X-Git-Url: http://wamblee.org/gitweb/?a=blobdiff_plain;ds=sidebyside;f=crawler%2Fbasic%2Fsrc%2Forg%2Fwamblee%2Fcrawler%2FAbstractPageRequest.java;h=2e598005ff3a453150a50884cccfd649889ee3c6;hb=5685a836b9208ff8babfe5ac5b30c5f86d27cf96;hp=dd9e8ae71a5d94dc9856a27a253daaa4f4a3181b;hpb=0c7e22e06b8aa3e5e0e516f2f3c46eee6215bd85;p=utils

diff --git a/crawler/basic/src/org/wamblee/crawler/AbstractPageRequest.java b/crawler/basic/src/org/wamblee/crawler/AbstractPageRequest.java
index dd9e8ae7..2e598005 100644
--- a/crawler/basic/src/org/wamblee/crawler/AbstractPageRequest.java
+++ b/crawler/basic/src/org/wamblee/crawler/AbstractPageRequest.java
@@ -17,9 +17,10 @@
 package org.wamblee.crawler;
 
 import java.io.ByteArrayOutputStream;
-import java.io.File;
 import java.io.IOException;
-import java.io.PrintStream;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.List;
 
 import javax.xml.transform.OutputKeys;
 import javax.xml.transform.Transformer;
@@ -41,9 +42,8 @@ import org.apache.xml.serialize.OutputFormat;
 import org.apache.xml.serialize.XMLSerializer;
 import org.w3c.dom.Document;
 import org.w3c.tidy.Tidy;
-import org.wamblee.io.FileResource;
-import org.wamblee.xml.DOMUtility;
-import org.wamblee.xml.XSLT;
+import org.wamblee.xml.DomUtils;
+import org.wamblee.xml.XslTransformer;
 
 /**
  * General support claas for all kinds of requests.
@@ -59,10 +59,12 @@ public abstract class AbstractPageRequest implements PageRequest {
     private int _maxDelay;
 
     private NameValuePair[] _params;
+    
+    private NameValuePair[] _headers;
 
     private String _xslt;
-
-    private PrintStream _os;
+    
+    private XslTransformer _transformer; 
 
     /**
      * Constructs the request.
@@ -73,24 +75,28 @@ public abstract class AbstractPageRequest implements PageRequest {
      *            Maximum delay before executing a request.
      * @param aParams
      *            Request parameters to use.
+     * @param aHeaders
+     *            Request headers to use. 
      * @param aXslt
      *            XSLT used to convert the response.
-     * @param aOs
-     *            Output stream for logging (if null then no logging is done).
      */
     protected AbstractPageRequest(int aMaxTries, int aMaxDelay,
-            NameValuePair[] aParams, String aXslt, PrintStream aOs) {
+            NameValuePair[] aParams, NameValuePair[] aHeaders, String aXslt, XslTransformer aTransformer) {
         if (aParams == null) {
             throw new IllegalArgumentException("aParams is null");
         }
+        if (aHeaders == null) {
+            throw new IllegalArgumentException("aHeaders is null");
+        }
         if (aXslt == null) {
             throw new IllegalArgumentException("aXslt is null");
         }
         _maxTries = aMaxTries;
         _maxDelay = aMaxDelay;
         _params = aParams;
+        _headers = aHeaders;
         _xslt = aXslt;
-        _os = aOs;
+        _transformer = aTransformer;
     }
 
     /*
@@ -105,10 +111,23 @@ public abstract class AbstractPageRequest implements PageRequest {
     /**
      * Gets the parameters for the request.
      * 
+     * @param aParams Additional parameters to use, obtained from another page, most likely as
+     *    hidden form fields. 
      * @return Request parameters.
      */
-    protected NameValuePair[] getParameters() {
-        return _params;
+    protected NameValuePair[] getParameters(NameValuePair[] aParams) {
+        List<NameValuePair> params = new ArrayList<NameValuePair>(); 
+        params.addAll(Arrays.asList(_params));
+        params.addAll(Arrays.asList(aParams));
+        return params.toArray(new NameValuePair[0]);
+    }
+    
+    /**
+     * Gets the headers for the request. 
+     * @return Request headers. 
+     */
+    protected NameValuePair[] getHeaders() { 
+        return _headers;
     }
 
     /**
@@ -127,6 +146,11 @@ public abstract class AbstractPageRequest implements PageRequest {
      */
     protected Document executeMethod(HttpClient aClient, HttpMethod aMethod)
             throws IOException, TransformerException {
+        
+        for (NameValuePair header: getHeaders()) { 
+            aMethod.setRequestHeader(header.getName(), header.getValue());
+        }
+        
         int triesLeft = _maxTries;
         while (triesLeft > 0) {
             triesLeft--;
@@ -160,20 +184,21 @@ public abstract class AbstractPageRequest implements PageRequest {
         try {
             aMethod = executeWithRedirects(aClient, aMethod);
             byte[] xhtmlData = getXhtml(aMethod);
-
-            Document transformed = new XSLT().transform(xhtmlData,
-                    new FileResource(new File(_xslt)));
-            _os.println("Transformed result is: ");
+            
+     
+            Document transformed = _transformer.transform(xhtmlData,
+                    _transformer.resolve(_xslt));
+            ByteArrayOutputStream os = new ByteArrayOutputStream(); 
             Transformer transformer = TransformerFactory.newInstance()
                     .newTransformer();
             transformer.setParameter(OutputKeys.INDENT, "yes");
             transformer.setParameter(OutputKeys.METHOD, "xml");
             transformer.transform(new DOMSource(transformed), new StreamResult(
-                    _os));
-
+                    os));
+            LOG.debug("Transformed result is \n" + os.toString());
             return transformed;
         } catch (TransformerConfigurationException e) {
-            throw new RuntimeException(e.getMessage(), e);
+            throw new TransformerException("Transformer configuration problem", e);
         } finally {
             // Release the connection.
             aMethod.releaseConnection();
@@ -195,24 +220,21 @@ public abstract class AbstractPageRequest implements PageRequest {
         tidy.setXHTML(true);
         tidy.setQuiet(true);
         tidy.setShowWarnings(false);
-        if (_os != null) {
-            _os.println("Content of '" + aMethod.getURI() + "'");
-            _os.println();
-        }
+      
         // We write the jtidy output to XML since the DOM tree it produces is
         // not namespace aware and namespace awareness is required by XSLT.
         // An alternative is to configure namespace awareness of the XML parser
         // in a system wide way.
-        Document w3cDoc = tidy.parseDOM(aMethod.getResponseBodyAsStream(), _os);
-        DOMUtility.removeDuplicateAttributes(w3cDoc);
+        ByteArrayOutputStream os = new ByteArrayOutputStream(); 
+        Document w3cDoc = tidy.parseDOM(aMethod.getResponseBodyAsStream(), os);
+        DomUtils.removeDuplicateAttributes(w3cDoc);
+        LOG.debug("Content of response is \n" + os.toString()); 
 
         ByteArrayOutputStream xhtml = new ByteArrayOutputStream();
         XMLSerializer serializer = new XMLSerializer(xhtml, new OutputFormat());
         serializer.serialize(w3cDoc);
         xhtml.flush();
-        if (_os != null) {
-            _os.println();
-        }
+
         return xhtml.toByteArray();
     }
 
@@ -260,7 +282,7 @@ public abstract class AbstractPageRequest implements PageRequest {
             // recursion.
         }
         default: {
-            throw new RuntimeException("Method failed: "
+            throw new IOException("Method failed: "
                     + aMethod.getStatusLine());
         }
         }