1 | /*
|
---|
2 | * ServiceRack.java
|
---|
3 | * Copyright (C) 2002 New Zealand Digital Library, http://www.nzdl.org
|
---|
4 | *
|
---|
5 | * This program is free software; you can redistribute it and/or modify
|
---|
6 | * it under the terms of the GNU General Public License as published by
|
---|
7 | * the Free Software Foundation; either version 2 of the License, or
|
---|
8 | * (at your option) any later version.
|
---|
9 | *
|
---|
10 | * This program is distributed in the hope that it will be useful,
|
---|
11 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
12 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
13 | * GNU General Public License for more details.
|
---|
14 | *
|
---|
15 | * You should have received a copy of the GNU General Public License
|
---|
16 | * along with this program; if not, write to the Free Software
|
---|
17 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
18 | */
|
---|
19 | package org.greenstone.gsdl3.service;
|
---|
20 |
|
---|
21 | // greenstone classes
|
---|
22 | import java.io.StringReader;
|
---|
23 | import java.util.HashMap;
|
---|
24 | import java.util.Properties;
|
---|
25 | import java.util.regex.Matcher;
|
---|
26 | import java.util.regex.Pattern;
|
---|
27 |
|
---|
28 | import javax.xml.parsers.DocumentBuilder;
|
---|
29 | import javax.xml.parsers.DocumentBuilderFactory;
|
---|
30 |
|
---|
31 | import org.apache.log4j.Logger;
|
---|
32 | import org.greenstone.gs3client.dlservices.DigitalLibraryServicesAPIA;
|
---|
33 | import org.greenstone.gs3client.dlservices.FedoraServicesAPIA;
|
---|
34 | import org.greenstone.gsdl3.core.MessageRouter;
|
---|
35 | import org.greenstone.gsdl3.util.Dictionary;
|
---|
36 | import org.greenstone.gsdl3.util.GSPath;
|
---|
37 | import org.greenstone.gsdl3.util.GSXML;
|
---|
38 | import org.greenstone.gsdl3.util.MacroResolver;
|
---|
39 | import org.greenstone.gsdl3.util.OID;
|
---|
40 | import org.greenstone.gsdl3.util.XMLConverter;
|
---|
41 | import org.w3c.dom.Document;
|
---|
42 | import org.w3c.dom.Element;
|
---|
43 | import org.w3c.dom.Node;
|
---|
44 | import org.w3c.dom.NodeList;
|
---|
45 | import org.xml.sax.InputSource;
|
---|
46 |
|
---|
47 | import org.apache.commons.lang3.StringUtils;
|
---|
48 |
|
---|
49 | /*
|
---|
50 | // greenstone classes
|
---|
51 | import org.greenstone.gsdl3.util.*;
|
---|
52 | import org.greenstone.gsdl3.core.*;
|
---|
53 |
|
---|
54 | // for fedora
|
---|
55 | import org.greenstone.gs3client.dlservices.*;
|
---|
56 | import org.greenstone.fedora.services.FedoraGS3Exception.CancelledException;
|
---|
57 |
|
---|
58 | // xml classes
|
---|
59 | import org.w3c.dom.Node;
|
---|
60 | import org.w3c.dom.NodeList;
|
---|
61 | import org.w3c.dom.Element;
|
---|
62 | import org.w3c.dom.Document;
|
---|
63 | import org.xml.sax.InputSource;
|
---|
64 | import javax.xml.parsers.*;
|
---|
65 | import org.apache.xpath.XPathAPI;
|
---|
66 |
|
---|
67 | // general java classes
|
---|
68 | import java.io.Reader;
|
---|
69 | import java.io.StringReader;
|
---|
70 | import java.io.File;
|
---|
71 | import java.util.HashMap;
|
---|
72 | import java.util.Locale;
|
---|
73 | import java.util.Properties;
|
---|
74 | import java.util.ResourceBundle;
|
---|
75 | import java.util.regex.*;
|
---|
76 | import java.lang.reflect.Method;
|
---|
77 | */
|
---|
78 |
|
---|
79 | import org.apache.log4j.*;
|
---|
80 |
|
---|
81 | /**
|
---|
82 | * FedoraServiceProxy - communicates with the FedoraGS3 interface.
|
---|
83 | *
|
---|
84 | * @author Anupama Krishnan
|
---|
85 | */
|
---|
86 | public class FedoraServiceProxy
|
---|
87 | extends ServiceRack implements OID.OIDTranslatable
|
---|
88 | {
|
---|
89 |
|
---|
90 | static Logger logger = Logger.getLogger(org.greenstone.gsdl3.service.FedoraServiceProxy.class.getName());
|
---|
91 | protected MacroResolver macro_resolver = null;
|
---|
92 |
|
---|
93 |
|
---|
94 | /** The handle to the fedora connection */
|
---|
95 | private DigitalLibraryServicesAPIA fedoraServicesAPIA;
|
---|
96 |
|
---|
97 | private String prevLanguage = "";
|
---|
98 |
|
---|
99 | public void cleanUp() {
|
---|
100 | super.cleanUp();
|
---|
101 | }
|
---|
102 |
|
---|
103 | /** sets the message router */
|
---|
104 | public void setMessageRouter(MessageRouter m) {
|
---|
105 | this.router = m;
|
---|
106 | setLibraryName(m.getLibraryName());
|
---|
107 | }
|
---|
108 |
|
---|
109 | /** the no-args constructor */
|
---|
110 | public FedoraServiceProxy() {
|
---|
111 | super();
|
---|
112 |
|
---|
113 | this.macro_resolver = new BasicTextMacroResolver();
|
---|
114 | }
|
---|
115 |
|
---|
116 |
|
---|
117 | /* configure the service module
|
---|
118 | *
|
---|
119 | * @param info the XML node <serviceRack name="XXX"/> with name equal
|
---|
120 | * to the class name (of the subclass)
|
---|
121 | *
|
---|
122 | * must configure short_service_info_ and service_info_map_
|
---|
123 | * @return true if configured ok
|
---|
124 | * must be implemented in subclasses
|
---|
125 | */
|
---|
126 | /*public boolean configure(Element info) {
|
---|
127 | return configure(info, null);
|
---|
128 | }*/
|
---|
129 |
|
---|
130 | public boolean configure(Element info, Element extra_info) {
|
---|
131 | // set up the class loader
|
---|
132 |
|
---|
133 | if (!super.configure(info, extra_info)){
|
---|
134 | return false;
|
---|
135 | }
|
---|
136 |
|
---|
137 | // Try to instantiate a Fedora dl handle
|
---|
138 | try {
|
---|
139 | // Fedora connection settings defaults.
|
---|
140 | // Read host and port from global.properties, since by default, we expect the Greenstone server to be used
|
---|
141 | Properties globalProperties = new Properties();
|
---|
142 | globalProperties.load(Class.forName("org.greenstone.util.GlobalProperties").getClassLoader().getResourceAsStream("global.properties"));
|
---|
143 | String host = globalProperties.getProperty("tomcat.server", "localhost");
|
---|
144 | String port = globalProperties.getProperty("tomcat.port", "8383");
|
---|
145 | String protocol = "http";
|
---|
146 | String username = "fedoraIntCallUser"; //"fedoraAdmin"
|
---|
147 | String password = "changeme"; //"<user password>"
|
---|
148 |
|
---|
149 | // See if buildConfig.xml overrides any of the defaults
|
---|
150 | // info is the <serviceRack> Element from buildConfig.xml (extra_info are the Elements of collectionConfig.xml)
|
---|
151 |
|
---|
152 | NodeList nodes = info.getElementsByTagName("fedoraConnection");
|
---|
153 | if(nodes != null && nodes.getLength() > 0) {
|
---|
154 |
|
---|
155 | Element fedoraElement = (Element)nodes.item(0);
|
---|
156 | if(fedoraElement.hasAttribute("protocol")) {
|
---|
157 | protocol = fedoraElement.getAttribute("protocol");
|
---|
158 | }
|
---|
159 | if(fedoraElement.hasAttribute("host")) {
|
---|
160 | host = fedoraElement.getAttribute("host");
|
---|
161 | }
|
---|
162 | if(fedoraElement.hasAttribute("port")) {
|
---|
163 | port = fedoraElement.getAttribute("port");
|
---|
164 | }
|
---|
165 | if(fedoraElement.hasAttribute("username")) {
|
---|
166 | username = fedoraElement.getAttribute("username");
|
---|
167 | }
|
---|
168 | if(fedoraElement.hasAttribute("password")) {
|
---|
169 | password = fedoraElement.getAttribute("password");
|
---|
170 | }
|
---|
171 | }
|
---|
172 |
|
---|
173 | fedoraServicesAPIA = new FedoraServicesAPIA(protocol, host, Integer.parseInt(port), username, password);
|
---|
174 |
|
---|
175 | } catch(org.greenstone.fedora.services.FedoraGS3Exception.CancelledException e) {
|
---|
176 | // The user pressed cancel in the fedora services instantiation dialog
|
---|
177 | return false;
|
---|
178 | } catch(Exception e) {
|
---|
179 | logger.error("Error instantiating the interface to the Fedora Repository:\n", e); // second parameter prints e's stacktrace
|
---|
180 | return false;
|
---|
181 | }
|
---|
182 |
|
---|
183 |
|
---|
184 | // Need to put the available services into short_service_info
|
---|
185 | // This is used by DefaultReceptionist.process() has an exception. But DefaultReceptionist.addExtraInfo()
|
---|
186 | // isn't helpful, and the problem actually already occurs in
|
---|
187 | // Receptionist.process() -> PageAction.process() -> MessageRouter.process()
|
---|
188 | // -> Collection/ServiceCluster.process() -> ServiceCluster.configureServiceRackList()
|
---|
189 | // -> ServiceRack.process() -> ServiceRack.processDescribe() -> ServiceRack.getServiceList().
|
---|
190 | // ServiceRack.getServiceList() requires this ServiceRack's services to be filled into the
|
---|
191 | // short_service_info Element which needs to be done in this FedoraServiceProxy.configure().
|
---|
192 |
|
---|
193 | // get the display and format elements from the coll config file for
|
---|
194 | // the classifiers
|
---|
195 | AbstractBrowse.extractExtraClassifierInfo(info, extra_info);
|
---|
196 |
|
---|
197 | // Copied from IViaProxy.java:
|
---|
198 | String collection = fedoraServicesAPIA.describeCollection(this.cluster_name);
|
---|
199 |
|
---|
200 | Element collNode = getResponseAsDOM(collection);
|
---|
201 | Element serviceList = (Element)collNode.getElementsByTagName(GSXML.SERVICE_ELEM+GSXML.LIST_MODIFIER).item(0);
|
---|
202 |
|
---|
203 | //this.short_service_info.appendChild(short_service_info.getOwnerDocument().importNode(serviceList, true));
|
---|
204 | // we want the individual service Elements, not the serviceList Element which will wrap it later
|
---|
205 | NodeList services = collNode.getElementsByTagName(GSXML.SERVICE_ELEM);
|
---|
206 | for(int i = 0; i < services.getLength(); i++) {
|
---|
207 | Node service = services.item(i);
|
---|
208 | this.short_service_info.appendChild(short_service_info.getOwnerDocument().importNode(service, true));
|
---|
209 | }
|
---|
210 |
|
---|
211 | // add some format info to service map if there is any
|
---|
212 | String path = GSPath.appendLink(GSXML.SEARCH_ELEM, GSXML.FORMAT_ELEM);
|
---|
213 | Element search_format = (Element) GSXML.getNodeByPath(extra_info, path);
|
---|
214 | if (search_format != null) {
|
---|
215 | this.format_info_map.put("TextQuery", this.desc_doc.importNode(search_format, true));
|
---|
216 | this.format_info_map.put("FieldQuery", this.desc_doc.importNode(search_format, true));
|
---|
217 | }
|
---|
218 |
|
---|
219 | // look for document display format
|
---|
220 | path = GSPath.appendLink(GSXML.DISPLAY_ELEM, GSXML.FORMAT_ELEM);
|
---|
221 | Element display_format = (Element)GSXML.getNodeByPath(extra_info, path);
|
---|
222 | if (display_format != null) {
|
---|
223 | this.format_info_map.put("DocumentContentRetrieve", this.desc_doc.importNode(display_format, true));
|
---|
224 | // should we make a copy?
|
---|
225 | }
|
---|
226 |
|
---|
227 | // the format info
|
---|
228 | Element cb_format_info = this.desc_doc.createElement(GSXML.FORMAT_ELEM);
|
---|
229 | boolean format_found = false;
|
---|
230 |
|
---|
231 | // look for classifier <browse><format>
|
---|
232 | path = GSPath.appendLink(GSXML.BROWSE_ELEM, GSXML.FORMAT_ELEM);
|
---|
233 | Element browse_format = (Element)GSXML.getNodeByPath(extra_info, path);
|
---|
234 | if (browse_format != null) {
|
---|
235 | cb_format_info.appendChild(GSXML.duplicateWithNewName(this.desc_doc, browse_format, GSXML.DEFAULT_ELEM, true));
|
---|
236 | format_found = true;
|
---|
237 | }
|
---|
238 |
|
---|
239 | // add in to the description a simplified list of classifiers
|
---|
240 | Element browse = (Element)GSXML.getChildByTagName(extra_info, "browse"); // the <browse>
|
---|
241 | NodeList classifiers = browse.getElementsByTagName(GSXML.CLASSIFIER_ELEM);
|
---|
242 | for(int i=0; i<classifiers.getLength(); i++) {
|
---|
243 | Element cl = (Element)classifiers.item(i);
|
---|
244 | Element new_cl = (Element)this.desc_doc.importNode(cl, false); // just import this node, not the children
|
---|
245 |
|
---|
246 | // get the format info out, and put inside a classifier element
|
---|
247 | Element format_cl = (Element)new_cl.cloneNode(false);
|
---|
248 | Element format = (Element)GSXML.getChildByTagName(cl, GSXML.FORMAT_ELEM);
|
---|
249 | if (format != null) {
|
---|
250 |
|
---|
251 | //copy all the children
|
---|
252 | NodeList elems = format.getChildNodes();
|
---|
253 | for (int j=0; j<elems.getLength();j++) {
|
---|
254 | format_cl.appendChild(this.desc_doc.importNode(elems.item(j), true));
|
---|
255 | }
|
---|
256 | cb_format_info.appendChild(format_cl);
|
---|
257 | format_found = true;
|
---|
258 | }
|
---|
259 |
|
---|
260 | }
|
---|
261 |
|
---|
262 | if (format_found) {
|
---|
263 | this.format_info_map.put("ClassifierBrowse", cb_format_info);
|
---|
264 | }
|
---|
265 |
|
---|
266 |
|
---|
267 | // set up the macro resolver
|
---|
268 | macro_resolver.setSiteDetails(this.site_http_address, this.cluster_name, this.getLibraryName());
|
---|
269 | Element replacement_elem = (Element)GSXML.getChildByTagName(extra_info, "replaceList");
|
---|
270 | if (replacement_elem != null) {
|
---|
271 | macro_resolver.addMacros(replacement_elem);
|
---|
272 | }
|
---|
273 | // look for any refs to global replace lists
|
---|
274 | NodeList replace_refs_elems = extra_info.getElementsByTagName("replaceListRef");
|
---|
275 | for (int i=0; i<replace_refs_elems.getLength(); i++) {
|
---|
276 | String id = ((Element)replace_refs_elems.item(i)).getAttribute("id");
|
---|
277 | if (!id.equals("")) {
|
---|
278 | Element replace_list = GSXML.getNamedElement(this.router.config_info, "replaceList", "id", id);
|
---|
279 | if (replace_list != null) {
|
---|
280 | macro_resolver.addMacros(replace_list);
|
---|
281 | }
|
---|
282 | }
|
---|
283 | }
|
---|
284 |
|
---|
285 | // configured ok
|
---|
286 | return true;
|
---|
287 | }
|
---|
288 |
|
---|
289 |
|
---|
290 | /* "DocumentContentRetrieve", "DocumentMetadataRetrieve", "DocumentStructureRetrieve",
|
---|
291 | "TextQuery", "FieldQuery", "ClassifierBrowse", "ClassifierBrowseMetadataRetrieve" */
|
---|
292 |
|
---|
293 | protected Element processDocumentContentRetrieve(Element request) {
|
---|
294 | String[] docIDs = parse(request, GSXML.DOC_NODE_ELEM, GSXML.NODE_ID_ATT);
|
---|
295 | String[] relLinks = parse(request, GSXML.DOC_NODE_ELEM, "externalURL");
|
---|
296 |
|
---|
297 | //logger.error("### request:");
|
---|
298 | //logger.error(GSXML.elementToString(request, true));
|
---|
299 |
|
---|
300 | if(docIDs == null) {
|
---|
301 | logger.error("DocumentContentRetrieve request specified no doc nodes.\n");
|
---|
302 | return XMLConverter.newDOM().createElement(GSXML.RESPONSE_ELEM); // empty response
|
---|
303 | } else {
|
---|
304 | for(int i = 0; i < docIDs.length; i++) {
|
---|
305 | //logger.error("BEFORE: docIDs[" + i + "]: " + docIDs[i]);
|
---|
306 | if(relLinks[i] != null && docIDs[i].startsWith("http://")) { // need to do a look up
|
---|
307 | docIDs[i] = translateExternalId(docIDs[i]);
|
---|
308 | } else {
|
---|
309 | docIDs[i] = translateId(docIDs[i]);
|
---|
310 | }
|
---|
311 | //logger.error("AFTER: docIDs[" + i + "]: " + docIDs[i]);
|
---|
312 | }
|
---|
313 | }
|
---|
314 |
|
---|
315 | String lang = request.getAttribute(GSXML.LANG_ATT);
|
---|
316 | if(!lang.equals(prevLanguage)) {
|
---|
317 | prevLanguage = lang;
|
---|
318 | fedoraServicesAPIA.setLanguage(lang);
|
---|
319 | }
|
---|
320 |
|
---|
321 | // first param (the collection) is not used by Fedora
|
---|
322 | Element response = getResponseAsDOM(fedoraServicesAPIA.retrieveDocumentContent(this.cluster_name, docIDs));
|
---|
323 |
|
---|
324 |
|
---|
325 | // resolve any collection specific macros
|
---|
326 | NodeList nodeContents = response.getElementsByTagName(GSXML.NODE_CONTENT_ELEM);
|
---|
327 | for(int i = 0; i < nodeContents.getLength(); i++) {
|
---|
328 | Element nodeContent = (Element)nodeContents.item(i);
|
---|
329 | /*if(nodeContent != null) {
|
---|
330 | nodeContent = (Element)nodeContent.getFirstChild(); // textNode
|
---|
331 | }*/
|
---|
332 | //logger.error("GIRAFFE 1. content retrieve response - nodeContent: " + GSXML.nodeToFormattedString(nodeContent));
|
---|
333 | String docContent = nodeContent.getFirstChild().getNodeValue(); // getTextNode and get its contents.
|
---|
334 | //logger.error("GIRAFFE 2. content retrieve response - docContent: " + docContent);
|
---|
335 |
|
---|
336 | if(docContent != null) {
|
---|
337 | // get document text and resolve and macros. Rel and external links have _httpextlink_ set by HTMLPlugin
|
---|
338 | docContent = macro_resolver.resolve(docContent, lang, MacroResolver.SCOPE_TEXT, ""); // doc_id
|
---|
339 | nodeContent.getFirstChild().setNodeValue(docContent);
|
---|
340 | //logger.error("GIRAFFE 3. content retrieve response. Updated docContent: " + docContent);
|
---|
341 | }
|
---|
342 | }
|
---|
343 |
|
---|
344 | return (Element)response.getElementsByTagName(GSXML.RESPONSE_ELEM).item(0);
|
---|
345 | }
|
---|
346 |
|
---|
347 | protected Element processDocumentStructureRetrieve(Element request) {
|
---|
348 | String[] docIDs = parse(request, GSXML.DOC_NODE_ELEM, GSXML.NODE_ID_ATT);
|
---|
349 | String[] relLinks = parse(request, GSXML.DOC_NODE_ELEM, "externalURL");
|
---|
350 |
|
---|
351 | if(docIDs == null) {
|
---|
352 | logger.error("DocumentStructureRetrieve request specified no doc nodes.\n");
|
---|
353 | return XMLConverter.newDOM().createElement(GSXML.RESPONSE_ELEM); // empty response
|
---|
354 | } else {
|
---|
355 | for(int i = 0; i < docIDs.length; i++) {
|
---|
356 | //logger.error("BEFORE: docIDs[" + i + "]: " + docIDs[i]);
|
---|
357 | if(relLinks[i] != null && docIDs[i].startsWith("http://")) { // need to do a look up
|
---|
358 | docIDs[i] = translateExternalId(docIDs[i]);
|
---|
359 | } else {
|
---|
360 | docIDs[i] = translateId(docIDs[i]);
|
---|
361 | }
|
---|
362 | }
|
---|
363 | }
|
---|
364 |
|
---|
365 | NodeList params = request.getElementsByTagName(GSXML.PARAM_ELEM);
|
---|
366 | String structure="";
|
---|
367 | String info="";
|
---|
368 | for(int i = 0; i < params.getLength(); i++) {
|
---|
369 | Element param = (Element)params.item(i);
|
---|
370 | if(param.getAttribute("name").equals("structure")) {
|
---|
371 | structure = structure + param.getAttribute("value") + "|";
|
---|
372 | } else if(param.getAttribute("name").equals("info")) {
|
---|
373 | info = info + param.getAttribute("value") + "|";
|
---|
374 | }
|
---|
375 | }
|
---|
376 |
|
---|
377 | String lang = request.getAttribute(GSXML.LANG_ATT);
|
---|
378 | if(!lang.equals(prevLanguage)) {
|
---|
379 | prevLanguage = lang;
|
---|
380 | fedoraServicesAPIA.setLanguage(lang);
|
---|
381 | }
|
---|
382 | Element response = getResponseAsDOM(fedoraServicesAPIA.retrieveDocumentStructure(
|
---|
383 | this.cluster_name, docIDs, new String[]{structure}, new String[]{info}));
|
---|
384 | return (Element)response.getElementsByTagName(GSXML.RESPONSE_ELEM).item(0);
|
---|
385 | }
|
---|
386 |
|
---|
387 | protected Element processDocumentMetadataRetrieve(Element request) {
|
---|
388 | String[] docIDs = parse(request, GSXML.DOC_NODE_ELEM, GSXML.NODE_ID_ATT);
|
---|
389 | String[] relLinks = parse(request, GSXML.DOC_NODE_ELEM, "externalURL");
|
---|
390 |
|
---|
391 | if(docIDs == null) {
|
---|
392 | logger.error("DocumentMetadataRetrieve request specified no doc nodes.\n");
|
---|
393 | return XMLConverter.newDOM().createElement(GSXML.RESPONSE_ELEM); // empty response
|
---|
394 | } else {
|
---|
395 | for(int i = 0; i < docIDs.length; i++) {
|
---|
396 | //logger.error("**** relLinks[i]: " + relLinks[i]);
|
---|
397 | //logger.error("**** docIDs[i]: " + docIDs[i]);
|
---|
398 | if(relLinks[i] != null && docIDs[i].startsWith("http://")) { // need to do a look up
|
---|
399 | docIDs[i] = translateExternalId(docIDs[i]);
|
---|
400 | } else {
|
---|
401 | docIDs[i] = translateId(docIDs[i]);
|
---|
402 | }
|
---|
403 | //logger.error("AFTER: docIDs[" + i + "]: " + docIDs[i]);
|
---|
404 | }
|
---|
405 | }
|
---|
406 |
|
---|
407 | NodeList params = request.getElementsByTagName(GSXML.PARAM_ELEM);
|
---|
408 | String[] metafields = {};
|
---|
409 | if(params.getLength() > 0) {
|
---|
410 | metafields = new String[params.getLength()];
|
---|
411 | for(int i = 0; i < metafields.length; i++) {
|
---|
412 | Element param = (Element)params.item(i);
|
---|
413 | //if(param.hasAttribute(GSXML.NAME_ATT) && param.getAttribute(GSXML.NAME_ATT).equals("metadata") && param.hasAttribute(GSXML.VALUE_ATT)) {
|
---|
414 | if(param.hasAttribute(GSXML.VALUE_ATT)){
|
---|
415 | metafields[i] = param.getAttribute(GSXML.VALUE_ATT);
|
---|
416 | } else {
|
---|
417 | metafields[i] = "";
|
---|
418 | }
|
---|
419 | }
|
---|
420 | }
|
---|
421 |
|
---|
422 | String lang = request.getAttribute(GSXML.LANG_ATT);
|
---|
423 | if(!lang.equals(prevLanguage)) {
|
---|
424 | prevLanguage = lang;
|
---|
425 | fedoraServicesAPIA.setLanguage(lang);
|
---|
426 | }
|
---|
427 | Element response = getResponseAsDOM(fedoraServicesAPIA.retrieveDocumentMetadata(
|
---|
428 | this.cluster_name, docIDs, metafields));
|
---|
429 | return (Element)response.getElementsByTagName(GSXML.RESPONSE_ELEM).item(0);
|
---|
430 | }
|
---|
431 |
|
---|
432 | protected Element processClassifierBrowseMetadataRetrieve(Element request) {
|
---|
433 | String[] classIDs = parse(request, GSXML.CLASS_NODE_ELEM, GSXML.NODE_ID_ATT);
|
---|
434 | //String[] relLinks = parse(request, GSXML.CLASS_NODE_ELEM, "externalURL");
|
---|
435 |
|
---|
436 | if(classIDs == null) {
|
---|
437 | logger.error("ClassifierBrowseMetadataRetrieve request specified no classifier nodes.\n");
|
---|
438 | return XMLConverter.newDOM().createElement(GSXML.RESPONSE_ELEM); // empty response
|
---|
439 | } else {
|
---|
440 | for(int i = 0; i < classIDs.length; i++) {
|
---|
441 | classIDs[i] = translateId(classIDs[i]);
|
---|
442 | }
|
---|
443 | }
|
---|
444 |
|
---|
445 | NodeList params = request.getElementsByTagName(GSXML.PARAM_ELEM);
|
---|
446 | String[] metafields = {};
|
---|
447 | if(params.getLength() > 0) {
|
---|
448 | metafields = new String[params.getLength()];
|
---|
449 | for(int i = 0; i < metafields.length; i++) {
|
---|
450 | Element param = (Element)params.item(i);
|
---|
451 | if(param.hasAttribute(GSXML.VALUE_ATT)){
|
---|
452 | metafields[i] = param.getAttribute(GSXML.VALUE_ATT);
|
---|
453 | } else {
|
---|
454 | metafields[i] = "";
|
---|
455 | }
|
---|
456 | }
|
---|
457 | }
|
---|
458 |
|
---|
459 | String lang = request.getAttribute(GSXML.LANG_ATT);
|
---|
460 | if(!lang.equals(prevLanguage)) {
|
---|
461 | prevLanguage = lang;
|
---|
462 | fedoraServicesAPIA.setLanguage(lang);
|
---|
463 | }
|
---|
464 | Element response = getResponseAsDOM(fedoraServicesAPIA.retrieveBrowseMetadata(
|
---|
465 | this.cluster_name, "ClassifierBrowseMetadataRetrieve", classIDs, metafields));
|
---|
466 | //logger.error("**** Response from retrieveBrowseMeta: " + GSXML.elementToString(response, true));
|
---|
467 | return (Element)response.getElementsByTagName(GSXML.RESPONSE_ELEM).item(0);
|
---|
468 | }
|
---|
469 |
|
---|
470 | protected Element processClassifierBrowse(Element request) {
|
---|
471 | String collection = this.cluster_name;
|
---|
472 | String lang = request.getAttribute(GSXML.LANG_ATT);
|
---|
473 | if(!lang.equals(prevLanguage)) {
|
---|
474 | prevLanguage = lang;
|
---|
475 | fedoraServicesAPIA.setLanguage(lang);
|
---|
476 | }
|
---|
477 |
|
---|
478 | NodeList classNodes = request.getElementsByTagName(GSXML.CLASS_NODE_ELEM);
|
---|
479 | if(classNodes == null || classNodes.getLength() <= 0) {
|
---|
480 | logger.error("ClassifierBrowse request specified no classifier IDs.\n");
|
---|
481 | return XMLConverter.newDOM().createElement(GSXML.RESPONSE_ELEM); // empty response
|
---|
482 | }
|
---|
483 | String classifierIDs[] = new String[classNodes.getLength()];
|
---|
484 | for(int i = 0; i < classifierIDs.length; i++) {
|
---|
485 | Element e = (Element)classNodes.item(i);
|
---|
486 | classifierIDs[i] = e.getAttribute(GSXML.NODE_ID_ATT);
|
---|
487 | classifierIDs[i] = translateId(classifierIDs[i]);
|
---|
488 | }
|
---|
489 |
|
---|
490 | NodeList params = request.getElementsByTagName(GSXML.PARAM_ELEM);
|
---|
491 | String structure="";
|
---|
492 | String info="";
|
---|
493 | for(int i = 0; i < params.getLength(); i++) {
|
---|
494 | Element param = (Element)params.item(i);
|
---|
495 | if(param.getAttribute("name").equals("structure")) {
|
---|
496 | structure = structure + param.getAttribute("value") + "|";
|
---|
497 | } else if(param.getAttribute("name").equals("info")) {
|
---|
498 | info = info + param.getAttribute("value") + "|";
|
---|
499 | }
|
---|
500 | }
|
---|
501 | ///structure = structure + "siblings"; //test for getting with classifier browse structure: siblings
|
---|
502 |
|
---|
503 | Element response
|
---|
504 | = getResponseAsDOM(fedoraServicesAPIA.retrieveBrowseStructure(collection, "ClassifierBrowse", classifierIDs,
|
---|
505 | new String[] {structure}, new String[] {info}));
|
---|
506 | //logger.error("**** FedoraServiceProxy - Response from retrieveBrowseStructure: " + GSXML.elementToString(response, true));
|
---|
507 |
|
---|
508 | return (Element)response.getElementsByTagName(GSXML.RESPONSE_ELEM).item(0);
|
---|
509 | }
|
---|
510 |
|
---|
511 | protected Element processTextQuery(Element request) {
|
---|
512 | return processQuery(request, "TextQuery");
|
---|
513 | }
|
---|
514 |
|
---|
515 | protected Element processFieldQuery(Element request) {
|
---|
516 | return processQuery(request, "FieldQuery");
|
---|
517 | }
|
---|
518 |
|
---|
519 | protected Element processQuery(Element request, String querytype) {
|
---|
520 | String collection = this.cluster_name;
|
---|
521 |
|
---|
522 | String lang = request.getAttribute(GSXML.LANG_ATT);
|
---|
523 | if(!lang.equals(prevLanguage)) {
|
---|
524 | prevLanguage = lang;
|
---|
525 | fedoraServicesAPIA.setLanguage(lang);
|
---|
526 | }
|
---|
527 |
|
---|
528 | NodeList paramNodes = request.getElementsByTagName(GSXML.PARAM_ELEM);
|
---|
529 | if(paramNodes.getLength() > 0) {
|
---|
530 | HashMap<String, String> params = new HashMap<String, String>(paramNodes.getLength());
|
---|
531 | for(int i = 0; i < paramNodes.getLength(); i++) {
|
---|
532 | Element param = (Element)paramNodes.item(i);
|
---|
533 | params.put(param.getAttribute(GSXML.NAME_ATT), param.getAttribute(GSXML.VALUE_ATT));
|
---|
534 | }
|
---|
535 |
|
---|
536 | Element response = getResponseAsDOM(fedoraServicesAPIA.query(collection, querytype, params));
|
---|
537 | return (Element)response.getElementsByTagName(GSXML.RESPONSE_ELEM).item(0);
|
---|
538 | } else {
|
---|
539 | logger.error("TextQuery request specified no parameters.\n");
|
---|
540 | return XMLConverter.newDOM().createElement(GSXML.RESPONSE_ELEM); // empty response
|
---|
541 | }
|
---|
542 | }
|
---|
543 |
|
---|
544 | // get the requested nodeIDs out of a request message
|
---|
545 | protected String[] parse(Element request, String nodeType, String attribute) {
|
---|
546 | String[] nodevalues = null;
|
---|
547 | int count = 0;
|
---|
548 |
|
---|
549 | Element docList = (Element) GSXML.getChildByTagName(request, nodeType+GSXML.LIST_MODIFIER);
|
---|
550 | if (docList != null) {
|
---|
551 | NodeList docNodes = docList.getElementsByTagName(nodeType);
|
---|
552 | if(docNodes.getLength() > 0) {
|
---|
553 | nodevalues = new String[docNodes.getLength()];
|
---|
554 | for(int i = 0; i < nodevalues.length; i++) {
|
---|
555 | Element e = (Element)docNodes.item(i);
|
---|
556 | String id = e.getAttribute(attribute);
|
---|
557 | // Not sure why there are at times requests for hashXXX.dir, which is not a fedora PID
|
---|
558 | // To skip these: if not requesting an externalURL and if requesting a docNode,
|
---|
559 | // then the ID has to contain the : character special to fedora PIDs
|
---|
560 | if(attribute == "externalURL" || (nodeType != GSXML.DOC_NODE_ELEM || id.contains(":"))) {
|
---|
561 | nodevalues[count++] = id;
|
---|
562 | }
|
---|
563 | }
|
---|
564 | }
|
---|
565 | }
|
---|
566 |
|
---|
567 | if(count == 0) {
|
---|
568 | return null;
|
---|
569 | }
|
---|
570 |
|
---|
571 | String[] tmp = new String[count];
|
---|
572 | for(int i = 0; i < count; i++) {
|
---|
573 | tmp[i] = nodevalues[i];
|
---|
574 | }
|
---|
575 | nodevalues = null;
|
---|
576 | nodevalues = tmp;
|
---|
577 |
|
---|
578 | return nodevalues;
|
---|
579 | }
|
---|
580 |
|
---|
581 |
|
---|
582 | /** if id ends in .fc, .pc etc, then translate it to the correct id
|
---|
583 | * For now (for testing things work) the default implementation is to just remove the suffix */
|
---|
584 | protected String translateId(String id) {
|
---|
585 | if (OID.needsTranslating(id)) {
|
---|
586 | return OID.translateOID(this, id); //return translateOID(id);
|
---|
587 | }
|
---|
588 | return id;
|
---|
589 | }
|
---|
590 |
|
---|
591 | /** if an id is not a greenstone id (an external id) then translate
|
---|
592 | * it to a greenstone one
|
---|
593 | * default implementation: return the id. Custom implementation:
|
---|
594 | * the id is a url that maps to a fedorapid whose dc.title contains the required HASHID */
|
---|
595 | protected String translateExternalId(String id) {
|
---|
596 | //logger.error("*** to translate an external ID: " + id); /////return id;
|
---|
597 | return this.externalId2OID(id);
|
---|
598 | }
|
---|
599 |
|
---|
600 | /** converts an external id to greenstone OID. External ID is a URL link
|
---|
601 | * that, if relative, maps to a fedorapid that has an entry in fedora.
|
---|
602 | * The dc:title meta for that fedorapid will contain the required OID. */
|
---|
603 | public String externalId2OID(String extid) {
|
---|
604 | if(extid.endsWith(".rt") && (extid.indexOf('.') != extid.lastIndexOf('.'))) {
|
---|
605 | // .rt is not file extension, but Greenstone request for root of document
|
---|
606 | // not relevant for external ID
|
---|
607 | extid = extid.substring(0, extid.length()-3);
|
---|
608 | }
|
---|
609 |
|
---|
610 | // the following method is unique to FedoraServicesAPIA
|
---|
611 | String response = ((FedoraServicesAPIA)fedoraServicesAPIA).getDocIDforURL(extid, this.cluster_name);
|
---|
612 | if(response.indexOf(GSXML.ERROR_ELEM) != -1) {
|
---|
613 | logger.error("**** The following error occurred when trying to find externalID for ID " + extid);
|
---|
614 | logger.error(response);
|
---|
615 | return extid;
|
---|
616 | }
|
---|
617 | if(response.equals("")) {
|
---|
618 | return extid;
|
---|
619 | } else {
|
---|
620 | return response;
|
---|
621 | }
|
---|
622 | }
|
---|
623 |
|
---|
624 |
|
---|
625 | /** translates relative oids into proper oids:
|
---|
626 | * .pr (parent), .rt (root) .fc (first child), .lc (last child),
|
---|
627 | * .ns (next sibling), .ps (previous sibling)
|
---|
628 | * .np (next page), .pp (previous page) : links sections in the order that you'd read the document
|
---|
629 | * a suffix is expected to be present so test before using
|
---|
630 | */
|
---|
631 | public String processOID(String doc_id, String top, String suff, int sibling_num) {
|
---|
632 |
|
---|
633 | // send off request to get sibling etc. information from Fedora
|
---|
634 | Element response = null;
|
---|
635 | String[] children = null;
|
---|
636 | if(doc_id.startsWith("CL")) { // classifiernode
|
---|
637 | response = getResponseAsDOM(fedoraServicesAPIA.retrieveBrowseStructure(this.cluster_name, "ClassifierBrowse", new String[]{doc_id},
|
---|
638 | new String[]{"children"}, new String[]{"siblingPosition"}));
|
---|
639 | NodeList nl = response.getElementsByTagName(GSXML.NODE_STRUCTURE_ELEM);
|
---|
640 | if(nl.getLength() > 0) {
|
---|
641 | Element nodeStructure = (Element)nl.item(0);
|
---|
642 |
|
---|
643 | if(nodeStructure != null) {
|
---|
644 | Element root = (Element) GSXML.getChildByTagName(nodeStructure, GSXML.CLASS_NODE_ELEM);
|
---|
645 | if(root != null) { // get children
|
---|
646 | NodeList classNodes = root.getElementsByTagName(GSXML.CLASS_NODE_ELEM);
|
---|
647 | if(classNodes != null) {
|
---|
648 | children = new String[classNodes.getLength()];
|
---|
649 | for(int i = 0; i < children.length; i++) {
|
---|
650 | Element child = (Element)classNodes.item(i);
|
---|
651 | children[i] = child.getAttribute(GSXML.NODE_ID_ATT);
|
---|
652 | }
|
---|
653 | }
|
---|
654 | }
|
---|
655 | }
|
---|
656 | }
|
---|
657 | } else { // documentnode
|
---|
658 | response = getResponseAsDOM(fedoraServicesAPIA.retrieveDocumentStructure(this.cluster_name, new String[]{doc_id},
|
---|
659 | new String[]{"children"}, new String[]{"siblingPosition"}));
|
---|
660 | String path = GSPath.createPath(new String[]{GSXML.RESPONSE_ELEM, GSXML.DOC_NODE_ELEM+GSXML.LIST_MODIFIER,
|
---|
661 | GSXML.DOC_NODE_ELEM, GSXML.NODE_STRUCTURE_ELEM, GSXML.DOC_NODE_ELEM});
|
---|
662 | Element parentDocNode = (Element) GSXML.getNodeByPath(response, path);
|
---|
663 |
|
---|
664 | if (parentDocNode == null) {
|
---|
665 | return top;
|
---|
666 | } // else
|
---|
667 | NodeList docNodes = parentDocNode.getElementsByTagName(GSXML.DOC_NODE_ELEM); // only children should remain, since that's what we requested
|
---|
668 | if(docNodes.getLength() > 0) {
|
---|
669 | children = new String[docNodes.getLength()];
|
---|
670 |
|
---|
671 | for(int i = 0; i < children.length; i++) {
|
---|
672 | Element e = (Element)docNodes.item(i);
|
---|
673 | children[i] = e.getAttribute(GSXML.NODE_ID_ATT);
|
---|
674 | }
|
---|
675 | } else { // return root node
|
---|
676 | children = new String[]{doc_id};
|
---|
677 | }
|
---|
678 | }
|
---|
679 |
|
---|
680 | if (suff.equals("fc")) {
|
---|
681 | return children[0];
|
---|
682 | } else if (suff.equals("lc")) {
|
---|
683 | return children[children.length-1];
|
---|
684 | } else {
|
---|
685 | if (suff.equals("ss")) {
|
---|
686 | return children[sibling_num-1];
|
---|
687 | }
|
---|
688 | // find the position that we are at.
|
---|
689 | int i=0;
|
---|
690 | while(i<children.length) {
|
---|
691 | if (children[i].equals(top)) {
|
---|
692 | break;
|
---|
693 | }
|
---|
694 | i++;
|
---|
695 | }
|
---|
696 |
|
---|
697 | if (suff.equals("ns")) {
|
---|
698 | if (i==children.length-1) {
|
---|
699 | return children[i];
|
---|
700 | }
|
---|
701 | return children[i+1];
|
---|
702 | } else if (suff.equals("ps")) {
|
---|
703 | if (i==0) {
|
---|
704 | return children[i];
|
---|
705 | }
|
---|
706 | return children[i-1];
|
---|
707 | }
|
---|
708 | }
|
---|
709 |
|
---|
710 | return top;
|
---|
711 | }
|
---|
712 |
|
---|
713 |
|
---|
714 | protected Element getResponseAsDOM(String response) {
|
---|
715 | if(response == null) { // will not be the case, because an empty
|
---|
716 | return null; // response message will be sent instead
|
---|
717 | }
|
---|
718 |
|
---|
719 | Element message = null;
|
---|
720 | try{
|
---|
721 | // turn the String xml response into a DOM tree:
|
---|
722 | DocumentBuilder builder
|
---|
723 | = DocumentBuilderFactory.newInstance().newDocumentBuilder();
|
---|
724 | Document doc
|
---|
725 | = builder.parse(new InputSource(new StringReader(response)));
|
---|
726 | message = doc.getDocumentElement();
|
---|
727 | } catch(Exception e){
|
---|
728 | if(response == null) {
|
---|
729 | response = "";
|
---|
730 | }
|
---|
731 | logger.error("An error occurred while trying to parse the response: ");
|
---|
732 | logger.error(response);
|
---|
733 | logger.error(e.getMessage());
|
---|
734 | }
|
---|
735 |
|
---|
736 | // Error elements in message will be processed outside of here, just return the message
|
---|
737 | return message;
|
---|
738 | }
|
---|
739 |
|
---|
740 | /* //process method for stylesheet requests
|
---|
741 | protected Element processFormat(Element request) {} */
|
---|
742 |
|
---|
743 | /* returns the service list for the subclass */
|
---|
744 | /* protected Element getServiceList(String lang) {
|
---|
745 | // for now, it is static and has no language stuff
|
---|
746 | return (Element) this.short_service_info.cloneNode(true);
|
---|
747 | }*/
|
---|
748 |
|
---|
749 | /** returns a specific service description */
|
---|
750 | protected Element getServiceDescription(Document doc, String service, String lang, String subset) {
|
---|
751 | if(!lang.equals(prevLanguage)) {
|
---|
752 | prevLanguage = lang;
|
---|
753 | fedoraServicesAPIA.setLanguage(lang);
|
---|
754 | }
|
---|
755 | String serviceResponse = fedoraServicesAPIA.describeService(service);
|
---|
756 | Element response = getResponseAsDOM(serviceResponse);
|
---|
757 |
|
---|
758 | // should be no chance of an npe, since FedoraGS3 lists the services, so will have descriptions for each
|
---|
759 | Element e = (Element)response.getElementsByTagName(GSXML.SERVICE_ELEM).item(0);
|
---|
760 | e = (Element)doc.importNode(e, true);
|
---|
761 | return e;
|
---|
762 | }
|
---|
763 |
|
---|
764 | /** overloaded version for no args case */
|
---|
765 | protected String getTextString(String key, String lang) {
|
---|
766 | return getTextString(key, lang, null, null);
|
---|
767 | }
|
---|
768 |
|
---|
769 | protected String getTextString(String key, String lang, String dictionary) {
|
---|
770 | return getTextString(key, lang, dictionary, null);
|
---|
771 | }
|
---|
772 | protected String getTextString(String key, String lang, String [] args) {
|
---|
773 | return getTextString(key, lang, null, args);
|
---|
774 | }
|
---|
775 |
|
---|
776 | /** getTextString - retrieves a language specific text string for the given
|
---|
777 | key and locale, from the specified resource_bundle (dictionary)
|
---|
778 | */
|
---|
779 | protected String getTextString(String key, String lang, String dictionary, String[] args) {
|
---|
780 |
|
---|
781 | // we want to use the collection class loader in case there are coll specific files
|
---|
782 | if (dictionary != null) {
|
---|
783 | // just try the one specified dictionary
|
---|
784 | Dictionary dict = new Dictionary(dictionary, lang, this.class_loader);
|
---|
785 | String result = dict.get(key, args);
|
---|
786 | if (result == null) { // not found
|
---|
787 | return "_"+key+"_";
|
---|
788 | }
|
---|
789 | return result;
|
---|
790 | }
|
---|
791 |
|
---|
792 | // now we try class names for dictionary names
|
---|
793 | String class_name = this.getClass().getName();
|
---|
794 | class_name = class_name.substring(class_name.lastIndexOf('.')+1);
|
---|
795 | Dictionary dict = new Dictionary(class_name, lang, this.class_loader);
|
---|
796 | String result = dict.get(key, args);
|
---|
797 | if (result != null) {
|
---|
798 | return result;
|
---|
799 | }
|
---|
800 |
|
---|
801 | // we have to try super classes
|
---|
802 | Class c = this.getClass().getSuperclass();
|
---|
803 | while (result == null && c != null) {
|
---|
804 | class_name = c.getName();
|
---|
805 | class_name = class_name.substring(class_name.lastIndexOf('.')+1);
|
---|
806 | if (class_name.equals("ServiceRack")) {
|
---|
807 | // this is as far as we go
|
---|
808 | break;
|
---|
809 | }
|
---|
810 | dict = new Dictionary(class_name, lang, this.class_loader);
|
---|
811 | result = dict.get(key, args);
|
---|
812 | c = c.getSuperclass();
|
---|
813 | }
|
---|
814 | if (result == null) {
|
---|
815 | return "_"+key+"_";
|
---|
816 | }
|
---|
817 | return result;
|
---|
818 |
|
---|
819 | }
|
---|
820 |
|
---|
821 | protected String getMetadataNameText(String key, String lang) {
|
---|
822 |
|
---|
823 | String properties_name = "metadata_names";
|
---|
824 | Dictionary dict = new Dictionary(properties_name, lang);
|
---|
825 |
|
---|
826 | String result = dict.get(key);
|
---|
827 | if (result == null) { // not found
|
---|
828 | return null;
|
---|
829 | }
|
---|
830 | return result;
|
---|
831 | }
|
---|
832 |
|
---|
833 | public static class BasicTextMacroResolver extends MacroResolver {
|
---|
834 | private static final Pattern p_back_slash = Pattern.compile("\\\"");// create a pattern "\\\"", but it matches both " and \"
|
---|
835 |
|
---|
836 | public String resolve(String text, String lang, String scope, String doc_oid)
|
---|
837 | {
|
---|
838 |
|
---|
839 | if (text == null || text.equals("")) {
|
---|
840 | return text;
|
---|
841 | }
|
---|
842 | if (!scope.equals(SCOPE_TEXT) || text_macros.size()==0) {
|
---|
843 | return text;
|
---|
844 | }
|
---|
845 |
|
---|
846 | java.util.ArrayList macros = text_macros;
|
---|
847 | for (int i=0; i<macros.size(); i++) {
|
---|
848 | String new_text = null;
|
---|
849 | Macro m = (Macro)macros.get(i);
|
---|
850 |
|
---|
851 | if(m.type == TYPE_TEXT) {
|
---|
852 | // make sure we resolve any macros in the text
|
---|
853 |
|
---|
854 | if(text.contains(m.macro)) {
|
---|
855 | if (m.resolve) {
|
---|
856 | new_text = this.resolve(m.text, lang, scope, doc_oid);
|
---|
857 | } else {
|
---|
858 | new_text = m.text;
|
---|
859 | }
|
---|
860 | text = StringUtils.replace(text, m.macro, new_text);//text = text.replaceAll(m.macro, new_text);
|
---|
861 | if (m.macro.endsWith("\\\\")) { // to get rid of "\" from the string like: "src="http://www.greenstone.org:80/.../mw.gif\">"
|
---|
862 | Matcher m_slash = p_back_slash.matcher(text);
|
---|
863 | String clean_str = "";
|
---|
864 | int s=0;
|
---|
865 | while (m_slash.find()) {
|
---|
866 | if (!text.substring(m_slash.end()-2, m_slash.end()-1).equals("\\")) {
|
---|
867 | clean_str = clean_str + text.substring(s,m_slash.end()-1); // it matches ", so get a substring before "
|
---|
868 | }else{
|
---|
869 | clean_str = clean_str + text.substring(s,m_slash.end()-2);// it matches \", so get a substring before \
|
---|
870 | }
|
---|
871 | s = m_slash.end();// get the index of the last match
|
---|
872 | clean_str = clean_str + "\"";
|
---|
873 | }
|
---|
874 | text = clean_str + text.substring(s,text.length());
|
---|
875 | }
|
---|
876 | }
|
---|
877 | }
|
---|
878 | }
|
---|
879 | return text;
|
---|
880 | }
|
---|
881 | }
|
---|
882 |
|
---|
883 |
|
---|
884 | }
|
---|
885 |
|
---|