I am creating a test Go HTTP server, and I am sending a response header of Transfer-Encoding: chunked so I can continually send new data as I retrieve it. This server should
It looks like httputil provides a NewChunkedReader function
The trick appears to be that you simply need to call Flusher.Flush() after each chunk is written. Note also that the "Transfer-Encoding" header will be handled by the writer implicitly, so no need to set it.
func main() {
http.HandleFunc("/", func(w http.ResponseWriter, r *http.Request) {
flusher, ok := w.(http.Flusher)
if !ok {
panic("expected http.ResponseWriter to be an http.Flusher")
}
w.Header().Set("X-Content-Type-Options", "nosniff")
for i := 1; i <= 10; i++ {
fmt.Fprintf(w, "Chunk #%d\n", i)
flusher.Flush() // Trigger "chunked" encoding and send a chunk...
time.Sleep(500 * time.Millisecond)
}
})
log.Print("Listening on localhost:8080")
log.Fatal(http.ListenAndServe(":8080", nil))
}
You can verify by using telnet:
$ telnet localhost 8080
Trying ::1...
Connected to localhost.
Escape character is '^]'.
GET / HTTP/1.1
HTTP/1.1 200 OK
Date: Tue, 02 Jun 2015 18:16:38 GMT
Content-Type: text/plain; charset=utf-8
Transfer-Encoding: chunked
9
Chunk #1
9
Chunk #2
...
You might need to do some research to verify that http.ResponseWriters support concurrent access for use by multiple goroutines.
Also, see this question for more information about the "X-Content-Type-Options" header.