Revert " Downloader 提供刷新组件的api,方便在spider中操作"

This reverts commit 2e2a0fd.

Author: sutra

webmagic-core/src/main/java/us/codecraft/webmagic/Site.java

@@ -40,20 +40,15 @@ public class Site {
 
     private static final Set<Integer> DEFAULT_STATUS_CODE_SET = new HashSet<Integer>();
 
-    private static final Set<Integer> DEFAULT_REFRESH_CODE_SET = new HashSet<>();
-
-    private Set<Integer> refreshCode = DEFAULT_REFRESH_CODE_SET;
     private Set<Integer> acceptStatCode = DEFAULT_STATUS_CODE_SET;
 
-
     private Map<String, String> headers = new HashMap<String, String>();
 
     private boolean useGzip = true;
 
     private boolean disableCookieManagement = false;
 
     static {
-        DEFAULT_REFRESH_CODE_SET.add(HttpConstant.StatusCode.FORBIDDEN);
         DEFAULT_STATUS_CODE_SET.add(HttpConstant.StatusCode.CODE_200);
     }
 
@@ -202,15 +197,6 @@ public Site setAcceptStatCode(Set<Integer> acceptStatCode) {
         return this;
     }
 
-    public Site setRefreshCode(Set<Integer> refreshCode){
-        this.refreshCode = refreshCode;
-        return this;
-    }
-    public Set<Integer> getRefreshCode(){
-        return refreshCode;
-
-    }
-
     /**
      * get acceptStatCode
      *

webmagic-core/src/main/java/us/codecraft/webmagic/Spider.java

@@ -424,10 +424,7 @@ private void onDownloadSuccess(Request request, Page page) {
                     pipeline.process(page.getResultItems(), this);
                 }
             }
-        } else if(site.getRefreshCode().contains(page.getStatusCode())) {
-            logger.info("page status code error, page {} , code: {}, start refresh downloader", request.getUrl(), page.getStatusCode());
-            downloader.refreshComponent(this);
-        }else {
+        } else {
             logger.info("page status code error, page {} , code: {}", request.getUrl(), page.getStatusCode());
         }
         sleep(site.getSleepTime());

webmagic-core/src/main/java/us/codecraft/webmagic/downloader/Downloader.java

@@ -18,18 +18,14 @@ public interface Downloader {
      * Downloads web pages and store in Page object.
      *
      * @param request request
-     * @param task    task
+     * @param task task
      * @return page
      */
-    Page download(Request request, Task task);
+    public Page download(Request request, Task task);
 
     /**
      * Tell the downloader how many threads the spider used.
-     *
      * @param threadNum number of threads
      */
-    void setThread(int threadNum);
-
-
-    void refreshComponent(Task task);
+    public void setThread(int threadNum);
 }

webmagic-core/src/main/java/us/codecraft/webmagic/downloader/HttpClientDownloader.java

@@ -111,17 +111,6 @@ public Page download(Request request, Task task) {
         }
     }
 
-
-    @Override
-    public void refreshComponent(Task task) {
-        if (proxyProvider != null ) {
-            proxyProvider.refreshProxy(task);
-        }
-
-            httpClients.remove(task.getSite().getDomain());
-
-    }
-
     @Override
     public void setThread(int thread) {
         httpClientGenerator.setPoolSize(thread);

webmagic-core/src/main/java/us/codecraft/webmagic/downloader/HttpClientGenerator.java

@@ -1,17 +1,13 @@
 package us.codecraft.webmagic.downloader;
 
-import java.io.File;
 import java.io.IOException;
 import java.security.KeyManagementException;
-import java.security.KeyStore;
-import java.security.KeyStoreException;
 import java.security.NoSuchAlgorithmException;
 import java.security.cert.CertificateException;
 import java.security.cert.X509Certificate;
 import java.util.Map;
 
 import javax.net.ssl.SSLContext;
-import javax.net.ssl.SSLContextSpi;
 import javax.net.ssl.TrustManager;
 import javax.net.ssl.X509TrustManager;
 
@@ -28,7 +24,6 @@
 import org.apache.http.conn.socket.PlainConnectionSocketFactory;
 import org.apache.http.conn.ssl.DefaultHostnameVerifier;
 import org.apache.http.conn.ssl.SSLConnectionSocketFactory;
-import org.apache.http.conn.ssl.TrustSelfSignedStrategy;
 import org.apache.http.impl.client.BasicCookieStore;
 import org.apache.http.impl.client.CloseableHttpClient;
 import org.apache.http.impl.client.DefaultHttpRequestRetryHandler;
@@ -37,7 +32,6 @@
 import org.apache.http.impl.conn.PoolingHttpClientConnectionManager;
 import org.apache.http.impl.cookie.BasicClientCookie;
 import org.apache.http.protocol.HttpContext;
-import org.apache.http.ssl.SSLContexts;
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
 
@@ -75,16 +69,16 @@ private SSLConnectionSocketFactory buildSSLConnectionSocketFactory() {
             return new SSLConnectionSocketFactory(sslContext, supportedProtocols,
                     null,
                     new DefaultHostnameVerifier()); // 优先绕过安全证书
-        } catch (KeyManagementException | CertificateException | KeyStoreException | IOException e) {
+        } catch (KeyManagementException e) {
             logger.error("ssl connection fail", e);
         } catch (NoSuchAlgorithmException e) {
             logger.error("ssl connection fail", e);
         }
         return SSLConnectionSocketFactory.getSocketFactory();
 	}
 
-    private SSLContext createIgnoreVerifySSL() throws NoSuchAlgorithmException, KeyManagementException, CertificateException, KeyStoreException, IOException {
-// 实现一个X509TrustManager接口，用于绕过验证，不用修改里面的方法
+    private SSLContext createIgnoreVerifySSL() throws NoSuchAlgorithmException, KeyManagementException {
+        // 实现一个X509TrustManager接口，用于绕过验证，不用修改里面的方法
         X509TrustManager trustManager = new X509TrustManager() {
 
             @Override
@@ -102,10 +96,10 @@ public X509Certificate[] getAcceptedIssuers() {
 
         };
 
-        SSLContext sc = SSLContext.getInstance("SSLv3");
+        SSLContext sc = SSLContext.getInstance("TLS");
         sc.init(null, new TrustManager[] { trustManager }, null);
         return sc;
-    }
+	}
 
     public HttpClientGenerator setPoolSize(int poolSize) {
         connectionManager.setMaxTotal(poolSize);

webmagic-core/src/main/java/us/codecraft/webmagic/utils/HttpConstant.java

@@ -28,7 +28,6 @@ public static abstract class Method {
     public static abstract class StatusCode {
 
         public static final int CODE_200 = 200;
-        public static final int FORBIDDEN = 403;
 
     }
 

webmagic-core/src/test/java/us/codecraft/webmagic/SpiderTest.java

@@ -57,11 +57,6 @@ public Site getSite() {
                 return Site.me().setSleepTime(0);
             }
         }).setDownloader(new Downloader() {
-            @Override
-            public void refreshComponent(Task task) {
-
-            }
-
             @Override
             public Page download(Request request, Task task) {
                 return new Page().setRawText("");

webmagic-core/src/test/java/us/codecraft/webmagic/downloader/MockGithubDownloader.java

@@ -28,11 +28,6 @@ public Page download(Request request, Task task) {
         return page;
     }
 
-    @Override
-    public void refreshComponent(Task task) {
-
-    }
-
     @Override
     public void setThread(int threadNum) {
     }

webmagic-extension/src/main/java/us/codecraft/webmagic/downloader/PhantomJSDownloader.java

@@ -42,12 +42,7 @@ public PhantomJSDownloader(String phantomJsCommand) {
         this.initPhantomjsCrawlPath();
         PhantomJSDownloader.phantomJsCommand = phantomJsCommand;
     }
-
-    @Override
-    public void refreshComponent(Task task) {
-
-    }
-
+    
     /**
      * 新增构造函数，支持crawl.js路径自定义，因为当其他项目依赖此jar包时，runtime.exec()执行phantomjs命令时无使用法jar包中的crawl.js
      * <pre>

webmagic-extension/src/test/java/us/codecraft/webmagic/downloader/MockGithubDownloader.java

@@ -9,10 +9,6 @@
  * @author code4crafter@gmail.com
  */
 public class MockGithubDownloader implements Downloader{
-    @Override
-    public void refreshComponent(Task task) {
-
-    }
 
     private String html = "\n" +
             "\n" +