房价信息爬取

  这次准备试试goquery库，来爬取的房价信息。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：" + 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)