I have a handler function which looks like below:
func DownloadFileHandler(w http.ResponseWriter, r *http.Request) {
SetResponseHeaders(w, r)
if r.Method == "OPTIONS" {
return
}
var err error
var responseCode int = http.StatusBadRequest
vars := mux.Vars(r)
file := vars["file"]
var n int64 = 0
var reader io.Reader
Filename := "images/" + file
f, err := os.Open(Filename)
if err != nil {
err = errors.New("File not found")
goto Error
}
reader = bufio.NewReader(f)
settings.WriteDebugLog(time.Now())
n, err = io.Copy(w, reader) // this takes a long time if the file is large
settings.WriteDebugLog(time.Now())
settings.WriteDebugLog(n / 1024 / 1024)
responseCode = http.StatusOK
settings.WriteLog(r, responseCode, nil, GetCurrentFuncName())
return
Error:
WriteErrorResponse(w, responseCode, err)
settings.WriteLog(r, responseCode, err, GetCurrentFuncName())
}
But if it is a large file (more than 1GB), the io.Copy takes quite a long time (best case: 1G 13 seconds shown below) and the download process in browser doesn't show up until the process of io.Copy completes. Is there a way to achieve my goal more efficiently? I want the download process to show up in my browser right after I send the download request.
2017/05/16 22:32:29 Debug: 2017-05-16 22:32:29.967499719 +0900 JST
2017/05/16 22:32:36 Debug: 2017-05-16 22:32:36.980914163 +0900 JST
2017/05/16 22:32:36 Debug: 1097
I tried calling io.Copy with "go io.Copy" then the download process shows up soon but the file size is 0.