Ejemplo n.º 1
0
func (this *Spider) pageProcess(req *page.Request) {
	var p *page.Page
	//下载页面
	for i := 0; i < 3; i++ {
		p = this.m_downLoader.DownLoad(req)
		if p.IsSucc() {
			break
		}
		time.Sleep(time.Microsecond * 1000)
	}

	if !p.IsSucc() {
		this.finishForReqProcesser(req.GetUrl())
		return
	}

	//分析页面内容
	this.m_pageProcesser.Process(p)

	//获取新的链接
	if p.CountNewUrls() > 0 {
		newUrls := p.GetNewUrls()
		for tmpUrl, tmpUrlTag := range newUrls {
			this.AddUrl(tmpUrl, "html", tmpUrlTag)
		}
	}

	this.finishForReqProcesser(req.GetUrl())

	//输出
	for _, tmpOut := range this.m_outputs {
		tmpOut.Process(p.GetPageItemsList(), p.GetRequest().GetUrl())
	}

}
Ejemplo n.º 2
0
func (this *HttpDownLoader) downloadHtml(p *page.Page, req *page.Request) *page.Page {
	p, destBody := this.downloadFile(p, req)
	if !p.IsSucc() {
		return p
	}
	p.SetBody(destBody)
	return p
}
Ejemplo n.º 3
0
func (this *YouxiduoProcesser) Process(p *page.Page) {
	if !p.IsSucc() {
		println(p.GetErrMsg())
		return
	}
	var body string = p.GetBody()
	var urlTag string = p.GetRequest().GetUrlTag()
	p.SetUrlTag(urlTag)
	//分析这个页面是LIST页面还是内容页面
	// <div class="infroList"><ul><li>...</div>===>LIST
	// <div class="pagebreak">...</div>===>LIST
	// CONTENT
	//<div class="article"

	if urlTag == "list" {
		//
		//1.寻找news-brief的content
		regList, err := regexp.Compile(`<div class=\"infroList\">(\s|.)*<\/ul>(\s|.)*<div class=\"pagebreak\">`)
		if err != nil {
			logs.GetFirstLogger().Error("分析页面出错,正则表达式错误了,url = " + p.GetRequest().GetUrl())
		}
		var infroList []string = regList.FindAllString(body, -1)

		if len(infroList) > 0 {
			this.parseNewsBreifInfo(infroList[0], p)
		} else {
			logs.GetFirstLogger().Info("No more list items")
		}
		//先寻找额外的LIST页面
		if !p.IsBreak() {
			regPageBreak, err := regexp.Compile(`<div class=\"pagebreak\">(\s|.)+<li class=\"lastPage\">`)
			if err != nil {
				logs.GetFirstLogger().Error("分析页面出错,翻页正则表达式错误,url = " + p.GetRequest().GetUrl())
			}
			var pageBreakList []string = regPageBreak.FindAllString(body, -1)
			if len(pageBreakList) > 0 {
				this.parseNewsLinkListInfo(pageBreakList[0], p)
			} else {
				logs.GetFirstLogger().Info("No more links")
			}

		}

	} else {
		//CONTENT
		this.parseNewsDetail(body, p)
	}

}