房价信息爬取

  这次准备试试goquery库，来爬取我房网的房价信息。首先要安装goquery,参考当go get遇到墙时，安装好库就可以开始。
  比较重要的就是观察页面布局，元素特征。
  本次爬虫只要有房价的楼盘信息，代码如下：

package main

import (
    "fmt"
    "strconv"
    "time"
    "github.com/PuerkitoBio/goquery"
    "log"
    "bytes"
    "encoding/csv"
    "os"
)
func p() {
    a:=0
    fileName := "wofang.csv"
    buf := new(bytes.Buffer)
    r2 := csv.NewWriter(buf)
    for i := 1; i <=202; i++ {
        fmt.Println("正在抓取第" + strconv.Itoa(i) + "页......")
        url := "http://www.wofang.com/building/p/" + strconv.Itoa(i) + "/"
        if i==1{
            url= "http://www.wofang.com/building/"
        }
        doc, err := goquery.NewDocument(url)
        if err != nil {
            log.Fatal(err)
        }
        doc.Find(".m ul li").Each(func(i int, s *goquery.Selection) {
                name:= s.Find(".title a").Text()
                location:= s.Find(".time").Text()
                price:=s.Find(".sale-price font").Text()
                if price!="" {
                    a++
                    s := make([]string,3)
                    s[0] = name
                    s[1] = price
                    s[2] = location
                    r2.Write(s)
                    r2.Flush()
                    fmt.Printf("%s,%s,%s\n", name,price, location)
                }
        })
    }
    fout,err := os.Create(fileName)
    defer fout.Close()
    if err != nil {
        fmt.Println(fileName,err)
        return
    }
    fout.WriteString(buf.String())
    fmt.Print(a)
}
func main() {
    t1 := time.Now() // get current time
    p()
    elapsed := time.Since(t1)
    fmt.Println("")
    fmt.Println("爬虫结束,总共耗时: ", elapsed)
}

  最后根据链接规律(大致就是：http://www.wofang.com/building/" + 地市键值+ "-te_住宅/"，键值用的比较笨的方法手动一个个点出来的)爬一下各地市的信息并用echart做可视化。(本文仅供参考)

转载于:https://www.cnblogs.com/polong/p/9831261.html