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