add huaban processor

pull/17/head
yihua.huang 12 years ago
parent fe224cbf66
commit 42508af041

@ -15,6 +15,9 @@ public class ConsolePipeline implements Pipeline{
@Override
public void process(ResultItems resultItems,Task task) {
if (resultItems.isSkip()){
return;
}
System.out.println("get page: "+resultItems.getRequest().getUrl());
for (Map.Entry<String, Object> entry : resultItems.getAll().entrySet()) {
System.out.println(entry.getKey()+":\t"+entry.getValue());

@ -45,6 +45,9 @@ public class FilePipeline implements Pipeline {
if (!file.exists()) {
file.mkdirs();
}
if (resultItems.isSkip()){
return;
}
try {
PrintWriter printWriter = new PrintWriter(new FileWriter(path + DigestUtils.md5Hex(resultItems.getRequest().getUrl())+".html"));
printWriter.println("url:\t" + resultItems.getRequest().getUrl());

@ -53,6 +53,7 @@ public class SeleniumDownloader implements Downloader,Destroyable {
logger.warn("interrupted", e);
return null;
}
logger.info("downloading page " + request.getUrl());
webDriver.get(request.getUrl());
WebDriver.Options manage = webDriver.manage();
Site site = task.getSite();

@ -7,8 +7,6 @@ import org.openqa.selenium.WebDriver;
import org.openqa.selenium.WebElement;
import org.openqa.selenium.chrome.ChromeDriver;
import java.util.List;
/**
* @author yihua.huang@dianping.com <br>
* @date: 13-7-26 <br>
@ -18,14 +16,12 @@ public class SeleniumTest {
@Ignore("need chrome driver")
@Test
public void test(){
System.getProperties().setProperty("webdriver.chrome.driver","/Users/yihua/Downloads/chromedriver");
public void testSelenium() {
System.getProperties().setProperty("webdriver.chrome.driver", "/Users/yihua/Downloads/chromedriver");
WebDriver webDriver = new ChromeDriver();
webDriver.get("http://huaban.com/");
List<WebElement> elements = webDriver.findElements(By.xpath("/html"));
for (WebElement element : elements) {
System.out.println(element.getAttribute("outerHTML"));
}
WebElement webElement = webDriver.findElement(By.xpath("/html"));
System.out.println(webElement.getAttribute("outerHTML"));
webDriver.close();
}
}

@ -22,6 +22,11 @@
<artifactId>webmagic-misc</artifactId>
<version>${project.version}</version>
</dependency>
<dependency>
<groupId>us.codecraft</groupId>
<artifactId>webmagic-selenium</artifactId>
<version>${project.version}</version>
</dependency>
<dependency>
<groupId>junit</groupId>
<artifactId>junit</artifactId>

@ -0,0 +1,45 @@
package us.codecraft.webmagic.samples;
import us.codecraft.webmagic.Page;
import us.codecraft.webmagic.Site;
import us.codecraft.webmagic.Spider;
import us.codecraft.webmagic.pipeline.FilePipeline;
import us.codecraft.webmagic.processor.PageProcessor;
import us.codecraft.webmagic.scheduler.RedisScheduler;
import us.codecraft.webmagic.selenium.downloader.SeleniumDownloader;
/**
* @author yihua.huang@dianping.com <br>
* @date: 13-7-26 <br>
* Time: 4:08 <br>
*/
public class HuabanProcessor implements PageProcessor {
private Site site;
@Override
public void process(Page page) {
page.addTargetRequests(page.getHtml().links().regex("http://huaban\\.com/.*").all());
if (page.getUrl().toString().contains("pins")) {
page.putField("img", page.getHtml().xpath("//div[@id='pin_img']/img/@src").toString());
} else {
page.getResultItems().setSkip(true);
}
}
@Override
public Site getSite() {
if (site == null) {
site = Site.me().setDomain("huaban.com").addStartUrl("http://huaban.com/");
}
return site;
}
public static void main(String[] args) {
Spider.create(new HuabanProcessor())
.scheduler(new RedisScheduler("localhost"))
.pipeline(new FilePipeline("/data/webmagic/test/"))
.downloader(new SeleniumDownloader("/Users/yihua/Downloads/chromedriver"))
.runAsync();
}
}
Loading…
Cancel
Save