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;
import org.apache.commons.httpclient.Header;
import org.apache.commons.httpclient.HttpClient;
-import org.apache.commons.httpclient.HttpException;
import org.apache.commons.httpclient.HttpMethod;
import org.apache.commons.httpclient.HttpStatus;
import org.apache.commons.httpclient.NameValuePair;
import org.apache.commons.httpclient.methods.GetMethod;
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
+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.XSLT;
+import org.wamblee.xml.DomUtils;
+import org.wamblee.xml.XslTransformer;
/**
* General support claas for all kinds of requests.
private int _maxDelay;
private NameValuePair[] _params;
+
+ private NameValuePair[] _headers;
private String _xslt;
-
- private PrintStream _os;
+
+ private XslTransformer _transformer;
/**
- * Constructs the request.
- * @param aMaxTries Maximum retries to perform.
- * @param aMaxDelay Maximum delay before executing a request.
- * @param aParams 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).
+ * Constructs the request.
+ *
+ * @param aMaxTries
+ * Maximum retries to perform.
+ * @param aMaxDelay
+ * 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.
*/
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;
}
/*
}
/**
- * Gets the parameters for the request.
- * @return Request parameters.
+ * Gets the parameters for the request.
+ *
+ * @return Request parameters.
*/
protected NameValuePair[] getParameters() {
return _params;
}
+
+ /**
+ * Gets the headers for the request.
+ * @return Request headers.
+ */
+ protected NameValuePair[] getHeaders() {
+ return _headers;
+ }
/**
- * Executes the request with a random delay and with a maximum number of
- * retries.
- * @param aClient HTTP client to use.
- * @param aMethod Method representing the request.
- * @return XML document describing the response.
- * @throws TransformerException In case transformation of the HTML to XML fails.
+ * Executes the request with a random delay and with a maximum number of
+ * retries.
+ *
+ * @param aClient
+ * HTTP client to use.
+ * @param aMethod
+ * Method representing the request.
+ * @return XML document describing the response.
+ * @throws IOException
+ * In case of IO problems.
+ * @throws TransformerException
+ * In case transformation of the HTML to XML fails.
*/
protected Document executeMethod(HttpClient aClient, HttpMethod aMethod)
- throws TransformerException {
+ throws IOException, TransformerException {
+
+ for (NameValuePair header: getHeaders()) {
+ aMethod.setRequestHeader(header.getName(), header.getValue());
+ }
+
int triesLeft = _maxTries;
while (triesLeft > 0) {
triesLeft--;
}
/**
- * Executes the request without doing any retries in case XSLT transformation
- * fails.
- * @param aClient HTTP client to use.
- * @param aMethod Method to execute.
- * @return XML document containing the result.
- * @throws TransformerException In case transformation of the result to XML fails.
+ * Executes the request without doing any retries in case XSLT
+ * transformation fails.
+ *
+ * @param aClient
+ * HTTP client to use.
+ * @param aMethod
+ * Method to execute.
+ * @return XML document containing the result.
+ * @throws IOException
+ * In case of IO problems.
+ * @throws TransformerException
+ * In case transformation of the result to XML fails.
*/
protected Document executeMethodWithoutRetries(HttpClient aClient,
- HttpMethod aMethod) throws TransformerException {
+ HttpMethod aMethod) throws IOException, TransformerException {
try {
- // Execute the method.
aMethod = executeWithRedirects(aClient, aMethod);
-
- // Transform the HTML into wellformed XML.
- Tidy tidy = new Tidy();
- tidy.setXHTML(true);
- tidy.setQuiet(true);
- tidy.setShowWarnings(false);
- if (_os != null) {
- _os.println("Content of '" + aMethod.getURI() + "'");
- _os.println();
- }
- // We let jtidy produce raw output because the DOM it produces is
- // is not namespace aware. We let the XSLT processor parse the XML
- // again
- // to ensure that the XSLT uses a namespace aware DOM tree. An
- // alternative
- // is to configure namespace awareness of the XML parser in a system
- // wide way.
- ByteArrayOutputStream xhtml = new ByteArrayOutputStream();
- tidy.parse(aMethod.getResponseBodyAsStream(), xhtml);
- _os.print(new String(xhtml.toByteArray()));
- // Obtaining the XML as dom is not used.
- // Document w3cDoc = tidy.parseDOM(method.getResponseBodyAsStream(),
- // _os);
- if (_os != null) {
- _os.println();
- }
- xhtml.flush();
- byte[] xhtmlData = xhtml.toByteArray();
- Document transformed = new XSLT().transform(xhtmlData,
- new FileResource(new File(_xslt)));
- _os.println("Transformed result is: ");
+ byte[] xhtmlData = getXhtml(aMethod);
+
+ 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 (HttpException e) {
- throw new RuntimeException(e.getMessage(), e);
- } catch (IOException e) {
- throw new RuntimeException(e.getMessage(), e);
} catch (TransformerConfigurationException e) {
- throw new RuntimeException(e.getMessage(), e);
+ throw new TransformerException("Transformer configuration problem", e);
} finally {
// Release the connection.
aMethod.releaseConnection();
}
}
+ /**
+ * Gets the result of the HTTP method as an XHTML document.
+ *
+ * @param aMethod
+ * Method to invoke.
+ * @return XHTML as a byte array.
+ * @throws IOException
+ * In case of problems obtaining the XHTML.
+ */
+ private byte[] getXhtml(HttpMethod aMethod) throws IOException {
+ // Transform the HTML into wellformed XML.
+ Tidy tidy = new Tidy();
+ tidy.setXHTML(true);
+ tidy.setQuiet(true);
+ tidy.setShowWarnings(false);
+
+ // 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.
+ 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();
+
+ return xhtml.toByteArray();
+ }
+
/**
* Sleeps for a random time but no more than the maximum delay.
- *
+ *
*/
private void delay() {
try {
}
/**
- * Executes the request and follows redirects if needed.
- * @param aClient HTTP client to use.
- * @param aMethod Method to use.
- * @return Final HTTP method used (differs from the parameter passed in in case
- * of redirection).
- * @throws IOException In case of network problems.
+ * Executes the request and follows redirects if needed.
+ *
+ * @param aClient
+ * HTTP client to use.
+ * @param aMethod
+ * Method to use.
+ * @return Final HTTP method used (differs from the parameter passed in in
+ * case of redirection).
+ * @throws IOException
+ * In case of network problems.
*/
private HttpMethod executeWithRedirects(HttpClient aClient,
HttpMethod aMethod) throws IOException {
Header header = aMethod.getResponseHeader(REDIRECT_HEADER);
aMethod = new GetMethod(header.getValue());
return executeWithRedirects(aClient, aMethod); // TODO protect
- // against infinite
- // recursion.
+ // against infinite
+ // recursion.
}
default: {
- throw new RuntimeException("Method failed: "
+ throw new IOException("Method failed: "
+ aMethod.getStatusLine());
}
}
git://wamblee.org / utils / blobdiff