such file or directory
make: *** [setup] Error 1

Question: Can I use gRPC for ..

Can I use gRPC not for communication between different applications, but between two instances of the same application residing in different computers, for splitting a job achieving parallelization?
example:

1. App1: {job1, job2, job,...jobn} // run in serial the processing of job1 to jobn, for example using threads in batches
2. App1: { {job1, job_n1}, {job_n2, ... jobm1} , ...{ job_j, ... jobn}} send requests using gRPC
   run_to_remote_1({job1, job_n1})
   run_to_remote_2({job_n2, ... jobm1})
   ...
   run_to_remote_k({ job_j, ... jobn})

and then gather the results from each machine? Will this timeout?