Golang Page.AddTargetRequest方法代码示例

// Parse html dom here and record the parse result that we want to crawl.
// Package simplejson (https://github.com/bitly/go-simplejson) is used to parse data of json.
func (this *MyPageProcesser) Process(p *page.Page) {
	if !p.IsSucc() {
		println(p.Errormsg())
		return
	}

	query := p.GetJson()
	status, err := query.GetPath("result", "status", "code").Int()
	if status != 0 || err != nil {
		log.Panicf("page is crawled error : errorinfo=%s : status=%d : startNewsId=%d", err.Error(), status, this.startNewsId)
	}
	num, err := query.GetPath("result", "pageStr", "pageSize").Int()
	if num == 0 || err != nil {
		// Add url of next crawl
		startIdstr := strconv.Itoa(this.startNewsId)
		p.AddTargetRequest("http://live.sina.com.cn/zt/api/l/get/finance/globalnews1/index.htm?format=json&id="+startIdstr+"&pagesize=10&dire=f", "json")
		return
	}

	var idint, nextid int
	var nextidstr string
	query = query.Get("result").Get("data")
	for i := 0; i < num; i++ {
		id, err := query.GetIndex(i).Get("id").String()
		if id == "" || err != nil {
			continue
		}
		idint, err = strconv.Atoi(id)
		if err != nil {
			continue
		}
		if idint <= this.startNewsId {
			break
		}
		if i == 0 {
			nextid = idint
			nextidstr = id
		}
		content, err := query.GetIndex(i).Get("content").String()
		if content == "" || err != nil {
			continue
		}
		time, err := query.GetIndex(i).Get("created_at").String()
		if err != nil {
			continue
		}

		p.AddField(id+"_id", id)
		p.AddField(id+"_content", content)
		p.AddField(id+"_time", time)
	}
	// Add url of next crawl
	this.startNewsId = nextid
	p.AddTargetRequest("http://live.sina.com.cn/zt/api/l/get/finance/globalnews1/index.htm?format=json&id="+nextidstr+"&pagesize=10&dire=f", "json")
	//println(p.GetTargetRequests())

}

// Parse html dom here and record the parse result that we want to Page.
// Package goquery (http://godoc.org/github.com/PuerkitoBio/goquery) is used to parse html.
func (this *MyPageProcesser) Process(p *page.Page) {
	if !p.IsSucc() {
		println(p.Errormsg())
		return
	}
	var crawok bool
	crawok = false
	//query := p.GetHtmlParser()
	//var urls []string

	//fmt.Println(p.GetBodyStr())
	re := regexp.MustCompile(`<a href="(.*?)">(.*?)`)

	sectUrlsTemp := re.FindAllSubmatch([]byte(p.GetBodyStr()), -1)

	for _, url := range sectUrlsTemp {
		for _, url1 := range url {
			crawok = true

			http_index := strings.Index(string(url1), "http://shinichr.diandian.com")

			http_note := strings.Index(string(url1), "\"")
			http_quote := strings.Index(string(url1), "#")

			if http_index >= 0 && http_quote < 0 {
				if http_note > 0 && http_note < http_index {
					continue
				}

				var http_url string
				if http_note <= 0 {
					http_url = string(url1)[http_index:]
				} else {
					http_url = string(url1)[http_index:http_note]
				}

				if this.visit_url[http_url] == 0 {
					this.visit_url[http_url] = 1

					fmt.Println("####unvisited:", http_url)
					//fmt.Println("###AddTargetRequest:", http_url)
					p.AddTargetRequest(http_url, "html")
				}
			}
		}
	}
	if crawok == false {
		fmt.Println("crawl false:*****************", p.GetRequest().GetUrl())
		http_page := strings.Index(p.GetRequest().GetUrl(), "http://shinichr.diandian.com/page")
		http_post := strings.Index(p.GetRequest().GetUrl(), "http://shinichr.diandian.com/post")
		fmt.Println("http_page:", http_page, "http_post:", http_post)
		if http_page >= 0 || http_post >= 0 {
			//this.visit_url[p.GetRequest().GetUrl()] = 0
			p.AddTargetRequest(p.GetRequest().GetUrl(), "html")
		}
	}
	http_index := strings.Index(p.GetRequest().GetUrl(), "http://shinichr.diandian.com/post/")

	//	rex, _ := regexp.Compile("\\/")
	//replaceurl := rex.ReplaceAllString(p.GetRequest().GetUrl(), ".")
	//fmt.Println("http_index=", http_index)
	//fmt.Println("replaceurl=", p.GetRequest().GetUrl()[http_index:], "....", http_index)
	if http_index >= 0 {

		cuturl := p.GetRequest().GetUrl()[34:]
		//fmt.Println("replaceurl=", cuturl)
		rex, _ := regexp.Compile("\\/")
		replaceurl := rex.ReplaceAllString(cuturl, ".")

		filedir := fmt.Sprintf("/home/shinichr/diandian_post/%s", replaceurl)

		fout, err := os.Create(filedir)
		if err != nil {
			fmt.Println(filedir, err)
			return
		}
		defer fout.Close()

		src := p.GetBodyStr()
		re, _ := regexp.Compile("\\<[\\S\\s]+?\\>")
		src = re.ReplaceAllStringFunc(src, strings.ToLower)

		//去除STYLE
		re, _ = regexp.Compile("\\<style[\\S\\s]+?\\</style\\>")
		src = re.ReplaceAllString(src, "")

		//去除SCRIPT
		re, _ = regexp.Compile("\\<script[\\S\\s]+?\\</script\\>")
		src = re.ReplaceAllString(src, "")

		//去除所有尖括号内的HTML代码，并换成换行符
		re, _ = regexp.Compile("\\<[\\S\\s]+?\\>")
		src = re.ReplaceAllString(src, "\n")

		//去除连续的换行符
		re, _ = regexp.Compile("\\s{2,}")
		src = re.ReplaceAllString(src, "\n")

//.........这里部分代码省略.........