// Spider is scheduler module for all the other modules, like downloader, pipeline, scheduler and etc.
// The taskname could be empty string too, or it can be used in Pipeline for record the result crawled by which task;
func NewSpider(pageinst page_processer.PageProcesser, taskname string) *Spider {
mlog.StraceInst().Open()
ap := &Spider{taskname: taskname, pPageProcesser: pageinst}
// init filelog.
ap.CloseFileLog()
ap.exitWhenComplete = true
ap.sleeptype = "fixed"
ap.startSleeptime = 0
// init spider
if ap.pScheduler == nil {
ap.SetScheduler(scheduler.NewQueueScheduler(false))
}
if ap.pDownloader == nil {
ap.SetDownloader(downloader.NewHttpDownloader())
}
mlog.StraceInst().Println("** start spider **")
ap.pPiplelines = make([]pipeline.Pipeline, 0)
return ap
}
func TestDownloadJson(t *testing.T) {
var req *request.Request
req = request.NewRequest("http://live.sina.com.cn/zt/api/l/get/finance/globalnews1/index.htm?format=json&id=23521&pagesize=4&dire=f&dpc=1", "json")
var dl downloader.Downloader
dl = downloader.NewHttpDownloader()
var p *page.Page
p = dl.Download(req)
var jsonMap interface{}
jsonMap = p.GetJsonMap()
fmt.Printf("%v", jsonMap)
//fmt.Println(doc)
//body := p.GetBodyStr()
//fmt.Println(body)
}
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())
*/
}
func TestCharSetChange(t *testing.T) {
var req *request.Request
//req = request.NewRequest("http://stock.finance.sina.com.cn/usstock/api/jsonp.php/t/US_CategoryService.getList?page=1&num=60", "jsonp")
req = request.NewRequest("http://soft.chinabyte.com/416/13164916.shtml", "html", "", "GET", "", nil, nil, nil, nil)
var dl downloader.Downloader
dl = downloader.NewHttpDownloader()
var p *page.Page
p = dl.Download(req)
//hp := p.GetHtmlParser()
//fmt.Printf("%v", jsonMap)
//fmt.Println(doc)
p.GetBodyStr()
body := p.GetBodyStr()
fmt.Println(body)
}
func (this *Spider) close() {
this.SetScheduler(scheduler.NewQueueScheduler(false))
this.SetDownloader(downloader.NewHttpDownloader())
this.pPiplelines = make([]pipeline.Pipeline, 0)
this.exitWhenComplete = true
}