1630848852
Fast HTTP implementation for Go.
Currently fasthttp is successfully used by VertaMedia in a production serving up to 200K rps from more than 1.5M concurrent keep-alive connections per physical server.
HTTP server performance comparison with net/http
In short, fasthttp server is up to 10 times faster than net/http. Below are benchmark results.
GOMAXPROCS=1
net/http server:
$ GOMAXPROCS=1 go test -bench=NetHTTPServerGet -benchmem -benchtime=10s
BenchmarkNetHTTPServerGet1ReqPerConn 1000000 12052 ns/op 2297 B/op 29 allocs/op
BenchmarkNetHTTPServerGet2ReqPerConn 1000000 12278 ns/op 2327 B/op 24 allocs/op
BenchmarkNetHTTPServerGet10ReqPerConn 2000000 8903 ns/op 2112 B/op 19 allocs/op
BenchmarkNetHTTPServerGet10KReqPerConn 2000000 8451 ns/op 2058 B/op 18 allocs/op
BenchmarkNetHTTPServerGet1ReqPerConn10KClients 500000 26733 ns/op 3229 B/op 29 allocs/op
BenchmarkNetHTTPServerGet2ReqPerConn10KClients 1000000 23351 ns/op 3211 B/op 24 allocs/op
BenchmarkNetHTTPServerGet10ReqPerConn10KClients 1000000 13390 ns/op 2483 B/op 19 allocs/op
BenchmarkNetHTTPServerGet100ReqPerConn10KClients 1000000 13484 ns/op 2171 B/op 18 allocs/op
fasthttp server:
$ GOMAXPROCS=1 go test -bench=kServerGet -benchmem -benchtime=10s
BenchmarkServerGet1ReqPerConn 10000000 1559 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet2ReqPerConn 10000000 1248 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet10ReqPerConn 20000000 797 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet10KReqPerConn 20000000 716 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet1ReqPerConn10KClients 10000000 1974 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet2ReqPerConn10KClients 10000000 1352 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet10ReqPerConn10KClients 20000000 789 ns/op 2 B/op 0 allocs/op
BenchmarkServerGet100ReqPerConn10KClients 20000000 604 ns/op 0 B/op 0 allocs/op
GOMAXPROCS=4
net/http server:
$ GOMAXPROCS=4 go test -bench=NetHTTPServerGet -benchmem -benchtime=10s
BenchmarkNetHTTPServerGet1ReqPerConn-4 3000000 4529 ns/op 2389 B/op 29 allocs/op
BenchmarkNetHTTPServerGet2ReqPerConn-4 5000000 3896 ns/op 2418 B/op 24 allocs/op
BenchmarkNetHTTPServerGet10ReqPerConn-4 5000000 3145 ns/op 2160 B/op 19 allocs/op
BenchmarkNetHTTPServerGet10KReqPerConn-4 5000000 3054 ns/op 2065 B/op 18 allocs/op
BenchmarkNetHTTPServerGet1ReqPerConn10KClients-4 1000000 10321 ns/op 3710 B/op 30 allocs/op
BenchmarkNetHTTPServerGet2ReqPerConn10KClients-4 2000000 7556 ns/op 3296 B/op 24 allocs/op
BenchmarkNetHTTPServerGet10ReqPerConn10KClients-4 5000000 3905 ns/op 2349 B/op 19 allocs/op
BenchmarkNetHTTPServerGet100ReqPerConn10KClients-4 5000000 3435 ns/op 2130 B/op 18 allocs/op
fasthttp server:
$ GOMAXPROCS=4 go test -bench=kServerGet -benchmem -benchtime=10s
BenchmarkServerGet1ReqPerConn-4 10000000 1141 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet2ReqPerConn-4 20000000 707 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet10ReqPerConn-4 30000000 341 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet10KReqPerConn-4 50000000 310 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet1ReqPerConn10KClients-4 10000000 1119 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet2ReqPerConn10KClients-4 20000000 644 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet10ReqPerConn10KClients-4 30000000 346 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet100ReqPerConn10KClients-4 50000000 282 ns/op 0 B/op 0 allocs/op
HTTP client comparison with net/http
In short, fasthttp client is up to 10 times faster than net/http. Below are benchmark results.
GOMAXPROCS=1
net/http client:
$ GOMAXPROCS=1 go test -bench='HTTPClient(Do|GetEndToEnd)' -benchmem -benchtime=10s
BenchmarkNetHTTPClientDoFastServer 1000000 12567 ns/op 2616 B/op 35 allocs/op
BenchmarkNetHTTPClientGetEndToEnd1TCP 200000 67030 ns/op 5028 B/op 56 allocs/op
BenchmarkNetHTTPClientGetEndToEnd10TCP 300000 51098 ns/op 5031 B/op 56 allocs/op
BenchmarkNetHTTPClientGetEndToEnd100TCP 300000 45096 ns/op 5026 B/op 55 allocs/op
BenchmarkNetHTTPClientGetEndToEnd1Inmemory 500000 24779 ns/op 5035 B/op 57 allocs/op
BenchmarkNetHTTPClientGetEndToEnd10Inmemory 1000000 26425 ns/op 5035 B/op 57 allocs/op
BenchmarkNetHTTPClientGetEndToEnd100Inmemory 500000 28515 ns/op 5045 B/op 57 allocs/op
BenchmarkNetHTTPClientGetEndToEnd1000Inmemory 500000 39511 ns/op 5096 B/op 56 allocs/op
fasthttp client:
$ GOMAXPROCS=1 go test -bench='kClient(Do|GetEndToEnd)' -benchmem -benchtime=10s
BenchmarkClientDoFastServer 20000000 865 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd1TCP 1000000 18711 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd10TCP 1000000 14664 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd100TCP 1000000 14043 ns/op 1 B/op 0 allocs/op
BenchmarkClientGetEndToEnd1Inmemory 5000000 3965 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd10Inmemory 3000000 4060 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd100Inmemory 5000000 3396 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd1000Inmemory 5000000 3306 ns/op 2 B/op 0 allocs/op
GOMAXPROCS=4
net/http client:
$ GOMAXPROCS=4 go test -bench='HTTPClient(Do|GetEndToEnd)' -benchmem -benchtime=10s
BenchmarkNetHTTPClientDoFastServer-4 2000000 8774 ns/op 2619 B/op 35 allocs/op
BenchmarkNetHTTPClientGetEndToEnd1TCP-4 500000 22951 ns/op 5047 B/op 56 allocs/op
BenchmarkNetHTTPClientGetEndToEnd10TCP-4 1000000 19182 ns/op 5037 B/op 55 allocs/op
BenchmarkNetHTTPClientGetEndToEnd100TCP-4 1000000 16535 ns/op 5031 B/op 55 allocs/op
BenchmarkNetHTTPClientGetEndToEnd1Inmemory-4 1000000 14495 ns/op 5038 B/op 56 allocs/op
BenchmarkNetHTTPClientGetEndToEnd10Inmemory-4 1000000 10237 ns/op 5034 B/op 56 allocs/op
BenchmarkNetHTTPClientGetEndToEnd100Inmemory-4 1000000 10125 ns/op 5045 B/op 56 allocs/op
BenchmarkNetHTTPClientGetEndToEnd1000Inmemory-4 1000000 11132 ns/op 5136 B/op 56 allocs/op
fasthttp client:
$ GOMAXPROCS=4 go test -bench='kClient(Do|GetEndToEnd)' -benchmem -benchtime=10s
BenchmarkClientDoFastServer-4 50000000 397 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd1TCP-4 2000000 7388 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd10TCP-4 2000000 6689 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd100TCP-4 3000000 4927 ns/op 1 B/op 0 allocs/op
BenchmarkClientGetEndToEnd1Inmemory-4 10000000 1604 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd10Inmemory-4 10000000 1458 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd100Inmemory-4 10000000 1329 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd1000Inmemory-4 10000000 1316 ns/op 5 B/op 0 allocs/op
Install
go get -u github.com/valyala/fasthttp
Switching from net/http to fasthttp
Unfortunately, fasthttp doesn't provide API identical to net/http. See the FAQ for details. There is net/http -> fasthttp handler converter, but it is better to write fasthttp request handlers by hand in order to use all of the fasthttp advantages (especially high performance :) ).
Important points:
Fasthttp works with RequestHandler functions instead of objects implementing Handler interface. Fortunately, it is easy to pass bound struct methods to fasthttp:
type MyHandler struct {
foobar string
}
// request handler in net/http style, i.e. method bound to MyHandler struct.
func (h *MyHandler) HandleFastHTTP(ctx *fasthttp.RequestCtx) {
// notice that we may access MyHandler properties here - see h.foobar.
fmt.Fprintf(ctx, "Hello, world! Requested path is %q. Foobar is %q",
ctx.Path(), h.foobar)
}
// request handler in fasthttp style, i.e. just plain function.
func fastHTTPHandler(ctx *fasthttp.RequestCtx) {
fmt.Fprintf(ctx, "Hi there! RequestURI is %q", ctx.RequestURI())
}
// pass bound struct method to fasthttp
myHandler := &MyHandler{
foobar: "foobar",
}
fasthttp.ListenAndServe(":8080", myHandler.HandleFastHTTP)
// pass plain function to fasthttp
fasthttp.ListenAndServe(":8081", fastHTTPHandler)
The RequestHandler accepts only one argument - RequestCtx. It contains all the functionality required for http request processing and response writing. Below is an example of a simple request handler conversion from net/http to fasthttp.
// net/http request handler
requestHandler := func(w http.ResponseWriter, r *http.Request) {
switch r.URL.Path {
case "/foo":
fooHandler(w, r)
case "/bar":
barHandler(w, r)
default:
http.Error(w, "Unsupported path", http.StatusNotFound)
}
}
// the corresponding fasthttp request handler
requestHandler := func(ctx *fasthttp.RequestCtx) {
switch string(ctx.Path()) {
case "/foo":
fooHandler(ctx)
case "/bar":
barHandler(ctx)
default:
ctx.Error("Unsupported path", fasthttp.StatusNotFound)
}
}
Fasthttp allows setting response headers and writing response body in an arbitrary order. There is no 'headers first, then body' restriction like in net/http. The following code is valid for fasthttp:
requestHandler := func(ctx *fasthttp.RequestCtx) {
// set some headers and status code first
ctx.SetContentType("foo/bar")
ctx.SetStatusCode(fasthttp.StatusOK)
// then write the first part of body
fmt.Fprintf(ctx, "this is the first part of body\n")
// then set more headers
ctx.Response.Header.Set("Foo-Bar", "baz")
// then write more body
fmt.Fprintf(ctx, "this is the second part of body\n")
// then override already written body
ctx.SetBody([]byte("this is completely new body contents"))
// then update status code
ctx.SetStatusCode(fasthttp.StatusNotFound)
// basically, anything may be updated many times before
// returning from RequestHandler.
//
// Unlike net/http fasthttp doesn't put response to the wire until
// returning from RequestHandler.
}
Fasthttp doesn't provide ServeMux, but there are more powerful third-party routers and web frameworks with fasthttp support:
Net/http code with simple ServeMux is trivially converted to fasthttp code:
// net/http code
m := &http.ServeMux{}
m.HandleFunc("/foo", fooHandlerFunc)
m.HandleFunc("/bar", barHandlerFunc)
m.Handle("/baz", bazHandler)
http.ListenAndServe(":80", m)
// the corresponding fasthttp code
m := func(ctx *fasthttp.RequestCtx) {
switch string(ctx.Path()) {
case "/foo":
fooHandlerFunc(ctx)
case "/bar":
barHandlerFunc(ctx)
case "/baz":
bazHandler.HandlerFunc(ctx)
default:
ctx.Error("not found", fasthttp.StatusNotFound)
}
}
fasthttp.ListenAndServe(":80", m)
net/http -> fasthttp conversion table:
var (
w http.ResponseWriter
r *http.Request
ctx *fasthttp.RequestCtx
)
VERY IMPORTANT! Fasthttp disallows holding references to RequestCtx or to its' members after returning from RequestHandler. Otherwise data races are inevitable. Carefully inspect all the net/http request handlers converted to fasthttp whether they retain references to RequestCtx or to its' members after returning. RequestCtx provides the following band aids for this case:
Use this brilliant tool - race detector - for detecting and eliminating data races in your program. If you detected data race related to fasthttp in your program, then there is high probability you forgot calling TimeoutError before returning from RequestHandler.
Blind switching from net/http to fasthttp won't give you performance boost. While fasthttp is optimized for speed, its' performance may be easily saturated by slow RequestHandler. So profile and optimize your code after switching to fasthttp. For instance, use quicktemplate instead of html/template.
See also fasthttputil, fasthttpadaptor and expvarhandler.
Performance optimization tips for multi-core systems
Fasthttp best practices
Tricks with []byte buffers
The following tricks are used by fasthttp. Use them in your code too.
var (
// both buffers are uninitialized
dst []byte
src []byte
)
dst = append(dst, src...) // is legal if dst is nil and/or src is nil
copy(dst, src) // is legal if dst is nil and/or src is nil
(string(src) == "") // is true if src is nil
(len(src) == 0) // is true if src is nil
src = src[:0] // works like a charm with nil src
// this for loop doesn't panic if src is nil
for i, ch := range src {
doSomething(i, ch)
}
So throw away nil checks for []byte buffers from you code. For example,
srcLen := 0
if src != nil {
srcLen = len(src)
}
becomes
srcLen := len(src)
String may be appended to []byte buffer with append
dst = append(dst, "foobar"...)
[]byte buffer may be extended to its' capacity.
All fasthttp functions accept nil []byte buffer
statusCode, body, err := fasthttp.Get(nil, "http://google.com/")
uintBuf := fasthttp.AppendUint(nil, 1234)
Related projects
FAQ
Why creating yet another http package instead of optimizing net/http?
Because net/http API limits many optimization opportunities. For example:
Why fasthttp API is incompatible with net/http?
Because net/http API limits many optimization opportunities. See the answer above for more details. Also certain net/http API parts are suboptimal for use:
Why fasthttp doesn't support HTTP/2.0 and WebSockets?
HTTP/2.0 support is in progress. WebSockets has been done already. Third parties also may use RequestCtx.Hijack for implementing these goodies.
Are there known net/http advantages comparing to fasthttp?
Yes:
Why fasthttp API prefers returning []byte instead of string?
Because []byte to string conversion isn't free - it requires memory allocation and copy. Feel free wrapping returned []byte result into string() if you prefer working with strings instead of byte slices. But be aware that this has non-zero overhead.
Which GO versions are supported by fasthttp?
Go1.5+. Older versions won't be supported, since their standard package miss useful functions.
NOTE: Go 1.9.7 is the oldest tested version. We recommend you to update as soon as you can. As of 1.11.3 we will drop 1.9.x support.
Please provide real benchmark data and server information
See this issue.
Are there plans to add request routing to fasthttp?
There are no plans to add request routing into fasthttp. Use third-party routers and web frameworks with fasthttp support:
I detected data race in fasthttp!
Cool! File a bug. But before doing this check the following in your code:
I didn't find an answer for my question here
Try exploring these questions.
Download Details:
Author: valyala
Download Link: Download The Source Code
Official Website: https://github.com/valyala/fasthttp
License: MIT
#go #golang #programming #developer
1599854400
Go announced Go 1.15 version on 11 Aug 2020. Highlighted updates and features include Substantial improvements to the Go linker, Improved allocation for small objects at high core counts, X.509 CommonName deprecation, GOPROXY supports skipping proxies that return errors, New embedded tzdata package, Several Core Library improvements and more.
As Go promise for maintaining backward compatibility. After upgrading to the latest Go 1.15 version, almost all existing Golang applications or programs continue to compile and run as older Golang version.
#go #golang #go 1.15 #go features #go improvement #go package #go new features
1630848852
Fast HTTP implementation for Go.
Currently fasthttp is successfully used by VertaMedia in a production serving up to 200K rps from more than 1.5M concurrent keep-alive connections per physical server.
HTTP server performance comparison with net/http
In short, fasthttp server is up to 10 times faster than net/http. Below are benchmark results.
GOMAXPROCS=1
net/http server:
$ GOMAXPROCS=1 go test -bench=NetHTTPServerGet -benchmem -benchtime=10s
BenchmarkNetHTTPServerGet1ReqPerConn 1000000 12052 ns/op 2297 B/op 29 allocs/op
BenchmarkNetHTTPServerGet2ReqPerConn 1000000 12278 ns/op 2327 B/op 24 allocs/op
BenchmarkNetHTTPServerGet10ReqPerConn 2000000 8903 ns/op 2112 B/op 19 allocs/op
BenchmarkNetHTTPServerGet10KReqPerConn 2000000 8451 ns/op 2058 B/op 18 allocs/op
BenchmarkNetHTTPServerGet1ReqPerConn10KClients 500000 26733 ns/op 3229 B/op 29 allocs/op
BenchmarkNetHTTPServerGet2ReqPerConn10KClients 1000000 23351 ns/op 3211 B/op 24 allocs/op
BenchmarkNetHTTPServerGet10ReqPerConn10KClients 1000000 13390 ns/op 2483 B/op 19 allocs/op
BenchmarkNetHTTPServerGet100ReqPerConn10KClients 1000000 13484 ns/op 2171 B/op 18 allocs/op
fasthttp server:
$ GOMAXPROCS=1 go test -bench=kServerGet -benchmem -benchtime=10s
BenchmarkServerGet1ReqPerConn 10000000 1559 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet2ReqPerConn 10000000 1248 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet10ReqPerConn 20000000 797 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet10KReqPerConn 20000000 716 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet1ReqPerConn10KClients 10000000 1974 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet2ReqPerConn10KClients 10000000 1352 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet10ReqPerConn10KClients 20000000 789 ns/op 2 B/op 0 allocs/op
BenchmarkServerGet100ReqPerConn10KClients 20000000 604 ns/op 0 B/op 0 allocs/op
GOMAXPROCS=4
net/http server:
$ GOMAXPROCS=4 go test -bench=NetHTTPServerGet -benchmem -benchtime=10s
BenchmarkNetHTTPServerGet1ReqPerConn-4 3000000 4529 ns/op 2389 B/op 29 allocs/op
BenchmarkNetHTTPServerGet2ReqPerConn-4 5000000 3896 ns/op 2418 B/op 24 allocs/op
BenchmarkNetHTTPServerGet10ReqPerConn-4 5000000 3145 ns/op 2160 B/op 19 allocs/op
BenchmarkNetHTTPServerGet10KReqPerConn-4 5000000 3054 ns/op 2065 B/op 18 allocs/op
BenchmarkNetHTTPServerGet1ReqPerConn10KClients-4 1000000 10321 ns/op 3710 B/op 30 allocs/op
BenchmarkNetHTTPServerGet2ReqPerConn10KClients-4 2000000 7556 ns/op 3296 B/op 24 allocs/op
BenchmarkNetHTTPServerGet10ReqPerConn10KClients-4 5000000 3905 ns/op 2349 B/op 19 allocs/op
BenchmarkNetHTTPServerGet100ReqPerConn10KClients-4 5000000 3435 ns/op 2130 B/op 18 allocs/op
fasthttp server:
$ GOMAXPROCS=4 go test -bench=kServerGet -benchmem -benchtime=10s
BenchmarkServerGet1ReqPerConn-4 10000000 1141 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet2ReqPerConn-4 20000000 707 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet10ReqPerConn-4 30000000 341 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet10KReqPerConn-4 50000000 310 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet1ReqPerConn10KClients-4 10000000 1119 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet2ReqPerConn10KClients-4 20000000 644 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet10ReqPerConn10KClients-4 30000000 346 ns/op 0 B/op 0 allocs/op
BenchmarkServerGet100ReqPerConn10KClients-4 50000000 282 ns/op 0 B/op 0 allocs/op
HTTP client comparison with net/http
In short, fasthttp client is up to 10 times faster than net/http. Below are benchmark results.
GOMAXPROCS=1
net/http client:
$ GOMAXPROCS=1 go test -bench='HTTPClient(Do|GetEndToEnd)' -benchmem -benchtime=10s
BenchmarkNetHTTPClientDoFastServer 1000000 12567 ns/op 2616 B/op 35 allocs/op
BenchmarkNetHTTPClientGetEndToEnd1TCP 200000 67030 ns/op 5028 B/op 56 allocs/op
BenchmarkNetHTTPClientGetEndToEnd10TCP 300000 51098 ns/op 5031 B/op 56 allocs/op
BenchmarkNetHTTPClientGetEndToEnd100TCP 300000 45096 ns/op 5026 B/op 55 allocs/op
BenchmarkNetHTTPClientGetEndToEnd1Inmemory 500000 24779 ns/op 5035 B/op 57 allocs/op
BenchmarkNetHTTPClientGetEndToEnd10Inmemory 1000000 26425 ns/op 5035 B/op 57 allocs/op
BenchmarkNetHTTPClientGetEndToEnd100Inmemory 500000 28515 ns/op 5045 B/op 57 allocs/op
BenchmarkNetHTTPClientGetEndToEnd1000Inmemory 500000 39511 ns/op 5096 B/op 56 allocs/op
fasthttp client:
$ GOMAXPROCS=1 go test -bench='kClient(Do|GetEndToEnd)' -benchmem -benchtime=10s
BenchmarkClientDoFastServer 20000000 865 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd1TCP 1000000 18711 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd10TCP 1000000 14664 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd100TCP 1000000 14043 ns/op 1 B/op 0 allocs/op
BenchmarkClientGetEndToEnd1Inmemory 5000000 3965 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd10Inmemory 3000000 4060 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd100Inmemory 5000000 3396 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd1000Inmemory 5000000 3306 ns/op 2 B/op 0 allocs/op
GOMAXPROCS=4
net/http client:
$ GOMAXPROCS=4 go test -bench='HTTPClient(Do|GetEndToEnd)' -benchmem -benchtime=10s
BenchmarkNetHTTPClientDoFastServer-4 2000000 8774 ns/op 2619 B/op 35 allocs/op
BenchmarkNetHTTPClientGetEndToEnd1TCP-4 500000 22951 ns/op 5047 B/op 56 allocs/op
BenchmarkNetHTTPClientGetEndToEnd10TCP-4 1000000 19182 ns/op 5037 B/op 55 allocs/op
BenchmarkNetHTTPClientGetEndToEnd100TCP-4 1000000 16535 ns/op 5031 B/op 55 allocs/op
BenchmarkNetHTTPClientGetEndToEnd1Inmemory-4 1000000 14495 ns/op 5038 B/op 56 allocs/op
BenchmarkNetHTTPClientGetEndToEnd10Inmemory-4 1000000 10237 ns/op 5034 B/op 56 allocs/op
BenchmarkNetHTTPClientGetEndToEnd100Inmemory-4 1000000 10125 ns/op 5045 B/op 56 allocs/op
BenchmarkNetHTTPClientGetEndToEnd1000Inmemory-4 1000000 11132 ns/op 5136 B/op 56 allocs/op
fasthttp client:
$ GOMAXPROCS=4 go test -bench='kClient(Do|GetEndToEnd)' -benchmem -benchtime=10s
BenchmarkClientDoFastServer-4 50000000 397 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd1TCP-4 2000000 7388 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd10TCP-4 2000000 6689 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd100TCP-4 3000000 4927 ns/op 1 B/op 0 allocs/op
BenchmarkClientGetEndToEnd1Inmemory-4 10000000 1604 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd10Inmemory-4 10000000 1458 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd100Inmemory-4 10000000 1329 ns/op 0 B/op 0 allocs/op
BenchmarkClientGetEndToEnd1000Inmemory-4 10000000 1316 ns/op 5 B/op 0 allocs/op
Install
go get -u github.com/valyala/fasthttp
Switching from net/http to fasthttp
Unfortunately, fasthttp doesn't provide API identical to net/http. See the FAQ for details. There is net/http -> fasthttp handler converter, but it is better to write fasthttp request handlers by hand in order to use all of the fasthttp advantages (especially high performance :) ).
Important points:
Fasthttp works with RequestHandler functions instead of objects implementing Handler interface. Fortunately, it is easy to pass bound struct methods to fasthttp:
type MyHandler struct {
foobar string
}
// request handler in net/http style, i.e. method bound to MyHandler struct.
func (h *MyHandler) HandleFastHTTP(ctx *fasthttp.RequestCtx) {
// notice that we may access MyHandler properties here - see h.foobar.
fmt.Fprintf(ctx, "Hello, world! Requested path is %q. Foobar is %q",
ctx.Path(), h.foobar)
}
// request handler in fasthttp style, i.e. just plain function.
func fastHTTPHandler(ctx *fasthttp.RequestCtx) {
fmt.Fprintf(ctx, "Hi there! RequestURI is %q", ctx.RequestURI())
}
// pass bound struct method to fasthttp
myHandler := &MyHandler{
foobar: "foobar",
}
fasthttp.ListenAndServe(":8080", myHandler.HandleFastHTTP)
// pass plain function to fasthttp
fasthttp.ListenAndServe(":8081", fastHTTPHandler)
The RequestHandler accepts only one argument - RequestCtx. It contains all the functionality required for http request processing and response writing. Below is an example of a simple request handler conversion from net/http to fasthttp.
// net/http request handler
requestHandler := func(w http.ResponseWriter, r *http.Request) {
switch r.URL.Path {
case "/foo":
fooHandler(w, r)
case "/bar":
barHandler(w, r)
default:
http.Error(w, "Unsupported path", http.StatusNotFound)
}
}
// the corresponding fasthttp request handler
requestHandler := func(ctx *fasthttp.RequestCtx) {
switch string(ctx.Path()) {
case "/foo":
fooHandler(ctx)
case "/bar":
barHandler(ctx)
default:
ctx.Error("Unsupported path", fasthttp.StatusNotFound)
}
}
Fasthttp allows setting response headers and writing response body in an arbitrary order. There is no 'headers first, then body' restriction like in net/http. The following code is valid for fasthttp:
requestHandler := func(ctx *fasthttp.RequestCtx) {
// set some headers and status code first
ctx.SetContentType("foo/bar")
ctx.SetStatusCode(fasthttp.StatusOK)
// then write the first part of body
fmt.Fprintf(ctx, "this is the first part of body\n")
// then set more headers
ctx.Response.Header.Set("Foo-Bar", "baz")
// then write more body
fmt.Fprintf(ctx, "this is the second part of body\n")
// then override already written body
ctx.SetBody([]byte("this is completely new body contents"))
// then update status code
ctx.SetStatusCode(fasthttp.StatusNotFound)
// basically, anything may be updated many times before
// returning from RequestHandler.
//
// Unlike net/http fasthttp doesn't put response to the wire until
// returning from RequestHandler.
}
Fasthttp doesn't provide ServeMux, but there are more powerful third-party routers and web frameworks with fasthttp support:
Net/http code with simple ServeMux is trivially converted to fasthttp code:
// net/http code
m := &http.ServeMux{}
m.HandleFunc("/foo", fooHandlerFunc)
m.HandleFunc("/bar", barHandlerFunc)
m.Handle("/baz", bazHandler)
http.ListenAndServe(":80", m)
// the corresponding fasthttp code
m := func(ctx *fasthttp.RequestCtx) {
switch string(ctx.Path()) {
case "/foo":
fooHandlerFunc(ctx)
case "/bar":
barHandlerFunc(ctx)
case "/baz":
bazHandler.HandlerFunc(ctx)
default:
ctx.Error("not found", fasthttp.StatusNotFound)
}
}
fasthttp.ListenAndServe(":80", m)
net/http -> fasthttp conversion table:
var (
w http.ResponseWriter
r *http.Request
ctx *fasthttp.RequestCtx
)
VERY IMPORTANT! Fasthttp disallows holding references to RequestCtx or to its' members after returning from RequestHandler. Otherwise data races are inevitable. Carefully inspect all the net/http request handlers converted to fasthttp whether they retain references to RequestCtx or to its' members after returning. RequestCtx provides the following band aids for this case:
Use this brilliant tool - race detector - for detecting and eliminating data races in your program. If you detected data race related to fasthttp in your program, then there is high probability you forgot calling TimeoutError before returning from RequestHandler.
Blind switching from net/http to fasthttp won't give you performance boost. While fasthttp is optimized for speed, its' performance may be easily saturated by slow RequestHandler. So profile and optimize your code after switching to fasthttp. For instance, use quicktemplate instead of html/template.
See also fasthttputil, fasthttpadaptor and expvarhandler.
Performance optimization tips for multi-core systems
Fasthttp best practices
Tricks with []byte buffers
The following tricks are used by fasthttp. Use them in your code too.
var (
// both buffers are uninitialized
dst []byte
src []byte
)
dst = append(dst, src...) // is legal if dst is nil and/or src is nil
copy(dst, src) // is legal if dst is nil and/or src is nil
(string(src) == "") // is true if src is nil
(len(src) == 0) // is true if src is nil
src = src[:0] // works like a charm with nil src
// this for loop doesn't panic if src is nil
for i, ch := range src {
doSomething(i, ch)
}
So throw away nil checks for []byte buffers from you code. For example,
srcLen := 0
if src != nil {
srcLen = len(src)
}
becomes
srcLen := len(src)
String may be appended to []byte buffer with append
dst = append(dst, "foobar"...)
[]byte buffer may be extended to its' capacity.
All fasthttp functions accept nil []byte buffer
statusCode, body, err := fasthttp.Get(nil, "http://google.com/")
uintBuf := fasthttp.AppendUint(nil, 1234)
Related projects
FAQ
Why creating yet another http package instead of optimizing net/http?
Because net/http API limits many optimization opportunities. For example:
Why fasthttp API is incompatible with net/http?
Because net/http API limits many optimization opportunities. See the answer above for more details. Also certain net/http API parts are suboptimal for use:
Why fasthttp doesn't support HTTP/2.0 and WebSockets?
HTTP/2.0 support is in progress. WebSockets has been done already. Third parties also may use RequestCtx.Hijack for implementing these goodies.
Are there known net/http advantages comparing to fasthttp?
Yes:
Why fasthttp API prefers returning []byte instead of string?
Because []byte to string conversion isn't free - it requires memory allocation and copy. Feel free wrapping returned []byte result into string() if you prefer working with strings instead of byte slices. But be aware that this has non-zero overhead.
Which GO versions are supported by fasthttp?
Go1.5+. Older versions won't be supported, since their standard package miss useful functions.
NOTE: Go 1.9.7 is the oldest tested version. We recommend you to update as soon as you can. As of 1.11.3 we will drop 1.9.x support.
Please provide real benchmark data and server information
See this issue.
Are there plans to add request routing to fasthttp?
There are no plans to add request routing into fasthttp. Use third-party routers and web frameworks with fasthttp support:
I detected data race in fasthttp!
Cool! File a bug. But before doing this check the following in your code:
I didn't find an answer for my question here
Try exploring these questions.
Download Details:
Author: valyala
Download Link: Download The Source Code
Official Website: https://github.com/valyala/fasthttp
License: MIT
#go #golang #programming #developer
1652177760
go-vcr
simplifies testing by recording your HTTP interactions and replaying them in future runs in order to provide fast, deterministic and accurate testing of your code.
go-vcr
was inspired by the VCR library for Ruby.
Install go-vcr
by executing the command below:
$ go get github.com/dnaeon/go-vcr/v2/recorder
Here is a simple example of recording and replaying etcd HTTP interactions.
You can find other examples in the example
directory of this repository as well.
package main
import (
"log"
"time"
"github.com/dnaeon/go-vcr/v2/recorder"
"github.com/coreos/etcd/client"
"golang.org/x/net/context"
)
func main() {
// Start our recorder
r, err := recorder.New("fixtures/etcd")
if err != nil {
log.Fatal(err)
}
defer r.Stop() // Make sure recorder is stopped once done with it
// Create an etcd configuration using our transport
cfg := client.Config{
Endpoints: []string{"http://127.0.0.1:2379"},
HeaderTimeoutPerRequest: time.Second,
Transport: r, // Inject as transport!
}
// Create an etcd client using the above configuration
c, err := client.New(cfg)
if err != nil {
log.Fatalf("Failed to create etcd client: %s", err)
}
// Get an example key from etcd
etcdKey := "/foo"
kapi := client.NewKeysAPI(c)
resp, err := kapi.Get(context.Background(), etcdKey, nil)
if err != nil {
log.Fatalf("Failed to get etcd key %s: %s", etcdKey, err)
}
log.Printf("Successfully retrieved etcd key %s: %s", etcdKey, resp.Node.Value)
}
During replay mode, You can customize the way incoming requests are matched against the recorded request/response pairs by defining a Matcher function. For example, the following matcher will match on method, URL and body:
r, err := recorder.New("fixtures/matchers")
if err != nil {
log.Fatal(err)
}
defer r.Stop() // Make sure recorder is stopped once done with it
r.SetMatcher(func(r *http.Request, i cassette.Request) bool {
if r.Body == nil {
return cassette.DefaultMatcher(r, i)
}
var b bytes.Buffer
if _, err := b.ReadFrom(r.Body); err != nil {
return false
}
r.Body = ioutil.NopCloser(&b)
return cassette.DefaultMatcher(r, i) && (b.String() == "" || b.String() == i.Body)
})
You often provide sensitive data, such as API credentials, when making requests against a service. By default, this data will be stored in the recorded data but you probably don't want this. Removing or replacing data before it is stored can be done by adding one or more Filter
s to your Recorder
. Here is an example that removes the Authorization
header from all requests:
r, err := recorder.New("fixtures/filters")
if err != nil {
log.Fatal(err)
}
defer r.Stop() // Make sure recorder is stopped once done with it
// Add a filter which removes Authorization headers from all requests:
r.AddFilter(func(i *cassette.Interaction) error {
delete(i.Request.Headers, "Authorization")
return nil
})
Filters added using *Recorder.AddFilter
are applied within VCR's custom http.Transport
. This means that if you edit a response in such a filter then subsequent test code will see the edited response. This may not be desirable in all cases. For instance, if a response body contains an OAuth access token that is needed for subsequent requests, then redact the access token in SaveFilter
will result in authorization failures.
Another way to edit recorded interactions is to use *Recorder.AddSaveFilter
. Filters added with this method are applied just before interactions are saved when *Recorder.Stop
is called.
r, err := recorder.New("fixtures/filters")
if err != nil {
log.Fatal(err)
}
defer r.Stop() // Make sure recorder is stopped once done with it
// Your test code will continue to see the real access token and
// it is redacted before the recorded interactions are saved
r.AddSaveFilter(func(i *cassette.Interaction) error {
if strings.Contains(i.URL, "/oauth/token") {
i.Response.Body = `{"access_token": "[REDACTED]"}`
}
return nil
})
Sometimes you want to allow specific requests to pass through to the remote server without recording anything.
Globally, you can use ModeDisabled
for this, but if you want to disable the recorder for individual requests, you can add Passthrough
functions to the recorder. The function takes a pointer to the original request, and returns a boolean, indicating if the request should pass through to the remote server.
Here's an example to pass through requests to a specific endpoint:
// Pass through the request to the remote server if the path matches "/login".
r.AddPassthrough(func(req *http.Request) bool {
return req.URL.Path == "/login"
})
Author: Dnaeon
Source Code: https://github.com/dnaeon/go-vcr
License: View license
1651097100
Request
HTTP Client for golang, Inspired by Javascript-axios Python-request. If you have experience about axios or requests, you will love it. No 3rd dependency.
go mod:
go get github.com/monaco-io/request
package main
import (
"github.com/monaco-io/request"
)
func main() {
var body = struct {
A string
B int
}{A: "A", B: 1}
var result interface{}
c := request.Client{
URL: "https://google.com",
Method: "POST",
Query: map[string]string{"hello": "world"},
JSON: body,
}
resp := c.Send().Scan(&result)
if !resp.OK(){
// handle error
log.Println(resp.Error())
}
// str := resp.String()
// bytes := resp.Bytes()
package main
import (
"github.com/monaco-io/request"
)
func main() {
c := request.Client{
URL: "https://google.com",
Method: "POST",
Query: map[string]string{"hello": "world"},
MultipartForm: MultipartForm{
Fields: map[string]string{"a": "1"},
Files: []string{"doc.txt"},
},
}
resp := c.Send().Scan(&result)
...
package main
import (
"github.com/monaco-io/request"
)
func main() {
var response interface{}
resp := request.
New().
POST("http://httpbin.org/post").
AddHeader(map[string]string{"Google": "google"}).
AddBasicAuth("google", "google").
AddURLEncodedForm(map[string]string{"data": "google"}).
Send().
Scan(&response)
...
package main
import (
"github.com/monaco-io/request"
"context"
)
func main() {
c := request.Client{
Context: context.Background(),
URL: "https://google.com",
Method: "POST",
BasicAuth: request.BasicAuth{
Username: "google",
Password: "google",
},
}
resp := c.Send()
...
package main
import (
"github.com/monaco-io/request"
"context"
)
func main() {
var response interface{}
resp := request.
NewWithContext(context.TODO()).
POST("http://httpbin.org/post").
AddHeader(map[string]string{"Google": "google"}).
AddBasicAuth("google", "google").
AddURLEncodedForm(map[string]string{"data": "google"}).
Send().
Scan(&response)
...
package main
import (
"github.com/monaco-io/request"
)
func main() {
c := request.Client{
URL: "https://google.com",
Method: "POST",
BasicAuth: request.BasicAuth{
Username: "google",
Password: "google",
},
}
resp := c.Send()
}
package main
import (
"github.com/monaco-io/request"
)
func main() {
c := request.Client{
URL: "https://google.com",
Method: "POST",
Timeout: time.Second*10,
}
}
package main
import (
"github.com/monaco-io/request"
)
func main() {
c := request.Client{
URL: "https://google.com",
CookiesMap: map[string]string{
"cookie_name": "cookie_value",
}
}
}
package main
import (
"github.com/monaco-io/request"
)
func main() {
c := request.Client{
URL: "https://google.com",
TLSConfig: &tls.Config{InsecureSkipVerify: true},
}
}
Author: Monaco-io
Source Code: https://github.com/monaco-io/request
License: MIT License
1651048380
go-retryablehttp
The retryablehttp
package provides a familiar HTTP client interface with automatic retries and exponential backoff. It is a thin wrapper over the standard net/http
client library and exposes nearly the same public API. This makes retryablehttp
very easy to drop into existing programs.
retryablehttp
performs automatic retries under certain conditions. Mainly, if an error is returned by the client (connection errors, etc.), or if a 500-range response code is received (except 501), then a retry is invoked after a wait period. Otherwise, the response is returned and left to the caller to interpret.
The main difference from net/http
is that requests which take a request body (POST/PUT et. al) can have the body provided in a number of ways (some more or less efficient) that allow "rewinding" the request body if the initial request fails so that the full request can be attempted again. See the godoc for more details.
Version 0.6.0 and before are compatible with Go prior to 1.12. From 0.6.1 onward, Go 1.12+ is required. From 0.6.7 onward, Go 1.13+ is required.
Example Use
Using this library should look almost identical to what you would do with net/http
. The most simple example of a GET request is shown below:
resp, err := retryablehttp.Get("/foo")
if err != nil {
panic(err)
}
The returned response object is an *http.Response
, the same thing you would usually get from net/http
. Had the request failed one or more times, the above call would block and retry with exponential backoff.
It's possible for a request to succeed in the sense that the expected response headers are received, but then to encounter network-level errors while reading the response body. In go-retryablehttp's most basic usage, this error would not be retryable, due to the out-of-band handling of the response body. In some cases it may be desirable to handle the response body as part of the retryable operation.
A toy example (which will retry the full request and succeed on the second attempt) is shown below:
c := retryablehttp.NewClient()
r := retryablehttp.NewRequest("GET", "://foo", nil)
handlerShouldRetry := true
r.SetResponseHandler(func(*http.Response) error {
if !handlerShouldRetry {
return nil
}
handlerShouldRetry = false
return errors.New("retryable error")
})
*http.Client
with retriesIt's possible to convert a *retryablehttp.Client
directly to a *http.Client
. This makes use of retryablehttp broadly applicable with minimal effort. Simply configure a *retryablehttp.Client
as you wish, and then call StandardClient()
:
retryClient := retryablehttp.NewClient()
retryClient.RetryMax = 10
standardClient := retryClient.StandardClient() // *http.Client
For more usage and examples see the godoc.
Author: Hashicorp
Source Code: https://github.com/hashicorp/go-retryablehttp
License: MPL-2.0 License