Esempio n. 1
0
// 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
	}

	query := p.GetHtmlParser()

	query.Find(`div[class="wx-rb bg-blue wx-rb_v1 _item"]`).Each(func(i int, s *goquery.Selection) {
		name := s.Find("div.txt-box > h3").Text()
		href, _ := s.Attr("href")

		fmt.Printf("WeName:%v link:http://http://weixin.sogou.com%v \r\n", name, href)
		// the entity we want to save by Pipeline
		p.AddField("name", name)
		p.AddField("href", href)
	})

	next_page_href, _ := query.Find("#sogou_next").Attr("href")
	if next_page_href == "" {
		p.SetSkip(true)
	} else {
		p.AddTargetRequestWithHeaderFile("http://weixin.sogou.com/weixin"+next_page_href, "html", "weixin.sogou.com.json")
	}

}
Esempio n. 2
0
// 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
	}

	query := p.GetHtmlParser()
	var urls []string
	query.Find("h3[class='repo-list-name'] a").Each(func(i int, s *goquery.Selection) {
		href, _ := s.Attr("href")
		urls = append(urls, "http://github.com/"+href)
	})
	// these urls will be saved and crawed by other coroutines.
	p.AddTargetRequests(urls, "html")

	name := query.Find(".entry-title .author").Text()
	name = strings.Trim(name, " \t\n")
	repository := query.Find(".entry-title .js-current-repository").Text()
	repository = strings.Trim(repository, " \t\n")
	//readme, _ := query.Find("#readme").Html()
	if name == "" {
		p.SetSkip(true)
	}
	// the entity we want to save by Pipeline
	p.AddField("author", name)
	p.AddField("project", repository)
	//p.AddField("readme", readme)
}
Esempio n. 3
0
// 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())

}
func (this *MyProcessor) Process(p *page.Page) {
	if !p.IsSucc() {
		mlog.LogInst().LogError(p.Errormsg())
		return
	}

	u, err := url.Parse(p.GetRequest().GetUrl())
	if err != nil {
		mlog.LogInst().LogError(err.Error())
		return
	}
	if !strings.HasSuffix(u.Host, "jiexieyin.org") {
		return
	}

	var urls []string
	query := p.GetHtmlParser()

	query.Find("a").Each(func(i int, s *goquery.Selection) {
		href, _ := s.Attr("href")
		reJavascript := regexp.MustCompile("^javascript\\:")
		reLocal := regexp.MustCompile("^\\#")
		reMailto := regexp.MustCompile("^mailto\\:")
		if reJavascript.MatchString(href) || reLocal.MatchString(href) || reMailto.MatchString(href) {
			return
		}

		//处理相对路径
		var absHref string
		urlHref, err := url.Parse(href)
		if err != nil {
			mlog.LogInst().LogError(err.Error())
			return
		}
		if !urlHref.IsAbs() {
			urlPrefix := p.GetRequest().GetUrl()
			absHref = urlPrefix + href
			urls = append(urls, absHref)
		} else {
			urls = append(urls, href)
		}

	})

	p.AddTargetRequests(urls, "html")

}
Esempio n. 5
0
// Parse html dom here and record the parse result that we want to crawl.
// 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
	}

	query := p.GetHtmlParser()

	name := query.Find(".lemmaTitleH1").Text()
	name = strings.Trim(name, " \t\n")

	summary := query.Find(".card-summary-content .para").Text()
	summary = strings.Trim(summary, " \t\n")

	// the entity we want to save by Pipeline
	p.AddField("name", name)
	p.AddField("summary", summary)
}
Esempio n. 6
0
// Parse html dom here and record the parse result that we want to crawl.
// 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
	}

	html := p.GetBodyStr()
	newUrls := urlutil.GetAllUrlIn(p.GetRequest().GetUrl(), html)
	for _, newUrl := range newUrls {
		newUrl = strings.Replace(newUrl, "//weibo.com/", "//tw.weibo.com/", -1)
		p.AddTargetRequest(newUrl, "html")
	}

	mailAddrList := mailaddrutil.GetAllMailAddrIn(html)
	for _, mailAddr := range mailAddrList {
		if _, ok := this.mailAddrMap[mailAddr]; !ok {
			this.mailAddrMap[mailAddr] = true
			this.mailLogger.WriteString(mailAddr + "\n")
			this.MailHandle.Push(mailAddr)
		}
	}
}
Esempio n. 7
0
// 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.GetUrlTag() == "site_login" {
		//fmt.Printf("%v\n", p.GetCookies())
		this.cookies = p.GetCookies()
		// AddTargetRequestWithParams Params:
		//  1. Url.
		//  2. Responce type is "html" or "json" or "jsonp" or "text".
		//  3. The urltag is name for marking url and distinguish different urls in PageProcesser and Pipeline.
		//  4. The method is POST or GET.
		//  5. The postdata is body string sent to sever.
		//  6. The header is header for http request.
		//  7. Cookies
		//  8. Http redirect function
		if len(this.cookies) != 0 {
			p.AddField("info", "get cookies success")
			req := request.NewRequest("http://backadmin.hucong.net/site/index", "html", "site_index", "GET", "", nil, this.cookies, nil, nil)
			p.AddTargetRequestWithParams(req)
		} else {
			p.AddField("info", "get cookies failed")
		}
	} else {
		//fmt.Printf("%v\n", p.GetBodyStr())
		query := p.GetHtmlParser()
		pageTitle := query.Find(".page-content .page-title").Text()

		if len(pageTitle) != 0 {
			p.AddField("page_title", pageTitle)
			p.AddField("info", "login success")
		} else {
			p.AddField("info", "login failed")
		}

	}

	return
	if !p.IsSucc() {
		println(p.Errormsg())
		return
	}

	query := p.GetHtmlParser()
	var urls []string
	query.Find("h3[class='repo-list-name'] a").Each(func(i int, s *goquery.Selection) {
		href, _ := s.Attr("href")
		urls = append(urls, "http://github.com/"+href)
	})
	// these urls will be saved and crawed by other coroutines.
	p.AddTargetRequests(urls, "html")

	name := query.Find(".entry-title .author").Text()
	name = strings.Trim(name, " \t\n")
	repository := query.Find(".entry-title .js-current-repository").Text()
	repository = strings.Trim(repository, " \t\n")
	//readme, _ := query.Find("#readme").Html()
	if name == "" {
		p.SetSkip(true)
	}
	// the entity we want to save by Pipeline
	p.AddField("author", name)
	p.AddField("project", repository)
	//p.AddField("readme", readme)
}