Catly

A file uploading service for your finest cat pictures 🐱

It currently supports the jpeg, png and gif image formats and has configurable storage backends.

Requirements

To run the client and server, you will first need to ensure that you have installed:

To get started you will need to build and run the client and server.

To run the docker version of this server, run these commands in the root of this repo:

λ docker build -t catly-server .
λ docker run -it -p "8000:8000" -p "8080:8080" catly-server

To compile the client, you will need to run:

λ go build -o grpc-upload cmd/client/main.go

You can then test an upload with:

λ ./grpc-upload ./cat.jpg

There a number of different options that can be supplied when running the client and the server

When starting the container via docker, any of the following environment variables can be passed in:

Name	Description	Default
CATLY_DOMAIN	The domain that you are running the service under	`http://127.0.0.1`
CATLY_HTTP_PORT	The port the HTTP service will run on	`8080`
CATLY_GRPC_PORT	The port the gRPC upload service will run on	`8000`
CATLY_STORAGE_PATH	The storage path in the container you wish to use. By default, only in memory storage will be used	`:memory:`
CATLY_MAX_REQUEST_SIZE	The maximum request size in bytes the server will accept. This can be used to restrict large files from being uploaded	`8388608` (~ 8MB)

The client supports the following flags:

Name	Description	Default
-server	Specifies the address for the catly server	`127.0.0.1:8000`

The project is broken into the following components

Package	Description
api	Contains an implementation of an HTTP server for serving files and a gRPC server for handling uploads
cmd/server	Contains the main setup logic for the gRPC/HTTP server
cmd/client	Contains the main setup logic for the gRPC upload client
protocol	Contains the protobuf bindings and definitions for the object service
storage	Contains different storage implementations for catly server. Currently there is an in memory store, as well as a filesystem store

Integration testing for client and server
CI stage for linting
Improved error messages in responses
Support for HTTP and secure gRPC
LRU cache for improving performance when serving popular images from file storage
Generate prebuilt server and client for github releases