中国用户因为墙的问题下载不到依赖库， 可以将GOPATH文件夹下的库移到您环境的GOPATH下
亲爱的，请用支付宝APP先扫红包码，然后如果你想打赏的话可以用红包给我打赏
If you go get difficult, you can move those files under GOPATH in this project to your Golang env's GOPATH.
HTTP Download Helper, Supports Many Features such as Cookie Persistence, HTTP(S) and SOCKS5 Proxy....
World-Wide-Web robot, also known as spiders and crawlers. The principle is to falsify network data by constructing appointed HTTP protocol data packet, then request resource to the specified host, goal is to access the data returned.
There are a large number of web information, human's hand movement such as copy-paste data from web page is time-consuming and laborious, thus inspired the data acquisition industry.
Batch access to public network data does not break the law, but because there is no difference, no control, very violent means will lead to other services is not stable, therefore, most of the resources provider will filtering some data packets(falsify), in this context, batch small data acquisition has become a problem. Integrated with various requirements, such as various API development, automated software testing(all this have similar technical principle). So this project come into the world(very simple).
The Marmot is very easy to understand, just like Python's library requests(Not yet Smile~ --| ). By enhancing native Golang HTTP library, help you deal with some trivial logic (such as collecting information, checking parameters), and add some fault-tolerant mechanisms (such as add lock, close time flow, ensure the high concurrent run without accident).
It provides a human friendly API interface, you can reuse it often. Very convenient to support Cookie Persistence, Crawler Proxy Settings, as well as others general settings, such as HTTP request header settings, timeout/pause settings, data upload/post settings.
It support all of the HTTP methods POST/PUT/GET/DELETE/... and has built-in spider pool and browser UA pool, easy to develop UA+Cookie persistence distributed spider.
The library is simple and practical, just a few lines of code to replace the previous Spaghetti code, has been applied in some large projects such as Full Golang Automatic Amazon Distributed crawler|spider, has withstood the test of two thousand long acting proxy IP and high concurrency, single machine every day to get millions of data.
The main uses: WeChat development/ API docking / Automated test / Rush Ticket Scripting / Vote Plug-in / Data Crawling
Now We support Default Worker, You can easy use:
lesson1.go
package main
import (
"fmt"
"github.com/hunterhug/marmot/miner"
)
func main() {
// Use Default Worker, You can Also New One:
// worker:=miner.New(nil)
miner.SetLogLevel(miner.DEBUG)
_, err := miner.SetUrl("https://www.whitehouse.gov").Go()
if err != nil {
fmt.Println(err.Error())
} else {
fmt.Println(miner.ToString())
}
}You can get it by:
go get -v github.com/hunterhug/marmot/miner
Or make your GOPATH sub dir: /src/github.com/hunterhug, and
cd src/github.com/hunterhug
git clone https://github.com/hunterhug/marmot
Suggest Golang1.8+
The most simple example such below, more see example/lesson dir:
lesson2.go
package main
// Example
import (
"fmt"
"github.com/hunterhug/marmot/miner"
)
func main() {
// 1. New a worker
worker, _ := miner.New(nil)
// 2. Set a URL And Fetch
html, err := worker.SetUrl("https://www.whitehouse.gov").SetUa(miner.RandomUa()).SetMethod(miner.GET).Go()
if err != nil {
fmt.Println(err.Error())
}
// 4.Print content equal to fmt.Println(worker.ToString())
fmt.Println(string(html))
}More detail Example is:
lesson3.go
package main
import (
// 1:import package
"github.com/hunterhug/marmot/miner"
"github.com/hunterhug/parrot/util"
)
func init() {
// 2:Optional global setting
miner.SetLogLevel(miner.DEBUG) // optional, set log to debug
miner.SetGlobalTimeout(3) // optional, http request timeout time
}
func main() {
log := miner.Log() // optional, miner log you can choose to use
// 3: Must new a Worker object, three ways
//worker, err := miner.NewWorker("http://smart:[email protected]:808") // proxy format: protocol://user(optional):password(optional)@ip:port
//worker, err := miner.NewWorker(nil) // normal worker, default keep Cookie
//worker, err := miner.NewAPI() // API worker, not keep Cookie
worker, err := miner.New(nil) // NewWorker alias
if err != nil {
panic(err)
}
// 4: Set the request Method/URL and some others, can chain set, only SetUrl is required.
// SetUrl: required, the Url
// SetMethod: optional, HTTP method: POST/GET/..., default GET
// SetWaitTime: optional, HTTP request wait/pause time
worker.SetUrl("http://cjhug.me/fuck.html").SetMethod(miner.GET).SetWaitTime(2)
worker.SetUa(miner.RandomUa()) // optional, browser user agent: IE/Firefox...
worker.SetRefer("https://www.whitehouse.gov") // optional, url refer
worker.SetHeaderParm("diyheader", "lenggirl") // optional, some other diy http header
//worker.SetBData([]byte("file data")) // optional, if you want post JSON data or upload file
//worker.SetFormParm("username","jinhan") // optional: if you want post form
//worker.SetFormParm("password","123")
// 5: Start Run
//worker.Get() // default GET
//worker.Post() // POST form request data, data can fill by SetFormParm()
//worker.PostJSON() // POST JSON dara, use SetBData()
//worker.PostXML() // POST XML, use SetBData()
//worker.PostFILE() // POST to Upload File, data in SetBData() too
//worker.OtherGo("OPTIONS", "application/x-www-form-urlencoded") // Other http method, Such as OPTIONS etcd
body, err := worker.Go() // if you use SetMethod(), otherwise equal to Get()
if err != nil {
log.Error(err.Error())
} else {
log.Infof("%s", string(body)) // Print return data
}
log.Debugf("%#v", worker.GetCookies) // if you not set log as debug, it will not appear
// You must Clear it! If you want to POST Data by SetFormParm()/SetBData() again
// After get the return data by post data, you can clear the data you fill
worker.Clear()
//worker.ClearAll() // you can also want to clear all, include http header you set
// Worker pool for concurrent, every Worker Object is serial as the browser. if you want collateral execution, use this.
miner.Pool.Set("myfirstworker", worker)
if w, ok := miner.Pool.Get("myfirstworker"); ok {
go func() {
data, _ := w.SetUrl("http://cjhug.me/fuck.html").Get()
log.Info(string(data))
}()
util.Sleep(10)
}
}Last example
lesson4.go
package main
import (
"context"
"fmt"
"strings"
"github.com/hunterhug/marmot/expert"
"github.com/hunterhug/marmot/miner"
)
func main() {
// We can debug, to see whether SetBeforeAction make sense
miner.SetLogLevel(miner.DEBUG)
// The url we want
url := "https://www.whitehouse.gov"
// IAM we can NewAPI
worker := miner.NewAPI()
// We can aop by context
// ctx, cancle := context.WithCancel(context.Background())
// ctx := context.TODO()
// worker.SetContext(ctx)
// Before we make some change, And every GET Or POST it will action
worker.SetBeforeAction(func(ctx context.Context, this *miner.Worker) {
fmt.Println("Before Action, I will add a HTTP header")
this.SetHeaderParm("Marmot", "v2")
this.SetHeaderParm("DUDUDUU", "DUDU")
// select {
// case <-ctx.Done():
// fmt.Println(ctx.Err()) // block in here util cancle()
// os.Exit(1)
// }
})
// we cancle it after 5 secord
// go func() {
// util.Sleep(5)
// cancle()
// }()
worker.SetAfterAction(func(ctx context.Context, this *miner.Worker) {
fmt.Println("After Action, I just print this sentence")
})
// Let's Go
body, err := worker.SetUrl(url).GoByMethod(miner.GET)
if err != nil {
fmt.Println(err.Error())
} else {
// Parse We want
fmt.Printf("Output:\n %s\n", parse(body))
}
// for {
// in here we loop util cancle() success
// }
}
// Parse HTML page
func parse(data []byte) string {
doc, err := expert.QueryBytes(data)
if err != nil {
fmt.Println(err.Error())
}
return strings.TrimSpace(doc.Find("#hero-caption").Text())
// return doc.Find("title").Text()
}Easy to use, you just need to New one Worker, and SetUrl, then make some settings and worker.Go().
There are four kinds of worker:
worker, err := miner.NewWorker("http://smart:[email protected]:808")// proxy worker, format:protocol://user(optional):password(optional)@ip:port, alias toNew(), support http(s), socks5worker, err := miner.NewWorker(nil)// normal worker, default keep Cookie, alias toNew()worker := miner.NewAPI()// API worker, will not keep Cookieworker, err := miner.NewWorkerByClient(&http.Client{})// You can also pass ahttp.Clientif you want
Camouflage our worker:
worker.SetUrl("https://www.whitehouse.gov")// required: set url you want toworker.SetMethod(miner.GET)// optional: set http methodPOST/GET/PUT/POSTJSONand so onworker.SetWaitTime(2)// optional: set timeout of http requestworker.SetUa(miner.RandomUa())// optional: set http browser user agent, you can see miner/config/ua.txtworker.SetRefer("https://www.whitehouse.gov")// optional: set http request Referworker.SetHeaderParm("diyheader", "lenggirl")// optional: set http diy headerworker.SetBData([]byte("file data"))// optional: set binary data for post or putworker.SetFormParm("username","jinhan")// optional: set form data for post or putworker.SetCookie("xx=dddd")// optional: you can set a init cookie, some website you can login and F12 copy the cookieworker.SetCookieByFile("/root/cookie.txt")// optional: set cookie which store in a file
Run our worker:
body, err := worker.Go()// if you use SetMethod(), auto use following ways, otherwise use Get()body, err := worker.Get()// defaultbody, err := worker.Post()// post form request, data fill by SetFormParm()body, err := worker.PostJSON()// post JSON request, data fill by SetBData()body, err := worker.PostXML()// post XML request, data fill by SetBData()body, err := worker.PostFILE()// upload file, data fill by SetBData(), and should set SetFileInfo(fileName, fileFormName string)body, err := worker.Delete()// you know!body, err := worker.Put()// ones http method...body, err := worker.PutJSON()// put JSON requestbody, err := worker.PutXML()body, err := worker.PutFILE()body, err := worker.OtherGo("OPTIONS", "application/x-www-form-urlencoded")// Other http method, Such as OPTIONS etc., can not sent binary.body, err := worker.OtherGoBinary("OPTIONS", "application/x-www-form-urlencoded")// Other http method, Such as OPTIONS etc., just sent binary.body, err := worker.GoByMethod("POST")// you can override SetMethod() By this, equal SetMethod() then Go()
Deal the return data, all data will be return as binary, You can immediately store it into a new variable:
fmt.Println(string(html))// type change directlyfmt.Println(worker.ToString())// use spider method, after http response, data will keep in the fieldRaw, just use ToStringfmt.Println(worker.JsonToString())// some json data will include chinese and other multibyte character, such as我爱你,我的小绵羊,사랑해
Attention: after every request for a url, the next request you can cover your http request header, otherwise header you set still exist,
if just want clear post data, use Clear(), and want clear HTTP header too please use ClearAll() .
Here is some practice in the example dir.
More see the code source.
Hook:
SetBeforeAction(fc func(context.Context, *Worker))SetAfterAction(fc func(context.Context, *Worker))
It has already used in many project(although some is very simple) :
- Full Golang Automatic Amazon Distributed crawler|spider // Just see Picture
- GoTaobao
- A lot closed source...
Project change you can see log, Install development environment you can refer:GoSpider-docker
Copyright 2017 by marmot author: [email protected].
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License