1 | package org.greenstone.gsdl3.service;
|
---|
2 |
|
---|
3 | // Greenstone classes
|
---|
4 | import org.greenstone.gsdl3.core.GSException;
|
---|
5 | import org.greenstone.util.Misc;
|
---|
6 | import org.greenstone.gsdl3.util.*;
|
---|
7 |
|
---|
8 | // XML classes
|
---|
9 | import org.w3c.dom.Element;
|
---|
10 | import org.w3c.dom.Document;
|
---|
11 | import org.w3c.dom.NodeList;
|
---|
12 |
|
---|
13 | import java.util.HashMap;
|
---|
14 | import java.util.ArrayList;
|
---|
15 | import java.io.File;
|
---|
16 | import java.io.InputStream;
|
---|
17 | import java.io.BufferedReader;
|
---|
18 | import java.io.InputStreamReader;
|
---|
19 | import java.io.IOException;
|
---|
20 | import java.net.HttpURLConnection;
|
---|
21 | import java.net.URLConnection;
|
---|
22 | import java.net.URL;
|
---|
23 | import java.net.Authenticator;
|
---|
24 | import java.net.MalformedURLException;
|
---|
25 |
|
---|
26 | import org.apache.log4j.*;
|
---|
27 |
|
---|
28 | /**
|
---|
29 | *
|
---|
30 | * @author <a href="mailto:[email protected]">Katherine Don</a>
|
---|
31 | * @version $Revision: 22085 $
|
---|
32 | * Modified by <a href="mailto:[email protected]">Chi-Yu Huang</a>
|
---|
33 | */
|
---|
34 |
|
---|
35 | public class IViaRetrieve
|
---|
36 | extends AbstractDocumentRetrieve {
|
---|
37 |
|
---|
38 | static Logger logger = Logger.getLogger(org.greenstone.gsdl3.service.IViaRetrieve.class.getName());
|
---|
39 |
|
---|
40 |
|
---|
41 | protected String ivia_server_url = null;
|
---|
42 |
|
---|
43 | public IViaRetrieve() {
|
---|
44 | does_structure = false;
|
---|
45 | }
|
---|
46 |
|
---|
47 | //Configure IViaRetrieve Service
|
---|
48 | public boolean configure(Element info, Element extra_info)
|
---|
49 | {
|
---|
50 | if (!super.configure(info, extra_info)) {
|
---|
51 | return false;
|
---|
52 | }
|
---|
53 |
|
---|
54 | Element server_elem = (Element)GSXML.getChildByTagName(info, "iViaServer");
|
---|
55 | if (server_elem == null) {
|
---|
56 | logger.error("no iViaServer element found");
|
---|
57 | return false;
|
---|
58 | }
|
---|
59 | ivia_server_url = server_elem.getAttribute("url");
|
---|
60 | if (ivia_server_url.equals("")) {
|
---|
61 | logger.error("no url for the iViaServer element");
|
---|
62 | return false;
|
---|
63 | }
|
---|
64 | return true;
|
---|
65 |
|
---|
66 | }
|
---|
67 |
|
---|
68 | /** gets a document by sending a request to iVia, then processes it and creates a documentNode around the text */
|
---|
69 | protected Element getNodeContent(String doc_id, String lang)
|
---|
70 | throws GSException {
|
---|
71 |
|
---|
72 | String url_string = ivia_server_url+"/cgi-bin/view_record?theme=gsdl3&record_id="+doc_id;
|
---|
73 |
|
---|
74 | StringBuffer buffer = new StringBuffer();
|
---|
75 | try {
|
---|
76 | BufferedReader reader = Misc.makeHttpConnection(url_string);
|
---|
77 | String line;
|
---|
78 | while((line = reader.readLine())!= null) {
|
---|
79 | buffer.append(line);
|
---|
80 | }
|
---|
81 | } catch (java.net.MalformedURLException e) {
|
---|
82 | throw new GSException("Malformed URL: "+url_string, GSXML.ERROR_TYPE_SYSTEM);
|
---|
83 | } catch (java.io.IOException e) {
|
---|
84 | throw new GSException("IOException during connection to "+url_string+": "+e.toString(),GSXML.ERROR_TYPE_SYSTEM);
|
---|
85 | }
|
---|
86 |
|
---|
87 | String node_content = buffer.toString();
|
---|
88 |
|
---|
89 | String escaped_content = GSXML.xmlSafe(node_content);
|
---|
90 |
|
---|
91 | StringBuffer processed_content = new StringBuffer(escaped_content.length());
|
---|
92 | processed_content.append("<nodeContent>");
|
---|
93 | int pos = 0;
|
---|
94 | int lastpos = 0;
|
---|
95 | while ((pos = escaped_content.indexOf("<a ", lastpos))!= -1) {
|
---|
96 | processed_content.append(escaped_content.substring(lastpos, pos));
|
---|
97 | int endpos = escaped_content.indexOf("</a>", pos);
|
---|
98 | if (endpos == -1) {
|
---|
99 | break;
|
---|
100 | }
|
---|
101 | String link = escaped_content.substring(pos, endpos+10);
|
---|
102 | link = convertLink(link);
|
---|
103 | processed_content.append(link);
|
---|
104 | lastpos = endpos+10;
|
---|
105 | }
|
---|
106 | processed_content.append(escaped_content.substring(lastpos)); // get the last bit
|
---|
107 | processed_content.append("</nodeContent>");
|
---|
108 |
|
---|
109 | Document content_doc = this.converter.getDOM(processed_content.toString());
|
---|
110 | if (content_doc == null) {
|
---|
111 | logger.error("Couldn't parse node content");
|
---|
112 | logger.error(processed_content.toString());
|
---|
113 | return null;
|
---|
114 | }
|
---|
115 |
|
---|
116 | Element content_element = content_doc.getDocumentElement();
|
---|
117 |
|
---|
118 | return (Element)this.doc.importNode(content_element,true);
|
---|
119 | }
|
---|
120 |
|
---|
121 | /** converts a url from an <a> element into a greenstone suitable one */
|
---|
122 | protected String convertLink(String aref) {
|
---|
123 | if (aref.indexOf("href="http") != -1) {
|
---|
124 | return aref; // an external link
|
---|
125 | }
|
---|
126 | String type = "other";
|
---|
127 | if (aref.indexOf("/cgi-bin/canned_search")!=-1) {
|
---|
128 | type="query";
|
---|
129 | } else if (aref.indexOf("/cgi-bin/click_through") != -1) {
|
---|
130 | type = "external";
|
---|
131 | } else if (aref.indexOf("/cgi-bin/view_record") != -1) {
|
---|
132 | type="document";
|
---|
133 | }
|
---|
134 |
|
---|
135 | int href_start = aref.indexOf("href="")+11;
|
---|
136 | int href_end = aref.indexOf(">", href_start);
|
---|
137 | String href = aref.substring(href_start, href_end);
|
---|
138 | String link_content = aref.substring(href_end+4, aref.length()-10);
|
---|
139 |
|
---|
140 | if (type.equals("external")) {
|
---|
141 | // the external link is everything after the http at the end.
|
---|
142 | String address = href.substring(href.lastIndexOf("http"));
|
---|
143 | address = address.replaceAll("%3[aA]", ":");
|
---|
144 | address = address.replaceAll("%2[fF]", "/");
|
---|
145 |
|
---|
146 | return "<a href=\""+address+"\">"+link_content+"</a>";
|
---|
147 | }
|
---|
148 | if (type.equals("other")) {
|
---|
149 | return "other type of link ("+link_content+")";
|
---|
150 | }
|
---|
151 | StringBuffer result = new StringBuffer();
|
---|
152 | result.append("<link type='");
|
---|
153 | result.append(type);
|
---|
154 | result.append("'");
|
---|
155 | if (type.equals("query")) {
|
---|
156 | result.append(" service='TextQuery'");
|
---|
157 | }
|
---|
158 | result.append(">");
|
---|
159 | // add in the parameters
|
---|
160 | href = href.substring(href.indexOf("?")+1);
|
---|
161 | String [] params = href.split("&");
|
---|
162 | for (int i=0; i<params.length; i++) {
|
---|
163 | String param = params[i];
|
---|
164 | int eq_pos = param.indexOf("=");
|
---|
165 | if (eq_pos != -1) {
|
---|
166 |
|
---|
167 | result.append("<param name='"+param.substring(0, eq_pos)+"' value='"+param.substring(eq_pos+1)+"'/>");
|
---|
168 | }
|
---|
169 | }
|
---|
170 | result.append(link_content);
|
---|
171 | result.append("</link>");
|
---|
172 |
|
---|
173 | return result.toString();
|
---|
174 | }
|
---|
175 |
|
---|
176 | // iVia craps out if we ask for a metadata which is not valid. So need
|
---|
177 | // to make sure we only ask for acceptable fields.
|
---|
178 | protected boolean isAcceptableMetadata(String meta) {
|
---|
179 | String valid_metadata = ",title,url,ivia_description,keywords,subjects,";
|
---|
180 | if (valid_metadata.indexOf(","+meta+",")!=-1) {
|
---|
181 | return true;
|
---|
182 | }
|
---|
183 | return false;
|
---|
184 | }
|
---|
185 |
|
---|
186 | protected String translateId(String oid){
|
---|
187 | int p = oid.lastIndexOf('.');
|
---|
188 | if (p != oid.length()-3) {
|
---|
189 | logger.info("translateoid error: '.' is not the third to last char!!");
|
---|
190 | return oid;
|
---|
191 | }
|
---|
192 | String top = oid.substring(0, p);
|
---|
193 | return top;
|
---|
194 | }
|
---|
195 |
|
---|
196 | protected String translateExternalId(String id){
|
---|
197 | return id;
|
---|
198 | }
|
---|
199 |
|
---|
200 | protected String getDocType(String node_id){
|
---|
201 | return GSXML.DOC_TYPE_SIMPLE;
|
---|
202 | }
|
---|
203 | protected String getRootId(String node_id){
|
---|
204 | return node_id;
|
---|
205 | }
|
---|
206 |
|
---|
207 | protected ArrayList getChildrenIds(String node_id){
|
---|
208 | return null;
|
---|
209 | }
|
---|
210 |
|
---|
211 | protected String getParentId(String node_id){
|
---|
212 | return null;
|
---|
213 | }
|
---|
214 |
|
---|
215 | protected Element getMetadataList (String doc_id,
|
---|
216 | boolean all_metadata,
|
---|
217 | ArrayList metadata_names)
|
---|
218 | throws GSException {
|
---|
219 |
|
---|
220 | Element meta_list = this.doc.createElement(GSXML.METADATA_ELEM+GSXML.LIST_MODIFIER);
|
---|
221 |
|
---|
222 | // do the query to the iVia server
|
---|
223 | StringBuffer field_list= new StringBuffer();
|
---|
224 | boolean metadata_found = false;
|
---|
225 |
|
---|
226 | for (int i=0; i<metadata_names.size();i++){
|
---|
227 | if (isAcceptableMetadata((String)metadata_names.get(i))){
|
---|
228 | metadata_found = true;
|
---|
229 | field_list.append((String)metadata_names.get(i));
|
---|
230 | field_list.append(",");
|
---|
231 | }
|
---|
232 | }
|
---|
233 | if (!metadata_found){
|
---|
234 | return meta_list;
|
---|
235 | }
|
---|
236 |
|
---|
237 | String url_string = ivia_server_url+"/cgi-bin/view_record_set?theme=gsdl3&record_id_list="+doc_id+"&field_list="+field_list.toString();
|
---|
238 | try {
|
---|
239 | BufferedReader reader = Misc.makeHttpConnection(url_string);
|
---|
240 | String line;
|
---|
241 | while ((line = reader.readLine()) != null) {
|
---|
242 | //metadata entry
|
---|
243 | int col_pos = line.indexOf(':');
|
---|
244 | if (col_pos == -1) {
|
---|
245 | // end of the metadata for this doc
|
---|
246 | break;
|
---|
247 | }
|
---|
248 | String name = line.substring(0,col_pos);
|
---|
249 | String value = line.substring(col_pos+2); // includes a space
|
---|
250 | GSXML.addMetadata(this.doc, meta_list, name, value);
|
---|
251 | }
|
---|
252 | } catch (java.net.MalformedURLException e) {
|
---|
253 | throw new GSException("Malformed URL: "+url_string, GSXML.ERROR_TYPE_SYSTEM);
|
---|
254 | } catch (java.io.IOException e) {
|
---|
255 | throw new GSException("IOException: "+e.toString(), GSXML.ERROR_TYPE_SYSTEM);
|
---|
256 | }
|
---|
257 | return meta_list;
|
---|
258 | }
|
---|
259 |
|
---|
260 | protected String getStructureInfo(String doc_id, String info_type){
|
---|
261 | return "";
|
---|
262 | }
|
---|
263 | }
|
---|