1 | /*
|
---|
2 | * GS2Retrieve.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 |
|
---|
22 | // Greenstone classes
|
---|
23 | import org.greenstone.gdbm.*;
|
---|
24 | import org.greenstone.gsdl3.util.*;
|
---|
25 |
|
---|
26 | // XML classes
|
---|
27 | import org.w3c.dom.Document;
|
---|
28 | import org.w3c.dom.Element;
|
---|
29 | import org.w3c.dom.NodeList;
|
---|
30 |
|
---|
31 | // General Java classes
|
---|
32 | import java.io.File;
|
---|
33 | import java.util.StringTokenizer;
|
---|
34 | import java.util.Vector;
|
---|
35 | import java.util.Set;
|
---|
36 | import java.util.Iterator;
|
---|
37 |
|
---|
38 | /** Implements the generic retrieval and classifier services for GS2
|
---|
39 | * collections.
|
---|
40 | *
|
---|
41 | * @author <a href="mailto:[email protected]">Katherine Don</a>
|
---|
42 | * @author <a href="mailto:[email protected]">Michael Dewsnip</a>
|
---|
43 | * @version $Revision: 4903 $
|
---|
44 | */
|
---|
45 |
|
---|
46 | public abstract class GS2Retrieve
|
---|
47 | extends ServiceRack {
|
---|
48 |
|
---|
49 | // the services on offer
|
---|
50 | // these strings must match what is found in the properties file
|
---|
51 | protected static final String DOCUMENT_STRUCTURE_RETRIEVE_SERVICE = "DocumentStructureRetrieve";
|
---|
52 | protected static final String DOCUMENT_METADATA_RETRIEVE_SERVICE = "DocumentMetadataRetrieve";
|
---|
53 | protected static final String DOCUMENT_CONTENT_RETRIEVE_SERVICE = "DocumentContentRetrieve";
|
---|
54 |
|
---|
55 |
|
---|
56 | // the browsing services - now in here, these will only be advertised if classifiers have been specified in the config file
|
---|
57 | private static final String CLASSIFIER_SERVICE = "ClassifierBrowse";
|
---|
58 | private static final String CLASSIFIER_METADATA_SERVICE = "ClassifierBrowseMetadataRetrieve";
|
---|
59 |
|
---|
60 | protected static final String STRUCT_PARAM = "structure";
|
---|
61 | protected static final String INFO_PARAM = "info";
|
---|
62 |
|
---|
63 | protected static final String STRUCT_ANCESTORS = "ancestors";
|
---|
64 | protected static final String STRUCT_PARENT = "parent";
|
---|
65 | protected static final String STRUCT_SIBS = "siblings";
|
---|
66 | protected static final String STRUCT_CHILDREN = "children";
|
---|
67 | protected static final String STRUCT_DESCENDS = "descendants";
|
---|
68 |
|
---|
69 | protected static final String INFO_NUM_SIBS = "numSiblings";
|
---|
70 | protected static final String INFO_NUM_CHILDREN = "numChildren";
|
---|
71 | protected static final String INFO_SIB_POS = "siblingPosition";
|
---|
72 |
|
---|
73 | protected static final int DOCUMENT=1;
|
---|
74 | protected static final int CLASSIFIER=2;
|
---|
75 |
|
---|
76 | protected GDBMWrapper gdbm_src_ = null;
|
---|
77 | protected Element config_info_ = null; // the xml from the config file
|
---|
78 |
|
---|
79 | /** constructor */
|
---|
80 | protected GS2Retrieve()
|
---|
81 | {
|
---|
82 | gdbm_src_ = new GDBMWrapper();
|
---|
83 | }
|
---|
84 |
|
---|
85 |
|
---|
86 | /** configure this service */
|
---|
87 | public boolean configure(Element info, Element extra_info)
|
---|
88 | {
|
---|
89 | System.out.println("Configuring GS2Retrieve...");
|
---|
90 | config_info_ = info;
|
---|
91 |
|
---|
92 | // set up short_service_info_ - for now just has name and type
|
---|
93 | Element dsr_service = doc_.createElement(GSXML.SERVICE_ELEM);
|
---|
94 | dsr_service.setAttribute(GSXML.TYPE_ATT, GSXML.SERVICE_TYPE_RETRIEVE);
|
---|
95 | dsr_service.setAttribute(GSXML.NAME_ATT, DOCUMENT_STRUCTURE_RETRIEVE_SERVICE);
|
---|
96 | short_service_info_.appendChild(dsr_service);
|
---|
97 |
|
---|
98 | Element dmr_service = doc_.createElement(GSXML.SERVICE_ELEM);
|
---|
99 | dmr_service.setAttribute(GSXML.TYPE_ATT, GSXML.SERVICE_TYPE_RETRIEVE);
|
---|
100 | dmr_service.setAttribute(GSXML.NAME_ATT, DOCUMENT_METADATA_RETRIEVE_SERVICE);
|
---|
101 | short_service_info_.appendChild(dmr_service);
|
---|
102 |
|
---|
103 | Element dcr_service = doc_.createElement(GSXML.SERVICE_ELEM);
|
---|
104 | dcr_service.setAttribute(GSXML.TYPE_ATT, GSXML.SERVICE_TYPE_RETRIEVE);
|
---|
105 | dcr_service.setAttribute(GSXML.NAME_ATT, DOCUMENT_CONTENT_RETRIEVE_SERVICE);
|
---|
106 | short_service_info_.appendChild(dcr_service);
|
---|
107 |
|
---|
108 | // Open GDBM database for querying
|
---|
109 | String gdbm_db_file = GSFile.GDBMDatabaseFile(site_home_, cluster_name_);
|
---|
110 | if (!gdbm_src_.openDatabase(gdbm_db_file, GDBMWrapper.READER)) {
|
---|
111 | System.err.println("Error: Could not open GDBM database!");
|
---|
112 | return false;
|
---|
113 | }
|
---|
114 |
|
---|
115 | // now do the classifier browse service
|
---|
116 |
|
---|
117 | // check that there are classifiers specified
|
---|
118 | Element class_list = (Element)GSXML.getChildByTagName(info, GSXML.CLASSIFIER_ELEM+GSXML.LIST_MODIFIER);
|
---|
119 | if (class_list == null) {
|
---|
120 | // no classifiers specified
|
---|
121 | return true;
|
---|
122 | }
|
---|
123 |
|
---|
124 | // get the display and format elements from the coll config file for
|
---|
125 | // the classifiers
|
---|
126 | extractExtraClassifierInfo(info, extra_info);
|
---|
127 | config_info_ = info;
|
---|
128 |
|
---|
129 | // short_service_info_ - the browse one
|
---|
130 | Element cb_service = doc_.createElement(GSXML.SERVICE_ELEM);
|
---|
131 | cb_service.setAttribute(GSXML.TYPE_ATT, GSXML.SERVICE_TYPE_BROWSE);
|
---|
132 | cb_service.setAttribute(GSXML.NAME_ATT, CLASSIFIER_SERVICE);
|
---|
133 | short_service_info_.appendChild(cb_service);
|
---|
134 |
|
---|
135 | // metadata retrieval for the browsing
|
---|
136 | Element cbmr_service = doc_.createElement(GSXML.SERVICE_ELEM);
|
---|
137 | cbmr_service.setAttribute(GSXML.TYPE_ATT, GSXML.SERVICE_TYPE_RETRIEVE);
|
---|
138 | cbmr_service.setAttribute(GSXML.NAME_ATT, CLASSIFIER_METADATA_SERVICE);
|
---|
139 | short_service_info_.appendChild(cbmr_service);
|
---|
140 |
|
---|
141 | // the format info
|
---|
142 | Element cb_format_info = doc_.createElement(GSXML.FORMAT_ELEM);
|
---|
143 | boolean format_found = false;
|
---|
144 | // add in to the description a simplified list of classifiers
|
---|
145 | NodeList classifiers = class_list.getElementsByTagName(GSXML.CLASSIFIER_ELEM);
|
---|
146 | for(int i=0; i<classifiers.getLength(); i++) {
|
---|
147 | Element cl = (Element)classifiers.item(i);
|
---|
148 | Element new_cl = (Element)doc_.importNode(cl, false); // just import this node, not the children
|
---|
149 |
|
---|
150 | // get the format info out, and put inside a classifier element
|
---|
151 | Element format_cl = (Element)new_cl.cloneNode(false);
|
---|
152 | Element format = (Element)GSXML.getChildByTagName(cl, GSXML.FORMAT_ELEM);
|
---|
153 | if (format != null) {
|
---|
154 |
|
---|
155 | //copy all the children
|
---|
156 | NodeList elems = format.getChildNodes();
|
---|
157 | for (int j=0; j<elems.getLength();j++) {
|
---|
158 | format_cl.appendChild(doc_.importNode(elems.item(j), true));
|
---|
159 | }
|
---|
160 | cb_format_info.appendChild(format_cl);
|
---|
161 | format_found = true;
|
---|
162 | }
|
---|
163 |
|
---|
164 |
|
---|
165 | }
|
---|
166 |
|
---|
167 | if (format_found) {
|
---|
168 | format_info_map_.put(CLASSIFIER_SERVICE, cb_format_info);
|
---|
169 | }
|
---|
170 |
|
---|
171 | // look for document display format
|
---|
172 | String path = GSPath.appendLink(GSXML.DISPLAY_ELEM, GSXML.FORMAT_ELEM);
|
---|
173 | Element display_format = (Element)GSXML.getNodeByPath(extra_info, path);
|
---|
174 | if (display_format != null) {
|
---|
175 | System.out.println("display format not null!");
|
---|
176 | format_info_map_.put(DOCUMENT_METADATA_RETRIEVE_SERVICE, doc_.importNode(display_format, true));
|
---|
177 | // shoudl we make a copy?
|
---|
178 | }
|
---|
179 | return true;
|
---|
180 | }
|
---|
181 |
|
---|
182 | protected Element getServiceDescription(String service_id, String lang) {
|
---|
183 |
|
---|
184 | if (service_id.equals(CLASSIFIER_SERVICE)) {
|
---|
185 |
|
---|
186 | Element class_list = (Element)GSXML.getChildByTagName(config_info_, GSXML.CLASSIFIER_ELEM+GSXML.LIST_MODIFIER);
|
---|
187 | if (class_list == null) {
|
---|
188 | // no classifiers specified
|
---|
189 | return null;
|
---|
190 | }
|
---|
191 |
|
---|
192 | Element cb_service = doc_.createElement(GSXML.SERVICE_ELEM);
|
---|
193 | cb_service.setAttribute(GSXML.TYPE_ATT, GSXML.SERVICE_TYPE_BROWSE);
|
---|
194 | cb_service.setAttribute(GSXML.NAME_ATT, CLASSIFIER_SERVICE);
|
---|
195 | cb_service.appendChild(GSXML.createDisplayTextElement(doc_, GSXML.DISPLAY_TEXT_NAME, getTextString(CLASSIFIER_SERVICE+".name", lang)));
|
---|
196 |
|
---|
197 | Element cl_list = doc_.createElement(GSXML.CLASSIFIER_ELEM+GSXML.LIST_MODIFIER);
|
---|
198 | cb_service.appendChild(cl_list);
|
---|
199 | NodeList classifiers = class_list.getElementsByTagName(GSXML.CLASSIFIER_ELEM);
|
---|
200 | for(int i=0; i<classifiers.getLength(); i++) {
|
---|
201 | Element cl = (Element)classifiers.item(i);
|
---|
202 | Element new_cl = (Element)doc_.importNode(cl, false); // just import this node, not the children
|
---|
203 |
|
---|
204 | cl_list.appendChild(new_cl);
|
---|
205 | String text = GSXML.getDisplayText(cl,
|
---|
206 | GSXML.DISPLAY_TEXT_NAME,
|
---|
207 | lang, "en");
|
---|
208 | if (text == null || text.equals("")) {
|
---|
209 | // no display element was specified, use the metadata name
|
---|
210 | // for now this looks in the class properties file
|
---|
211 | // this needs to use a general metadata thing instead
|
---|
212 | text = getTextString(cl.getAttribute(GSXML.CLASSIFIER_CONTENT_ATT), lang);
|
---|
213 | }
|
---|
214 |
|
---|
215 | Element cl_name = GSXML.createDisplayTextElement(doc_, GSXML.DISPLAY_TEXT_NAME, text);
|
---|
216 | new_cl.appendChild(cl_name);
|
---|
217 |
|
---|
218 | }
|
---|
219 | return cb_service;
|
---|
220 | }
|
---|
221 |
|
---|
222 | // these ones are probably never called, but put them here just in case
|
---|
223 |
|
---|
224 | if (service_id.equals(CLASSIFIER_METADATA_SERVICE)) {
|
---|
225 |
|
---|
226 | Element cbmr_service = doc_.createElement(GSXML.SERVICE_ELEM);
|
---|
227 | cbmr_service.setAttribute(GSXML.TYPE_ATT, GSXML.SERVICE_TYPE_RETRIEVE);
|
---|
228 | cbmr_service.setAttribute(GSXML.NAME_ATT, CLASSIFIER_METADATA_SERVICE);
|
---|
229 | return cbmr_service;
|
---|
230 | }
|
---|
231 |
|
---|
232 | if (service_id.equals(DOCUMENT_STRUCTURE_RETRIEVE_SERVICE)) {
|
---|
233 | Element dsr_service = doc_.createElement(GSXML.SERVICE_ELEM);
|
---|
234 | dsr_service.setAttribute(GSXML.TYPE_ATT, GSXML.SERVICE_TYPE_RETRIEVE);
|
---|
235 | dsr_service.setAttribute(GSXML.NAME_ATT, DOCUMENT_STRUCTURE_RETRIEVE_SERVICE);
|
---|
236 | return dsr_service;
|
---|
237 | }
|
---|
238 | if (service_id.equals(DOCUMENT_METADATA_RETRIEVE_SERVICE)) {
|
---|
239 | Element dmr_service = doc_.createElement(GSXML.SERVICE_ELEM);
|
---|
240 | dmr_service.setAttribute(GSXML.TYPE_ATT, GSXML.SERVICE_TYPE_RETRIEVE);
|
---|
241 | dmr_service.setAttribute(GSXML.NAME_ATT, DOCUMENT_METADATA_RETRIEVE_SERVICE);
|
---|
242 | return dmr_service;
|
---|
243 | }
|
---|
244 |
|
---|
245 | if (service_id.equals(DOCUMENT_CONTENT_RETRIEVE_SERVICE)) {
|
---|
246 | Element dcr_service = doc_.createElement(GSXML.SERVICE_ELEM);
|
---|
247 | dcr_service.setAttribute(GSXML.TYPE_ATT, GSXML.SERVICE_TYPE_RETRIEVE);
|
---|
248 | dcr_service.setAttribute(GSXML.NAME_ATT, DOCUMENT_CONTENT_RETRIEVE_SERVICE);
|
---|
249 | return dcr_service;
|
---|
250 | }
|
---|
251 |
|
---|
252 | return null;
|
---|
253 | }
|
---|
254 |
|
---|
255 | /** this looks for any classifier specific display or format info from extra_info and adds it in to the correct place in info */
|
---|
256 | protected boolean extractExtraClassifierInfo(Element info, Element extra_info) {
|
---|
257 |
|
---|
258 | if (extra_info == null) {
|
---|
259 | return false;
|
---|
260 | }
|
---|
261 |
|
---|
262 | Document owner = info.getOwnerDocument();
|
---|
263 | // so far we have display and format elements that we need for classifiers
|
---|
264 | NodeList classifiers = info.getElementsByTagName(GSXML.CLASSIFIER_ELEM);
|
---|
265 | Element config_browse = (Element)GSXML.getChildByTagName(extra_info, GSXML.BROWSE_ELEM);
|
---|
266 |
|
---|
267 | for (int i=0; i<classifiers.getLength();i++) {
|
---|
268 | Element cl = (Element)classifiers.item(i);
|
---|
269 | String name = cl.getAttribute(GSXML.NAME_ATT);
|
---|
270 | Element node_extra = GSXML.getNamedElement(config_browse,
|
---|
271 | GSXML.CLASSIFIER_ELEM,
|
---|
272 | GSXML.NAME_ATT,
|
---|
273 | name);
|
---|
274 | if (node_extra == null) {
|
---|
275 | System.err.println("GS2REtrieve: haven't found extra info for classifier named "+name);
|
---|
276 | continue;
|
---|
277 | }
|
---|
278 |
|
---|
279 | // get the display elements if any - displayName
|
---|
280 | NodeList display_names = node_extra.getElementsByTagName(GSXML.DISPLAY_TEXT_ELEM);
|
---|
281 | if (display_names !=null) {
|
---|
282 | Element display = owner.createElement(GSXML.DISPLAY_ELEM);
|
---|
283 | for (int j=0; j<display_names.getLength(); j++) {
|
---|
284 | Element e = (Element)display_names.item(j);
|
---|
285 | cl.appendChild(owner.importNode(e, true));
|
---|
286 |
|
---|
287 | }
|
---|
288 | }
|
---|
289 |
|
---|
290 | // get the format element if any
|
---|
291 | Element format = (Element)GSXML.getChildByTagName(node_extra, GSXML.FORMAT_ELEM);
|
---|
292 | if (format==null) { // try a generic one that applies to all classifiers
|
---|
293 | format = (Element)GSXML.getChildByTagName(extra_info,
|
---|
294 | GSXML.FORMAT_ELEM);
|
---|
295 | }
|
---|
296 | if (format!=null) { // append to index info
|
---|
297 | cl.appendChild(owner.importNode(format, true));
|
---|
298 | }
|
---|
299 | } // for each classifier
|
---|
300 | return true;
|
---|
301 | }
|
---|
302 |
|
---|
303 |
|
---|
304 | /** parent is true if this node is definitely the parent of something,
|
---|
305 | * child is true is it definitely is a child of something - just for efficiency purposes */
|
---|
306 | protected Element createDocNode(String node_id, boolean parent, boolean child) {
|
---|
307 |
|
---|
308 | // create this here or pass it in?
|
---|
309 | DBInfo info = gdbm_src_.getInfo(node_id);
|
---|
310 | Element node;
|
---|
311 | if (isClassifier(node_id)) {
|
---|
312 | node = doc_.createElement(GSXML.CLASS_NODE_ELEM);
|
---|
313 | //String childtype = info.getInfo("childtype");
|
---|
314 | //String orientation="";
|
---|
315 | //if (childtype.equals("HList")) {
|
---|
316 | // orientation = "horizontal";
|
---|
317 | //} else { // assume vertical
|
---|
318 | // orientation = "vertical";
|
---|
319 | //}
|
---|
320 | //node.setAttribute(GSXML.CLASS_NODE_ORIENTATION_ATT, orientation);
|
---|
321 | } else {
|
---|
322 |
|
---|
323 | node = doc_.createElement(GSXML.DOC_NODE_ELEM);
|
---|
324 |
|
---|
325 | String top_id = OID.getTop(node_id);
|
---|
326 | boolean is_top = (top_id.equals(node_id) ? true : false);
|
---|
327 |
|
---|
328 | String children = info.getInfo("contains");
|
---|
329 | boolean is_leaf = (children.equals("") ? true : false);
|
---|
330 |
|
---|
331 | // set teh node type att
|
---|
332 | if (is_top) {
|
---|
333 | node.setAttribute(GSXML.NODE_TYPE_ATT, GSXML.NODE_TYPE_ROOT);
|
---|
334 | } else if (is_leaf) {
|
---|
335 | node.setAttribute(GSXML.NODE_TYPE_ATT, GSXML.NODE_TYPE_LEAF);
|
---|
336 | } else {
|
---|
337 | node.setAttribute(GSXML.NODE_TYPE_ATT, GSXML.NODE_TYPE_INTERIOR);
|
---|
338 | }
|
---|
339 |
|
---|
340 | // set teh doc type att
|
---|
341 | if (is_top && is_leaf) { // a single section document
|
---|
342 | node.setAttribute(GSXML.DOC_TYPE_ATT, "simple");
|
---|
343 |
|
---|
344 | } else {
|
---|
345 |
|
---|
346 | if (!is_top) { // we need to look at the top info
|
---|
347 | info = gdbm_src_.getInfo(top_id);
|
---|
348 | }
|
---|
349 |
|
---|
350 | String childtype = info.getInfo("childtype");
|
---|
351 | if (childtype.equals("Paged")) {
|
---|
352 | node.setAttribute(GSXML.DOC_TYPE_ATT, "paged");
|
---|
353 | } else {
|
---|
354 | node.setAttribute(GSXML.DOC_TYPE_ATT, "hierarchy");
|
---|
355 | }
|
---|
356 | }
|
---|
357 |
|
---|
358 | }
|
---|
359 | node.setAttribute(GSXML.NODE_ID_ATT, node_id);
|
---|
360 | return node;
|
---|
361 |
|
---|
362 | }
|
---|
363 | /** Returns the parent of a specified documentID, or null if none exists */
|
---|
364 | protected Element getParent(String doc_id)
|
---|
365 | {
|
---|
366 | String parent_id = OID.getParent(doc_id);
|
---|
367 | if (parent_id.equals(doc_id))
|
---|
368 | return null;
|
---|
369 |
|
---|
370 | return createDocNode(parent_id, true, false);
|
---|
371 | }
|
---|
372 |
|
---|
373 |
|
---|
374 | /** adds all the children of doc_id the the doc element,
|
---|
375 | * and if recursive=true, adds all their children as well*/
|
---|
376 | protected void addDescendants(Element doc, String doc_id,
|
---|
377 | boolean recursive)
|
---|
378 | {
|
---|
379 | DBInfo info = gdbm_src_.getInfo(doc_id);
|
---|
380 | String contains = info.getInfo("contains");
|
---|
381 |
|
---|
382 | StringTokenizer st = new StringTokenizer(contains, ";");
|
---|
383 | while (st.hasMoreTokens()) {
|
---|
384 | String child_id = st.nextToken().replaceAll("\"", doc_id);
|
---|
385 | Element child = createDocNode(child_id, false, true);
|
---|
386 | doc.appendChild(child);
|
---|
387 |
|
---|
388 | // Apply recursively, if desired
|
---|
389 | if (recursive) {
|
---|
390 | addDescendants(child, child_id, recursive);
|
---|
391 | }
|
---|
392 |
|
---|
393 | }
|
---|
394 | }
|
---|
395 |
|
---|
396 | /** adds all the siblings of current_id to the parent element. */
|
---|
397 | protected Element addSiblings(Element parent, String parent_id, String current_id) {
|
---|
398 | Element current_node = (Element)parent.getFirstChild();
|
---|
399 | if (current_node ==null) {
|
---|
400 | // create a sensible error message
|
---|
401 | System.err.println("Error: there should be a first child.");
|
---|
402 | return null;
|
---|
403 | }
|
---|
404 | // remove the current child,- will add it in later in its correct place
|
---|
405 | parent.removeChild(current_node);
|
---|
406 |
|
---|
407 | // add in all the siblings,
|
---|
408 | addDescendants(parent, parent_id, false);
|
---|
409 |
|
---|
410 | // find the node that is now the current node
|
---|
411 | // this assumes that the new node that was created is the same as
|
---|
412 | // the old one that was removed - we may want to replace the new one
|
---|
413 | // with the old one.
|
---|
414 | Element new_current = GSXML.getNamedElement(parent, current_node.getNodeName(), GSXML.NODE_ID_ATT, current_id);
|
---|
415 | return new_current;
|
---|
416 |
|
---|
417 | }
|
---|
418 | /** Returns true if the OID specifies a leaf node, false otherwise
|
---|
419 | Note: this makes a request to the GDBM database so it may not be
|
---|
420 | a particularly cheap operation */
|
---|
421 | protected boolean isLeafNode(String oid)
|
---|
422 | {
|
---|
423 | DBInfo info = gdbm_src_.getInfo(oid);
|
---|
424 | String children = info.getInfo("contains");
|
---|
425 | return (children.equals(""));
|
---|
426 | }
|
---|
427 |
|
---|
428 | // for now just use CL for classifiers - should have a type? in teh gdbm
|
---|
429 | // database.
|
---|
430 | protected boolean isClassifier(String oid) {
|
---|
431 | if (oid.startsWith("CL")) {
|
---|
432 | return true;
|
---|
433 | }
|
---|
434 | return false;
|
---|
435 | }
|
---|
436 |
|
---|
437 | protected Element processDocumentStructureRetrieve(Element request) {
|
---|
438 | return genericStructureRetrieve(request, DOCUMENT);
|
---|
439 | }
|
---|
440 |
|
---|
441 | protected Element processClassifierBrowse(Element request) {
|
---|
442 | return genericStructureRetrieve(request, CLASSIFIER);
|
---|
443 | }
|
---|
444 |
|
---|
445 | /** Retrieve the structure of a document */
|
---|
446 | protected Element genericStructureRetrieve(Element request, int type)
|
---|
447 | {
|
---|
448 | // Create a new (empty) result message
|
---|
449 | Element result = doc_.createElement(GSXML.RESPONSE_ELEM);
|
---|
450 |
|
---|
451 | String node_name;
|
---|
452 | String service_name;
|
---|
453 | if (type==DOCUMENT) {
|
---|
454 | service_name = DOCUMENT_STRUCTURE_RETRIEVE_SERVICE;
|
---|
455 | node_name = GSXML.DOC_NODE_ELEM;
|
---|
456 | } else {
|
---|
457 | service_name = CLASSIFIER_SERVICE;
|
---|
458 | node_name = GSXML.CLASS_NODE_ELEM;
|
---|
459 | }
|
---|
460 |
|
---|
461 | result.setAttribute(GSXML.FROM_ATT, service_name);
|
---|
462 | result.setAttribute(GSXML.TYPE_ATT, GSXML.REQUEST_TYPE_PROCESS);
|
---|
463 |
|
---|
464 | // Get the parameters of the request
|
---|
465 | Element param_list = (Element) GSXML.getChildByTagName(request, GSXML.PARAM_ELEM+GSXML.LIST_MODIFIER);
|
---|
466 | if (param_list == null) {
|
---|
467 | System.err.println("Error: DocumentStructureRetrieve request had no paramList.");
|
---|
468 | return result; // Return the empty result
|
---|
469 | }
|
---|
470 |
|
---|
471 | // the type of info required
|
---|
472 | boolean want_structure = false;
|
---|
473 | boolean want_info = false;
|
---|
474 |
|
---|
475 | Vector info_types=new Vector();
|
---|
476 | // The document structure information desired
|
---|
477 | boolean want_ancestors = false;
|
---|
478 | boolean want_parent = false;
|
---|
479 | boolean want_siblings = false;
|
---|
480 | boolean want_children = false;
|
---|
481 | boolean want_descendants = false;
|
---|
482 |
|
---|
483 | // Process the request parameters
|
---|
484 | NodeList params = param_list.getElementsByTagName(GSXML.PARAM_ELEM);
|
---|
485 | for (int i=0; i<params.getLength();i++) {
|
---|
486 |
|
---|
487 | Element param = (Element)params.item(i);
|
---|
488 | String p_name = param.getAttribute(GSXML.NAME_ATT);
|
---|
489 | String p_value = GSXML.getValue(param);
|
---|
490 | // Identify the structure information desired
|
---|
491 | if (p_name.equals(STRUCT_PARAM)) {
|
---|
492 | want_structure = true;
|
---|
493 |
|
---|
494 | // This is NOT locale sensitive
|
---|
495 | if (p_value.equals(STRUCT_ANCESTORS))
|
---|
496 | want_ancestors = true;
|
---|
497 | else if (p_value.equals(STRUCT_PARENT))
|
---|
498 | want_parent = true;
|
---|
499 | else if (p_value.equals(STRUCT_SIBS))
|
---|
500 | want_siblings = true;
|
---|
501 | else if (p_value.equals(STRUCT_CHILDREN))
|
---|
502 | want_children = true;
|
---|
503 | else if (p_value.equals(STRUCT_DESCENDS))
|
---|
504 | want_descendants = true;
|
---|
505 | else
|
---|
506 | System.err.println("Warning: Unknown value \"" + p_value + "\".");
|
---|
507 | } else if (p_name.equals(INFO_PARAM)) {
|
---|
508 | want_info = true;
|
---|
509 | info_types.add(p_value);
|
---|
510 | }
|
---|
511 | }
|
---|
512 |
|
---|
513 | // Make sure there is no repeated information
|
---|
514 | if (want_ancestors)
|
---|
515 | want_parent = false;
|
---|
516 | if (want_descendants)
|
---|
517 | want_children = false;
|
---|
518 |
|
---|
519 |
|
---|
520 | Element query_doc_list = (Element) GSXML.getChildByTagName(request, node_name+GSXML.LIST_MODIFIER);
|
---|
521 | if (query_doc_list == null) {
|
---|
522 | System.err.println("Error: DocumentStructureRetrieve request specified no doc nodes.\n");
|
---|
523 | return result;
|
---|
524 | }
|
---|
525 |
|
---|
526 | Element doc_list = doc_.createElement(node_name+GSXML.LIST_MODIFIER);
|
---|
527 | result.appendChild(doc_list);
|
---|
528 |
|
---|
529 | // Get the documents
|
---|
530 | String[] doc_ids = GSXML.getAttributeValuesFromList(query_doc_list,
|
---|
531 | GSXML.NODE_ID_ATT);
|
---|
532 | for (int i = 0; i < doc_ids.length; i++) {
|
---|
533 | String doc_id = doc_ids[i];
|
---|
534 |
|
---|
535 | System.out.println("doc_id = "+doc_id);
|
---|
536 | if (OID.needsTranslating(doc_id)) {
|
---|
537 | doc_id = gdbm_src_.translateOID(doc_id);
|
---|
538 | System.out.println("translated doc_id = "+doc_id);
|
---|
539 | }
|
---|
540 |
|
---|
541 | // Add the document to the list
|
---|
542 | Element doc = doc_.createElement(node_name);
|
---|
543 | doc_list.appendChild(doc);
|
---|
544 | doc.setAttribute(GSXML.NODE_ID_ATT, doc_id);
|
---|
545 |
|
---|
546 |
|
---|
547 | if (want_info) {
|
---|
548 |
|
---|
549 | Element info_elem = doc_.createElement("nodeStructureInfo");
|
---|
550 | doc.appendChild(info_elem);
|
---|
551 |
|
---|
552 | for (int j=0; j<info_types.size(); j++) {
|
---|
553 | String info_type = (String)info_types.get(j);
|
---|
554 | Element inf = getInfo(doc_id, info_type);
|
---|
555 | if (inf != null) {
|
---|
556 | info_elem.appendChild(inf);
|
---|
557 | }
|
---|
558 | }
|
---|
559 | }
|
---|
560 | if (want_structure) {
|
---|
561 | // all structure info goes into a nodeStructure elem
|
---|
562 | Element structure_elem = doc_.createElement(GSXML.NODE_STRUCTURE_ELEM);
|
---|
563 | doc.appendChild(structure_elem);
|
---|
564 |
|
---|
565 | // Add the requested structure information
|
---|
566 | Element current = createDocNode(doc_id, false, false);
|
---|
567 |
|
---|
568 | //Ancestors: continually add parent nodes until the root is reached
|
---|
569 | Element top_node = current; // the top node so far
|
---|
570 | if (want_ancestors) {
|
---|
571 | String current_id = doc_id;
|
---|
572 | while (true) {
|
---|
573 | Element parent = getParent(current_id);
|
---|
574 | if (parent == null)
|
---|
575 | break;
|
---|
576 |
|
---|
577 | parent.appendChild(top_node);
|
---|
578 | current_id = parent.getAttribute(GSXML.NODE_ID_ATT);
|
---|
579 | top_node = parent;
|
---|
580 | }
|
---|
581 | }
|
---|
582 | // Parent: get the parent of the selected node
|
---|
583 | if (want_parent) {
|
---|
584 | Element parent = getParent(doc_id);
|
---|
585 | if (parent != null) {
|
---|
586 | parent.appendChild(current);
|
---|
587 | top_node = parent;
|
---|
588 | }
|
---|
589 | }
|
---|
590 |
|
---|
591 |
|
---|
592 | // now the top node is the root of the structure
|
---|
593 | structure_elem.appendChild(top_node);
|
---|
594 |
|
---|
595 | //Siblings: get the other descendants of the selected node's parent
|
---|
596 | if (want_siblings) {
|
---|
597 | Element parent = (Element)current.getParentNode(); // this may be the structure element if there has been no request for parents or ancestors
|
---|
598 | String parent_id = OID.getParent(doc_id);
|
---|
599 |
|
---|
600 | // add siblings, - returns a pointer to the new current node
|
---|
601 | current = addSiblings(parent, parent_id, doc_id);
|
---|
602 | }
|
---|
603 |
|
---|
604 | // Children: get the descendants, but only one level deep
|
---|
605 | if (want_children)
|
---|
606 | addDescendants(current, doc_id, false);
|
---|
607 | // Descendants: recursively get every descendant of the selected node
|
---|
608 | if (want_descendants)
|
---|
609 | addDescendants(current, doc_id, true);
|
---|
610 | } // if want structure
|
---|
611 | } // for each doc
|
---|
612 | return result;
|
---|
613 | }
|
---|
614 |
|
---|
615 |
|
---|
616 | protected Element processDocumentMetadataRetrieve(Element request) {
|
---|
617 | return genericMetadataRetrieve(request, DOCUMENT);
|
---|
618 | }
|
---|
619 |
|
---|
620 | protected Element processClassifierBrowseMetadataRetrieve(Element request) {
|
---|
621 | return genericMetadataRetrieve(request, CLASSIFIER);
|
---|
622 | }
|
---|
623 |
|
---|
624 |
|
---|
625 | /** Retrieve metadata associated with a document or classifier node*/
|
---|
626 | protected Element genericMetadataRetrieve(Element request, int type)
|
---|
627 | {
|
---|
628 | // Create a new (empty) result message
|
---|
629 | Element result = doc_.createElement(GSXML.RESPONSE_ELEM);
|
---|
630 |
|
---|
631 | String node_name;
|
---|
632 |
|
---|
633 | String service_name;
|
---|
634 | if (type==DOCUMENT) {
|
---|
635 | service_name = DOCUMENT_METADATA_RETRIEVE_SERVICE;
|
---|
636 | node_name = GSXML.DOC_NODE_ELEM;
|
---|
637 | } else {
|
---|
638 | service_name = CLASSIFIER_METADATA_SERVICE;
|
---|
639 | node_name = GSXML.CLASS_NODE_ELEM;
|
---|
640 | }
|
---|
641 | result.setAttribute(GSXML.FROM_ATT, service_name);
|
---|
642 | result.setAttribute(GSXML.TYPE_ATT, GSXML.REQUEST_TYPE_PROCESS);
|
---|
643 |
|
---|
644 | // Get the parameters of the request
|
---|
645 | Element param_list = (Element) GSXML.getChildByTagName(request, GSXML.PARAM_ELEM+GSXML.LIST_MODIFIER);
|
---|
646 | if (param_list == null) {
|
---|
647 | System.err.println("GS2Retrieve, DocumentMetadataRetrieve Error: missing paramList.\n");
|
---|
648 | return result; // Return the empty result
|
---|
649 | }
|
---|
650 |
|
---|
651 | // The metadata information required
|
---|
652 | Vector metadata_list = new Vector();
|
---|
653 | boolean all_metadata = false;
|
---|
654 | // Process the request parameters
|
---|
655 | Element param = (Element) param_list.getFirstChild();
|
---|
656 | while (param != null) {
|
---|
657 | // Identify the metadata information desired
|
---|
658 | if (param.getAttribute(GSXML.NAME_ATT).equals("metadata")) {
|
---|
659 | String metadata = GSXML.getValue(param);
|
---|
660 | if (metadata.equals("all")) {
|
---|
661 | all_metadata = true;
|
---|
662 | break;
|
---|
663 | }
|
---|
664 | metadata_list.add(metadata);
|
---|
665 | }
|
---|
666 | param = (Element) param.getNextSibling();
|
---|
667 | }
|
---|
668 |
|
---|
669 | Element node_list = doc_.createElement(node_name+GSXML.LIST_MODIFIER);
|
---|
670 | result.appendChild(node_list);
|
---|
671 |
|
---|
672 | // Get the documents
|
---|
673 | Element request_node_list = (Element) GSXML.getChildByTagName(request, node_name+GSXML.LIST_MODIFIER);
|
---|
674 | if (request_node_list == null) {
|
---|
675 | System.err.println("Error: DocumentMetadataRetrieve request had no "+node_name+"List.\n");
|
---|
676 | return result;
|
---|
677 | }
|
---|
678 |
|
---|
679 | NodeList request_nodes = request_node_list.getChildNodes();
|
---|
680 | for (int i = 0; i < request_nodes.getLength(); i++) {
|
---|
681 | Element request_node = (Element) request_nodes.item(i);
|
---|
682 | String node_id = request_node.getAttribute(GSXML.NODE_ID_ATT);
|
---|
683 |
|
---|
684 | if (OID.needsTranslating(node_id)) {
|
---|
685 | node_id = gdbm_src_.translateOID(node_id);
|
---|
686 | }
|
---|
687 |
|
---|
688 | // Add the document to the list
|
---|
689 | Element new_node = (Element)doc_.importNode(request_node, false);
|
---|
690 | node_list.appendChild(new_node);
|
---|
691 |
|
---|
692 | // Add the requested metadata information
|
---|
693 | Element node_meta_list = doc_.createElement(GSXML.METADATA_ELEM+GSXML.LIST_MODIFIER);
|
---|
694 | new_node.appendChild(node_meta_list);
|
---|
695 | DBInfo info = gdbm_src_.getInfo(node_id);
|
---|
696 | if (info == null) {// I have had a case where it is null!
|
---|
697 | continue;
|
---|
698 | }
|
---|
699 | System.out.println("found the info");
|
---|
700 | if (all_metadata) {
|
---|
701 | System.out.println("trying to get all the metadata");
|
---|
702 | // return everything out of the database
|
---|
703 | Set keys = info.getKeys();
|
---|
704 | Iterator it = keys.iterator();
|
---|
705 | while(it.hasNext()) {
|
---|
706 | String key = (String)it.next();
|
---|
707 | System.out.println("getting metadata "+key);
|
---|
708 | String value = info.getInfo(key);
|
---|
709 | GSXML.addMetadata(doc_, node_meta_list, key, value);
|
---|
710 | }
|
---|
711 | } else { // just get the selected ones
|
---|
712 |
|
---|
713 | for (int m = 0; m < metadata_list.size(); m++) {
|
---|
714 | String metadata = (String) metadata_list.get(m);
|
---|
715 | String value = getMetadata(node_id, info, metadata);
|
---|
716 | GSXML.addMetadata(doc_, node_meta_list, metadata, value);
|
---|
717 | }
|
---|
718 | }
|
---|
719 | //String v = getMetadata (node_id, info, "parent_Title");
|
---|
720 | //System.out.println("parent title = "+v);
|
---|
721 | //v = getMetadata (node_id, info, "ancestors': '_Title");
|
---|
722 | //System.out.println("ancestortitle = "+v);
|
---|
723 | }
|
---|
724 |
|
---|
725 | return result;
|
---|
726 | }
|
---|
727 |
|
---|
728 | protected final char RELATION_SEP_CHAR = '_';
|
---|
729 | protected final String SEPARATOR_SEP_STRING = "'";
|
---|
730 |
|
---|
731 | protected String getMetadata(String node_id, DBInfo info,
|
---|
732 | String metadata) {
|
---|
733 | int pos = metadata.indexOf(RELATION_SEP_CHAR);
|
---|
734 | if (pos ==-1) {
|
---|
735 | return info.getInfo(metadata);
|
---|
736 | }
|
---|
737 |
|
---|
738 | String relation = metadata.substring(0, pos);
|
---|
739 | if (relation.equals("parent") || relation.startsWith("ancestors")) {
|
---|
740 | String parent_id = OID.getParent(node_id);
|
---|
741 | if (parent_id.equals(node_id)){
|
---|
742 | // no parents or ancestors
|
---|
743 | return "";
|
---|
744 | }
|
---|
745 | DBInfo parent_info = gdbm_src_.getInfo(parent_id);
|
---|
746 | if (parent_info == null) return "";
|
---|
747 | String new_meta = metadata.substring(pos+1);
|
---|
748 | if (relation.equals("parent")) {
|
---|
749 | return parent_info.getInfo(new_meta);
|
---|
750 | }
|
---|
751 | // do ancestor stuff
|
---|
752 | // get teh separating string
|
---|
753 | String items [] = relation.split(SEPARATOR_SEP_STRING);
|
---|
754 | String separator = "";
|
---|
755 | if (items.length > 1) {
|
---|
756 | separator = items[1];
|
---|
757 | relation = items[0];
|
---|
758 | }
|
---|
759 |
|
---|
760 | StringBuffer value = new StringBuffer();
|
---|
761 | value.append(parent_info.getInfo(new_meta));
|
---|
762 | String current_id = parent_id;
|
---|
763 | parent_id = OID.getParent(current_id);
|
---|
764 | while (!parent_id.equals(current_id)) {
|
---|
765 | parent_info = gdbm_src_.getInfo(parent_id);
|
---|
766 | if (parent_info == null) return value.toString();
|
---|
767 | value.insert(0, separator);
|
---|
768 | value.insert(0, parent_info.getInfo(new_meta));
|
---|
769 |
|
---|
770 | current_id = parent_id;
|
---|
771 | parent_id = OID.getParent(current_id);
|
---|
772 | }
|
---|
773 |
|
---|
774 | return value.toString();
|
---|
775 | }
|
---|
776 |
|
---|
777 |
|
---|
778 | // its not a relation after all
|
---|
779 | return info.getInfo(metadata);
|
---|
780 |
|
---|
781 | }
|
---|
782 |
|
---|
783 | /** Retrieve the content of a document - implemented by concrete subclasses */
|
---|
784 | protected abstract Element processDocumentContentRetrieve(Element request);
|
---|
785 |
|
---|
786 | /** needs to get info from gdbm database - if the calling code gets it already it may pay to pass it in instead */
|
---|
787 | protected String resolveImages(String doc_content, String doc_id)
|
---|
788 | {
|
---|
789 | String top_doc_id = OID.getTop(doc_id);
|
---|
790 | DBInfo info = gdbm_src_.getInfo(top_doc_id);
|
---|
791 | String archivedir = info.getInfo("archivedir");
|
---|
792 | String image_dir = site_http_address_ + "/collect/"+cluster_name_+"/index/assoc/"+archivedir;
|
---|
793 |
|
---|
794 | // Resolve all "_httpdocimg_"s
|
---|
795 | doc_content = doc_content.replaceAll("_httpdocimg_", image_dir);
|
---|
796 | return doc_content;
|
---|
797 | }
|
---|
798 |
|
---|
799 | protected Element getInfo(String doc_id, String info_type) {
|
---|
800 |
|
---|
801 | String value="";
|
---|
802 | if (info_type.equals(INFO_NUM_SIBS)) {
|
---|
803 | String parent_id = OID.getParent(doc_id);
|
---|
804 | if (parent_id.equals(doc_id)) {
|
---|
805 | value="0";
|
---|
806 | } else {
|
---|
807 | value = String.valueOf(getNumChildren(parent_id));
|
---|
808 | }
|
---|
809 | } else if (info_type.equals(INFO_NUM_CHILDREN)) {
|
---|
810 | value = String.valueOf(getNumChildren(doc_id));
|
---|
811 | } else if (info_type.equals(INFO_SIB_POS)) {
|
---|
812 | String parent_id = OID.getParent(doc_id);
|
---|
813 | if (parent_id.equals(doc_id)) {
|
---|
814 | value="-1";
|
---|
815 | } else {
|
---|
816 | DBInfo info = gdbm_src_.getInfo(parent_id);
|
---|
817 | String contains = info.getInfo("contains");
|
---|
818 | contains = contains.replaceAll("\"", parent_id);
|
---|
819 | String [] children = contains.split(";");
|
---|
820 | for (int i=0;i<children.length;i++) {
|
---|
821 | String child_id = children[i];
|
---|
822 | System.out.println("child="+child_id+" doc="+doc_id);
|
---|
823 | if (child_id.equals(doc_id)) {
|
---|
824 | value = String.valueOf(i+1); // make it from 1 to length
|
---|
825 | break;
|
---|
826 | }
|
---|
827 | }
|
---|
828 | }
|
---|
829 | } else {
|
---|
830 | return null;
|
---|
831 | }
|
---|
832 | Element info_elem = doc_.createElement("info");
|
---|
833 | info_elem.setAttribute(GSXML.NAME_ATT, info_type);
|
---|
834 | info_elem.setAttribute(GSXML.VALUE_ATT, value);
|
---|
835 | return info_elem;
|
---|
836 | }
|
---|
837 |
|
---|
838 | protected int getNumChildren(String doc_id) {
|
---|
839 | DBInfo info = gdbm_src_.getInfo(doc_id);
|
---|
840 | String contains = info.getInfo("contains");
|
---|
841 | if (contains.equals("")) {
|
---|
842 | return 0;
|
---|
843 | }
|
---|
844 | String [] children = contains.split(";");
|
---|
845 | return children.length;
|
---|
846 | }
|
---|
847 |
|
---|
848 | }
|
---|