This package provides a simple framework for building high-performance ETL web servers in Go, compatible with AIStore.

import 	"github.com/NVIDIA/aistore/ext/etl/webserver"

type EchoServer struct {
	webserver.ETLServer
}

func (*EchoServer) Transform(input io.ReadCloser, _, _ string) (io.ReadCloser, error) {
	data, err := io.ReadAll(input)
	if err != nil {
		return nil, err
	}
	input.Close()
	return io.NopCloser(bytes.NewReader(data)), nil
}

func main() {
	svr := &EchoServer{}
	if err := webserver.Run(svr, "0.0.0.0", 8080); err != nil {
		log.Fatalf("Failed to start server: %v", err)
	}
}

FROM golang:1.26-alpine AS builder
RUN apk add --no-cache git
WORKDIR /app
COPY src/go.mod src/go.sum ./
RUN go mod download
COPY src/ ./
RUN go build -ldflags="-s -w" -o echo

FROM alpine:3.19
WORKDIR /app
COPY --from=builder /app/echo .
EXPOSE 80
ENTRYPOINT ["./echo"]

docker build -t <myrepo>/echo-etl:latest .
docker push <myrepo>/echo-etl:latest

# init_spec.yaml
apiVersion: v1
kind: Pod
metadata:
  name: etl-echo
  annotations:
    communication_type: "hpush://"
spec:
  containers:
    - name: server
      image: <myrepo>/echo-etl:latest
      ports: [{ name: default, containerPort: 8000 }]
      command: ["./echo", "-l", "0.0.0.0", "-p", "8000"]
      readinessProbe:
        httpGet: { path: /health, port: default }

ais etl init spec --name my-echo --from-file init_spec.yaml
ais etl bucket my-echo ais://<src-bucket> ais://<dst-bucket>

• Echo Returns the original input data on every request. Source
• Hello World Returns b"Hello World!" on every request. Source
• FFmpeg Transform audio files into WAV format with control over Audio Channels (AC) and Audio Rate (AR). Source

• GET /<object-path>: Fetch and transform an object
• PUT /<object-path>: Send object content to be transformed
• GET /health: Health check
• /ws: Establish WebSocket connection

• Direct PUT is supported: when AIS sends the ais-node-url header, the transformed result will be streamed directly to the destination.