2 * Copyright 2005 the original author or authors.
4 * Licensed under the Apache License, Version 2.0 (the "License");
5 * you may not use this file except in compliance with the License.
6 * You may obtain a copy of the License at
8 * http://www.apache.org/licenses/LICENSE-2.0
10 * Unless required by applicable law or agreed to in writing, software
11 * distributed under the License is distributed on an "AS IS" BASIS,
12 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13 * See the License for the specific language governing permissions and
14 * limitations under the License.
17 package org.wamblee.crawler;
19 import java.io.IOException;
20 import java.io.PrintStream;
22 import javax.xml.transform.TransformerException;
24 import org.apache.commons.httpclient.HttpClient;
25 import org.apache.commons.httpclient.HttpMethod;
26 import org.apache.commons.httpclient.NameValuePair;
27 import org.apache.commons.httpclient.methods.GetMethod;
28 import org.w3c.dom.Document;
31 * Gets a page by issueing a get request.
33 public class GetPageRequest extends AbstractPageRequest {
36 * Constructs the request.
37 * @param aMaxTries Maximum number of retries.
38 * @param aMaxDelay Maximum delay before executing the request.
39 * @param aParams Request parameters to use.
40 * @param aXslt XSLT to use.
42 public GetPageRequest(int aMaxTries, int aMaxDelay, NameValuePair[] aParams, String aXslt) {
43 super(aMaxTries, aMaxDelay, aParams, aXslt, null);
47 * Constructs the request.
48 * @param aMaxTries Maximum number of retries.
49 * @param aMaxDelay Maximum delay before executing the request.
50 * @param aParams Request parameters to use.
51 * @param aXslt XSLT to use.
52 * @param aOs Logging output stream to use.
54 public GetPageRequest(int aMaxTries, int aMaxDelay, NameValuePair[] aParams, String aXslt, PrintStream aOs) {
55 super(aMaxTries, aMaxDelay, aParams, aXslt, aOs);
61 * @see org.wamblee.crawler.PageRequest#getPage(org.apache.commons.httpclient.HttpClient)
63 public Document execute(String aUrl, HttpClient aClient)
64 throws PageException {
65 HttpMethod method = new GetMethod(aUrl);
66 if (getParameters().length > 0) {
67 String oldQueryString = method.getQueryString();
68 method.setQueryString(getParameters());
69 String queryString = method.getQueryString();
70 if (oldQueryString.length() > 0) {
71 queryString = queryString + '&' + oldQueryString;
72 method.setQueryString(queryString);
76 return executeMethod(aClient, method);
77 } catch (TransformerException e) {
78 throw new PageException(e.getMessage(), e);
79 } catch (IOException e) {
80 throw new PageException(e.getMessage(), e);