// 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)
}
// 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")
}
}
// 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) {
query := p.GetHtmlParser()
var urls []string
query.Find("#threadlisttableid tbody").Each(func(i int, s *goquery.Selection) {
if s.HasClass("emptb") {
return
}
href, _ := s.Find("tbody tr .icn a").Attr("href")
urls = append(urls, href)
})
// these urls will be saved and crawed by other coroutines.
p.AddTargetRequests(urls, "html")
title := query.Find("#thread_subject").Text()
title = strings.Trim(title, "\t\n\r")
author := query.Find("#postlist div .authi").Eq(0).Text()
author = strings.Trim(author, "\t\r\n")
if title == "" || author == "" {
p.SetSkip(true)
}
p.AddField("title", title)
p.AddField("author", author)
}
// 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) {
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)
}
func (this *PlantProcesser) Process(p *page.Page) {
if !p.IsSucc() {
println(p.Errormsg())
return
}
query := p.GetHtmlParser()
if !this.isPlant(query, p) {
p.SetSkip(true)
}
this.getName(query, p)
this.getSummary(query, p)
this.getCatalog(query, p)
p.AddTargetRequests(this.getUrls(query), "html")
}
// 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()
currentUrl := p.GetRequest().GetUrl()
var urls []string
query.Find("a").Each(func(i int, s *goquery.Selection) {
href, _ := s.Attr("href")
urlHref, err := url.Parse(href)
if err != nil {
mlog.LogInst().LogError(err.Error())
return
}
if !urlHref.IsAbs() {
href = currentUrl + href
}
// Temporarily check in crawler.go, it will be implemented in pattern package.
if checkMatchPattern(base, href) {
visited, _ := rep.CheckIfVisited(href)
if !visited {
rep.VisitedNewNode(href)
// urls = append(urls, href)
urlstr.UploadURL(href)
}
}
})
// store content to db
fmt.Printf("====store & commit : %s====\n\n\n", currentUrl)
content, _ := query.Html()
// content := ""
storage.StoreInsert(collection, storage.StoreFormat{currentUrl, content})
urlstr.CommitURL(currentUrl)
releaseSlot <- 1
url := GetOneURL()
if url != "" {
urls = append(urls, url)
}
p.AddTargetRequests(urls, "html")
}
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")
}
func TestDownloadHtml(t *testing.T) {
//return
//request := request.NewRequest("http://live.sina.com.cn/zt/api/l/get/finance/globalnews1/index.htm?format=json&callback=t13975294&id=23521&pagesize=45&dire=f&dpc=1")
var req *request.Request
req = request.NewRequest("http://live.sina.com.cn/zt/l/v/finance/globalnews1/", "html", "", "GET", "", nil, nil, nil, nil)
var dl downloader.Downloader
dl = downloader.NewHttpDownloader()
var p *page.Page
p = dl.Download(req)
var doc *goquery.Document
doc = p.GetHtmlParser()
//fmt.Println(doc)
//body := p.GetBodyStr()
//fmt.Println(body)
var s *goquery.Selection
s = doc.Find("body")
if s.Length() < 1 {
t.Error("html parse failed!")
}
/*
doc, err := goquery.NewDocument("http://live.sina.com.cn/zt/l/v/finance/globalnews1/")
if err != nil {
fmt.Printf("%v",err)
}
s := doc.Find("meta");
fmt.Println(s.Length())
resp, err := http.Get("http://live.sina.com.cn/zt/l/v/finance/globalnews1/")
if err != nil {
fmt.Printf("%v",err)
}
defer resp.Body.Close()
doc, err = goquery.NewDocumentFromReader(resp.Body)
s = doc.Find("meta");
fmt.Println(s.Length())
*/
}
/*
** 解析页面，把粉丝的信息存入dynamodb，同时把接下来要爬取的url存入sqs
*/
func (this *MyPageProcesser) Process(p *page.Page) {
if !p.IsSucc() {
glog.Errorln(p.Errormsg())
return
}
/*
** 打印爬取得页面
*/
glog.Infoln(p)
query := p.GetHtmlParser()
if Urls[i] == "weibo.cn" {
i = i + 1
}
if UrlsLevel[i] == 0 {
glog.Infoln("layer:", crawlUrl.Layer)
this.w.GetNextPageUrl(query, p)
this.w.GetFriendsUrl(query, p)
} else if UrlsLevel[i] == 1 {
this.w.GetFriendsInfo(query)
}
// if crawlUrl.Layer == 0 {
// } else if crawlUrl.Layer == 1 {
// glog.Infoln("layer:", crawlUrl.Layer)
// this.w.GetNextPageUrl(query, p)
// this.w.GetFFUrl(query)
// } else if crawlUrl.Layer == 2 {
// glog.Infoln("layer:", crawlUrl.Layer)
// this.w.GetFFInfo(query)
// }
//
header_num := rand.Intn(9)
header_json := headerJson[header_num]
i = i + 1
p.AddTargetRequestWithHeaderFile(Urls[i], "html", header_json)
}
func (this MyPageProcesser) Process(p *page.Page) {
query := p.GetHtmlParser()
if p.GetUrlTag() == "index" {
query.Find(`div[class="main area"] div[class="lc"] ul li a`).Each(func(i int, s *goquery.Selection) {
url, isExsit := s.Attr("href")
if isExsit {
reg := regexp.MustCompile(`^do not know what is this`)
var fmtStr string
if rxYule.MatchString(url) {
reg = rxYule
fmtStr = wkSohuYule
}
if rxPic.MatchString(url) {
reg = rxPic
fmtStr = wkSohuPic
}
regxpArrag := reg.FindStringSubmatch(url)
if len(regxpArrag) == 2 {
addRequest(p, "changyan", fmt.Sprintf(fmtStr, regxpArrag[1]), "", s.Text())
}
}
})
}
if p.GetUrlTag() == "changyan" {
jsonMap := ChangyanJson{}
err := json.NewDecoder(strings.NewReader(p.GetBodyStr())).Decode(&jsonMap)
if err == nil {
content, ok := p.GetRequest().GetMeta().(string)
if ok {
fmt.Println("Title:", content, " CommentCount:", jsonMap.ListData.OuterCmtSum, " ParticipationCount:", jsonMap.ListData.ParticipationSum)
}
}
}
}
func (this SitePageProcesser) Process(p *page.Page) {
fmt.Println("Site Page Processer")
if p.GetUrlTag() == "index" {
query := p.GetHtmlParser()
query.Find("ul[class='audioList fontYaHei'] li a").Each(func(i int, s *goquery.Selection) {
strTitle, _ := s.Attr("title")
strUrl, _ := s.Attr("data-url")
if !IsFileExist(strTitle) {
strFileName := fmt.Sprintf("%s.mp3", strTitle)
fmt.Println(strFileName)
cmd := exec.Command("/usr/local/bin/wget", strUrl, "-O", strFileName)
err := cmd.Run()
if err != nil {
fmt.Println(err)
}
d, _ := cmd.Output()
fmt.Println(string(d))
}
})
}
}
// 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 fetch_content string
query := p.GetHtmlParser()
content := p.GetBodyStr()
reg := regexp.MustCompile(`class="([0-9a-zA-Z_-]*content[0-9a-zA-Z_-]*)"`)
reg_res := reg.FindAllStringSubmatch(content, -1)
class_content := make([]string, 0)
for _, class := range reg_res {
submatch := class[1]
class_content = append(class_content, submatch)
}
removeDuplicate(&class_content)
for _, class := range class_content {
query.Find("." + class).Each(func(i int, s *goquery.Selection) {
text := strings.Trim(s.Text(), " \t\n")
text = strings.Replace(text, " ", "", -1)
text = strings.Replace(text, "\n", "", -1)
text = strings.Replace(text, "\t", "", -1)
if text != "" {
fetch_content = fetch_content + text
}
})
}
if fetch_content != "" {
p.AddField("content", fetch_content)
}
}
// 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)
}