Now using dependency injection for the XslTransformer instead of

[utils] / crawler / basic / src / org / wamblee / crawler / AbstractPageRequest.java

diff --git a/crawler/basic/src/org/wamblee/crawler/AbstractPageRequest.java b/crawler/basic/src/org/wamblee/crawler/AbstractPageRequest.java
index dd9e8ae71a5d94dc9856a27a253daaa4f4a3181b..5cb4fae6e40bc2f3b0215118bad67ca52028e657 100644 (file)
--- a/crawler/basic/src/org/wamblee/crawler/AbstractPageRequest.java
+++ b/crawler/basic/src/org/wamblee/crawler/AbstractPageRequest.java
@@ -17,9 +17,7 @@
 package org.wamblee.crawler;
 
 import java.io.ByteArrayOutputStream;
-import java.io.File;
 import java.io.IOException;
-import java.io.PrintStream;
 
 import javax.xml.transform.OutputKeys;
 import javax.xml.transform.Transformer;
@@ -41,9 +39,8 @@ 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.XslTransformer;
 
 /**
  * General support claas for all kinds of requests.
@@ -61,8 +58,8 @@ public abstract class AbstractPageRequest implements PageRequest {
     private NameValuePair[] _params;
 
     private String _xslt;
-
-    private PrintStream _os;
+    
+    private XslTransformer _transformer; 
 
     /**
      * Constructs the request.
@@ -75,11 +72,9 @@ public abstract class AbstractPageRequest implements PageRequest {
      *            Request parameters 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, String aXslt, XslTransformer aTransformer) {
         if (aParams == null) {
             throw new IllegalArgumentException("aParams is null");
         }
@@ -90,7 +85,7 @@ public abstract class AbstractPageRequest implements PageRequest {
         _maxDelay = aMaxDelay;
         _params = aParams;
         _xslt = aXslt;
-        _os = aOs;
+        _transformer = aTransformer;
     }
 
     /*
@@ -161,16 +156,16 @@ public abstract class AbstractPageRequest implements PageRequest {
             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);
@@ -195,24 +190,21 @@ 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);
+        ByteArrayOutputStream os = new ByteArrayOutputStream(); 
+        Document w3cDoc = tidy.parseDOM(aMethod.getResponseBodyAsStream(), os);
         DOMUtility.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();
     }