+sina blog
parent
900172c968
commit
c733046045
|
@ -1351,6 +1351,7 @@ public class XpathSelectorTest {
|
||||||
public void testOschina() {
|
public void testOschina() {
|
||||||
Html html1 = new Html(html);
|
Html html1 = new Html(html);
|
||||||
Assert.assertEquals("再次吐槽easyui", html1.xpath(".//*[@class='QTitle']/h1/a").toString());
|
Assert.assertEquals("再次吐槽easyui", html1.xpath(".//*[@class='QTitle']/h1/a").toString());
|
||||||
|
System.out.println(html1.regex("(<body>.*?</body>)").links().toStrings());
|
||||||
}
|
}
|
||||||
|
|
||||||
}
|
}
|
||||||
|
|
|
@ -1,7 +1,8 @@
|
||||||
package us.codecraft.webmagic.samples;
|
package us.codecraft.webmagic.samples;
|
||||||
|
|
||||||
import us.codecraft.webmagic.Site;
|
|
||||||
import us.codecraft.webmagic.Page;
|
import us.codecraft.webmagic.Page;
|
||||||
|
import us.codecraft.webmagic.Site;
|
||||||
|
import us.codecraft.webmagic.Spider;
|
||||||
import us.codecraft.webmagic.processor.PageProcessor;
|
import us.codecraft.webmagic.processor.PageProcessor;
|
||||||
|
|
||||||
/**
|
/**
|
||||||
|
@ -15,7 +16,7 @@ public class SinaBlogProcesser implements PageProcessor {
|
||||||
|
|
||||||
@Override
|
@Override
|
||||||
public void process(Page page) {
|
public void process(Page page) {
|
||||||
page.addTargetRequests(page.getHtml().links().regex("(http://blog\\.sina\\.com\\.cn/s/blog_.*)").toStrings());
|
page.addTargetRequests(page.getHtml().xpath("//div[@class='articalfrontback SG_j_linedot1 clearfix']").links().toStrings());
|
||||||
page.putField("title", page.getHtml().xpath("//div[@class='articalTitle']/h2"));
|
page.putField("title", page.getHtml().xpath("//div[@class='articalTitle']/h2"));
|
||||||
page.putField("content",page.getHtml().xpath("//div[@id='articlebody']//div[@class='articalContent']"));
|
page.putField("content",page.getHtml().xpath("//div[@id='articlebody']//div[@class='articalContent']"));
|
||||||
page.putField("id",page.getUrl().regex("http://blog\\.sina\\.com\\.cn/s/blog_(\\w+)"));
|
page.putField("id",page.getUrl().regex("http://blog\\.sina\\.com\\.cn/s/blog_(\\w+)"));
|
||||||
|
@ -26,9 +27,13 @@ public class SinaBlogProcesser implements PageProcessor {
|
||||||
@Override
|
@Override
|
||||||
public Site getSite() {
|
public Site getSite() {
|
||||||
if (site==null){
|
if (site==null){
|
||||||
site = Site.me().setDomain("blog.sina.com.cn").addStartUrl("http://blog.sina.com.cn/flashsword20").setSleepTime(3000).
|
site = Site.me().setDomain("blog.sina.com.cn").addStartUrl("http://blog.sina.com.cn/s/blog_4701280b0102egl0.html").setSleepTime(3000).
|
||||||
setUserAgent("Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_2) AppleWebKit/537.31 (KHTML, like Gecko) Chrome/26.0.1410.65 Safari/537.31");
|
setUserAgent("Mozilla/5.0 (Macintosh; Intel Mac OS X 10_7_2) AppleWebKit/537.31 (KHTML, like Gecko) Chrome/26.0.1410.65 Safari/537.31");
|
||||||
}
|
}
|
||||||
return site;
|
return site;
|
||||||
}
|
}
|
||||||
|
|
||||||
|
public static void main(String[] args) {
|
||||||
|
Spider.create(new SinaBlogProcesser()).run();
|
||||||
|
}
|
||||||
}
|
}
|
||||||
|
|
|
@ -1,5 +1,6 @@
|
||||||
package us.codecraft.webmagic.processor;
|
package us.codecraft.webmagic.processor;
|
||||||
|
|
||||||
|
import org.junit.Ignore;
|
||||||
import org.junit.Test;
|
import org.junit.Test;
|
||||||
import us.codecraft.webmagic.Spider;
|
import us.codecraft.webmagic.Spider;
|
||||||
import us.codecraft.webmagic.pipeline.FilePipeline;
|
import us.codecraft.webmagic.pipeline.FilePipeline;
|
||||||
|
@ -16,6 +17,7 @@ import java.io.IOException;
|
||||||
*/
|
*/
|
||||||
public class DiaoyuwengProcessorTest {
|
public class DiaoyuwengProcessorTest {
|
||||||
|
|
||||||
|
@Ignore
|
||||||
@Test
|
@Test
|
||||||
public void test() throws IOException {
|
public void test() throws IOException {
|
||||||
DiaoyuwengProcessor diaoyuwengProcessor = new DiaoyuwengProcessor();
|
DiaoyuwengProcessor diaoyuwengProcessor = new DiaoyuwengProcessor();
|
||||||
|
|
Loading…
Reference in New Issue