zip extractor

Package to make zip extraction easy (less codes) and efficient.

$ go get github.com/koron-go/zipx

import (
    "github.com/koron-go/zipx"
    "context"
)

func main() {
    err := zipx.New().ExtractFile(context.Background(), "archive.zip", zipx.Dir("outdir"))
    if err != nil {
        panic(err)
    }
}

See document for references.

Features

• Concurrent extraction

  • Auto concurrency. Use runtime.NumCPU() as default.

  • Configurable by manual

    // no concurrency (=sequential)
    x1 := zipx.New().WithConcurrency(1)
    
    // full concurrency, no limits. it would be slow.
    x0 := zipx.New().WithConcurrency(0)

• Customizable progress monitor

  var lastPercent = -1
  
  func monitor(p zipx.Progress) {
      curr := (p.NumDone * 10 / p.NumTotal) * 10
      if curr > lastPercent {
          lastPercent = curr
          report.Printf("progress %d%%", curr)
      }
  }
  
  func main() {
      err := zipx.New().
          WithMonitor(zipx.MonitorFunc(monitor)).
          ExtractFile(context.Background(), "archive.zip", zipx.Dir("outdir"))
      // ...(snip)...
  }

• Configurable zipx.Destination

  • Where to extract files are configurable by implementing zipx.Destination interface.

  • zipx provides two zipx.Destination implementaions.

    • for files: zipx.Dir
    • black hole: zipx.Discard

  • Using zipx.Dir, you can support names with other encodings rather than "UTF-8" easily by configuring zip.DefaultReinterpreter.

    // Example to support ShiftJIS (Windows or so)
    import "golang.org/x/text/encoding/japanese"
    
    func shiftjis(s string) (string, error) {
        d := japanese.ShiftJIS.NewDecoder()
        return d.String(s)
    }
    
    zipx.DefaultReinterpreter = ReinterpretFunc(shiftjis)
    
    // ...(snip)...
    
    err := zipx.New().ExtractFile(context.Background(), "archive.zip",
        zipx.Dir("outdir"))
    // ...(snip)...

• Cancelable by context.Context

  // Example to cancel extraction after 5 seconds.
  ctx, cancel := context.WithCancel(context.Background())
  defer cancel()
  go func() {
      time.Sleep(5 * time.Second)
      cancel()
  }()
  
  err := zipx.New().ExtractFile(ctx, "archive.zip", zipx.Dir("outdir"))
  if err != nil {
      // this would be "context canceled", if the extraction took over 5
      // seconds.
      log.Print(err)
  }

Why use zipx ?

golang では archive/zip 使えば容易にZIPの解凍は実装できます。しかしファイル 作ったり並列にして速度だしたりプログレスだしたりといったよくある実装は、自分で やらなきゃいけなくてちょっと面倒でした。

zipx ではZIPの解凍で頻繁に必要になるそれらの手続きをパッケージングしました。