How to Read Lines from GZIP-Compressed Files in Go

read_gzip_lines.go

package main

import (
	"bufio"
	"compress/gzip"
	"fmt"
	"io"
	"log"
	"os"
)

// GZLines iterates over lines of a file that's gzip-compressed.
// Iterating lines of an io.Reader is one of those things that Go
// makes needlessly complex.
func GZLines(rawf io.Reader) (chan []byte, chan error, error) {
	rawContents, err := gzip.NewReader(rawf)
	if err != nil {
		return nil, nil, err
	}
	bufferedContents := bufio.NewReader(rawContents)
	ch := make(chan []byte)
	errs := make(chan error)
	go func(ch chan []byte, errs chan error, contents *bufio.Reader) {
		defer func(ch chan []byte, errs chan error) {
			close(ch)
			close(errs)
		}(ch, errs)
		for {
			line, err := contents.ReadBytes('\n')
			ch <- line
			if err != nil {
				if err != io.EOF {
					errs <- err
				}
				return
			}
		}
	}(ch, errs, bufferedContents)
	return ch, errs, nil
}

func main() {
	if len(os.Args) < 2 {
		log.Fatal("Usage: read file.gz")
	}
	filename := os.Args[1]
	fmt.Printf("Called on: %+v\n", filename)
	rawf, err := os.Open(filename)
	if err != nil {
		log.Fatal(err)
	}
	lines, errors, err := GZLines(rawf)
	if err != nil {
		log.Fatal(err)
	}
	for {
		select {
		case line := <-lines:
			os.Stdout.Write(line)
		case err := <-errors:
			log.Fatal(err)
		}
	}
}