1 | /*
|
---|
2 | * AbstractGS2DocumentRetrieve.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 org.greenstone.gsdl3.core.GSException;
|
---|
23 | import org.greenstone.gsdl3.util.GSXML;
|
---|
24 | import org.greenstone.gsdl3.util.GSFile;
|
---|
25 | import org.greenstone.gsdl3.util.OID;
|
---|
26 | import org.greenstone.gsdl3.util.MacroResolver;
|
---|
27 | import org.greenstone.gsdl3.util.GS2MacroResolver;
|
---|
28 | import org.greenstone.gsdl3.util.GSConstants;
|
---|
29 | import org.greenstone.gsdl3.util.SimpleCollectionDatabase;
|
---|
30 | import org.greenstone.gsdl3.util.DBInfo;
|
---|
31 | // XML classes
|
---|
32 | import org.w3c.dom.Document;
|
---|
33 | import org.w3c.dom.Element;
|
---|
34 | import org.w3c.dom.NodeList;
|
---|
35 |
|
---|
36 | // General Java classes
|
---|
37 | import java.io.File;
|
---|
38 | import java.util.StringTokenizer;
|
---|
39 | import java.util.Vector;
|
---|
40 | import java.util.Set;
|
---|
41 | import java.util.Iterator;
|
---|
42 | import java.util.ArrayList;
|
---|
43 |
|
---|
44 | import org.apache.log4j.*;
|
---|
45 |
|
---|
46 | /** Implements the generic retrieval and classifier services for GS2
|
---|
47 | * collections.
|
---|
48 | *
|
---|
49 | * @author <a href="mailto:[email protected]">Katherine Don</a>
|
---|
50 | * @author <a href="mailto:[email protected]">Michael Dewsnip</a>
|
---|
51 | */
|
---|
52 |
|
---|
53 | public abstract class AbstractGS2DocumentRetrieve
|
---|
54 | extends AbstractDocumentRetrieve {
|
---|
55 |
|
---|
56 | static Logger logger = Logger.getLogger(org.greenstone.gsdl3.service.AbstractGS2DocumentRetrieve.class.getName());
|
---|
57 |
|
---|
58 | // protected static final String EXTLINK_PARAM = "ext"; here or in base??
|
---|
59 | protected String index_stem = null;
|
---|
60 |
|
---|
61 | protected SimpleCollectionDatabase coll_db = null;
|
---|
62 |
|
---|
63 |
|
---|
64 | /** constructor */
|
---|
65 | protected AbstractGS2DocumentRetrieve()
|
---|
66 | {
|
---|
67 | }
|
---|
68 |
|
---|
69 | public void cleanUp() {
|
---|
70 | super.cleanUp();
|
---|
71 | this.coll_db.closeDatabase();
|
---|
72 | }
|
---|
73 | /** configure this service */
|
---|
74 | public boolean configure(Element info, Element extra_info)
|
---|
75 | {
|
---|
76 | if (!super.configure(info, extra_info)){
|
---|
77 | return false;
|
---|
78 | }
|
---|
79 |
|
---|
80 | logger.info("Configuring AbstractGS2DocumentRetrieve...");
|
---|
81 | //this.config_info = info;
|
---|
82 |
|
---|
83 | // the index stem is either specified in the config file or is the collection name
|
---|
84 | Element index_stem_elem = (Element) GSXML.getChildByTagName(info, GSXML.INDEX_STEM_ELEM);
|
---|
85 | if (index_stem_elem != null) {
|
---|
86 | this.index_stem = index_stem_elem.getAttribute(GSXML.NAME_ATT);
|
---|
87 | }
|
---|
88 | if (this.index_stem == null || this.index_stem.equals("")) {
|
---|
89 | logger.error("AbstractGS2DocumentRetrieve.configure(): indexStem element not found, stem will default to collection name");
|
---|
90 | this.index_stem = this.cluster_name;
|
---|
91 | }
|
---|
92 |
|
---|
93 | // find out what kind of database we have
|
---|
94 | Element database_type_elem = (Element) GSXML.getChildByTagName(info, GSXML.DATABASE_TYPE_ELEM);
|
---|
95 | String database_type = null;
|
---|
96 | if (database_type_elem != null) {
|
---|
97 | database_type = database_type_elem.getAttribute(GSXML.NAME_ATT);
|
---|
98 | }
|
---|
99 | if (database_type == null || database_type.equals("")) {
|
---|
100 | database_type = "gdbm"; // the default
|
---|
101 | }
|
---|
102 | coll_db = new SimpleCollectionDatabase(database_type);
|
---|
103 | if (coll_db == null) {
|
---|
104 | logger.error("Couldn't create the collection database of type "+database_type);
|
---|
105 | return false;
|
---|
106 | }
|
---|
107 |
|
---|
108 | // Open database for querying
|
---|
109 | String coll_db_file = GSFile.collectionDatabaseFile(this.site_home, this.cluster_name, this.index_stem, database_type);
|
---|
110 | if (!this.coll_db.openDatabase(coll_db_file, SimpleCollectionDatabase.READ)) {
|
---|
111 | logger.error("Could not open collection database!");
|
---|
112 | return false;
|
---|
113 | }
|
---|
114 | this.macro_resolver = new GS2MacroResolver(this.coll_db);
|
---|
115 |
|
---|
116 | return true;
|
---|
117 | }
|
---|
118 |
|
---|
119 | /** if id ends in .fc, .pc etc, then translate it to the correct id */
|
---|
120 | protected String translateId(String node_id) {
|
---|
121 | return this.coll_db.translateOID(node_id);
|
---|
122 | }
|
---|
123 |
|
---|
124 | /** if an id is not a greenstone id (an external id) then translate
|
---|
125 | it to a greenstone one*/
|
---|
126 | protected String translateExternalId(String node_id){
|
---|
127 | return this.coll_db.externalId2OID(node_id);
|
---|
128 | }
|
---|
129 |
|
---|
130 | /** returns the id of the root node of the document containing node node_id. . may be the same as node_id */
|
---|
131 | protected String getRootId(String node_id) {
|
---|
132 | return OID.getTop(node_id);
|
---|
133 | }
|
---|
134 | /** returns a list of the child ids in order, null if no children */
|
---|
135 | protected ArrayList getChildrenIds(String node_id) {
|
---|
136 | DBInfo info = this.coll_db.getInfo(node_id);
|
---|
137 | if (info == null) {
|
---|
138 | return null;
|
---|
139 | }
|
---|
140 |
|
---|
141 | String contains = info.getInfo("contains");
|
---|
142 | if (contains.equals("")) {
|
---|
143 | return null;
|
---|
144 | }
|
---|
145 | ArrayList children = new ArrayList();
|
---|
146 | StringTokenizer st = new StringTokenizer(contains, ";");
|
---|
147 | while (st.hasMoreTokens()) {
|
---|
148 | String child_id = st.nextToken().replaceAll("\"", node_id);
|
---|
149 | children.add(child_id);
|
---|
150 | }
|
---|
151 | return children;
|
---|
152 |
|
---|
153 | }
|
---|
154 | /** returns the node id of the parent node, null if no parent */
|
---|
155 | protected String getParentId(String node_id){
|
---|
156 | String parent = OID.getParent(node_id);
|
---|
157 | if (parent.equals(node_id)) {
|
---|
158 | return null;
|
---|
159 | }
|
---|
160 | return parent;
|
---|
161 | }
|
---|
162 |
|
---|
163 | /** get the metadata for the classifier node node_id
|
---|
164 | * returns a metadataList element:
|
---|
165 | * <metadataList><metadata name="xxx">value</metadata></metadataList>
|
---|
166 | */
|
---|
167 | // assumes only one value per metadata
|
---|
168 | protected Element getMetadataList(String node_id, boolean all_metadata,
|
---|
169 | ArrayList metadata_names)
|
---|
170 | throws GSException {
|
---|
171 | Element metadata_list = this.doc.createElement(GSXML.METADATA_ELEM+GSXML.LIST_MODIFIER);
|
---|
172 | DBInfo info = this.coll_db.getInfo(node_id);
|
---|
173 | if (info == null) {
|
---|
174 | return null;
|
---|
175 | }
|
---|
176 | String lang = "en"; // why do we need this??
|
---|
177 | if (all_metadata) {
|
---|
178 | // return everything out of the database
|
---|
179 | Set keys = info.getKeys();
|
---|
180 | Iterator it = keys.iterator();
|
---|
181 | while(it.hasNext()) {
|
---|
182 | String key = (String)it.next();
|
---|
183 | //String value = info.getInfo(key);
|
---|
184 | Vector values = info.getMultiInfo(key);
|
---|
185 | for(int i=0; i<values.size(); i++) {
|
---|
186 | GSXML.addMetadata(this.doc, metadata_list, key, this.macro_resolver.resolve((String)values.elementAt(i), lang, MacroResolver.SCOPE_META, node_id));
|
---|
187 | }
|
---|
188 | }
|
---|
189 |
|
---|
190 | } else {
|
---|
191 | for (int i=0; i<metadata_names.size(); i++) {
|
---|
192 | String meta_name = (String) metadata_names.get(i);
|
---|
193 | String value = getMetadata(node_id, info, meta_name, lang);
|
---|
194 | GSXML.addMetadata(this.doc, metadata_list, meta_name, value);
|
---|
195 | }
|
---|
196 | }
|
---|
197 | return metadata_list;
|
---|
198 | }
|
---|
199 |
|
---|
200 | /** returns the structural information asked for.
|
---|
201 | * info_type may be one of
|
---|
202 | * INFO_NUM_SIBS, INFO_NUM_CHILDREN, INFO_SIB_POS
|
---|
203 | */
|
---|
204 | protected String getStructureInfo(String doc_id, String info_type) {
|
---|
205 | String value="";
|
---|
206 | if (info_type.equals(INFO_NUM_SIBS)) {
|
---|
207 | String parent_id = OID.getParent(doc_id);
|
---|
208 | if (parent_id.equals(doc_id)) {
|
---|
209 | value="0";
|
---|
210 | } else {
|
---|
211 | value = String.valueOf(getNumChildren(parent_id));
|
---|
212 | }
|
---|
213 | return value;
|
---|
214 | }
|
---|
215 |
|
---|
216 | if (info_type.equals(INFO_NUM_CHILDREN)) {
|
---|
217 | return String.valueOf(getNumChildren(doc_id));
|
---|
218 | }
|
---|
219 |
|
---|
220 |
|
---|
221 | if (info_type.equals(INFO_SIB_POS)) {
|
---|
222 | String parent_id = OID.getParent(doc_id);
|
---|
223 | if (parent_id.equals(doc_id)) {
|
---|
224 | return "-1";
|
---|
225 | }
|
---|
226 |
|
---|
227 | DBInfo info = this.coll_db.getInfo(parent_id);
|
---|
228 | if (info==null) {
|
---|
229 | return "-1";
|
---|
230 | }
|
---|
231 |
|
---|
232 | String contains = info.getInfo("contains");
|
---|
233 | contains = contains.replaceAll("\"", parent_id);
|
---|
234 | String [] children = contains.split(";");
|
---|
235 | for (int i=0;i<children.length;i++) {
|
---|
236 | String child_id = children[i];
|
---|
237 | if (child_id.equals(doc_id)) {
|
---|
238 | return String.valueOf(i+1); // make it from 1 to length
|
---|
239 |
|
---|
240 | }
|
---|
241 | }
|
---|
242 |
|
---|
243 | return "-1";
|
---|
244 | } else {
|
---|
245 | return null;
|
---|
246 | }
|
---|
247 |
|
---|
248 | }
|
---|
249 |
|
---|
250 | protected int getNumChildren(String node_id) {
|
---|
251 | DBInfo info = this.coll_db.getInfo(node_id);
|
---|
252 | if (info == null) {
|
---|
253 | return 0;
|
---|
254 | }
|
---|
255 | String contains = info.getInfo("contains");
|
---|
256 | if (contains.equals("")) {
|
---|
257 | return 0;
|
---|
258 | }
|
---|
259 | String [] children = contains.split(";");
|
---|
260 | return children.length;
|
---|
261 | }
|
---|
262 |
|
---|
263 | /** returns the document type of the doc that the specified node
|
---|
264 | belongs to. should be one of
|
---|
265 | GSXML.DOC_TYPE_SIMPLE,
|
---|
266 | GSXML.DOC_TYPE_PAGED,
|
---|
267 | GSXML.DOC_TYPE_HIERARCHY
|
---|
268 | */
|
---|
269 | protected String getDocType(String node_id) {
|
---|
270 | DBInfo info = this.coll_db.getInfo(node_id);
|
---|
271 | if (info == null) {
|
---|
272 | return GSXML.DOC_TYPE_SIMPLE;
|
---|
273 | }
|
---|
274 | String doc_type = info.getInfo("doctype");
|
---|
275 | if (!doc_type.equals("")&&!doc_type.equals("doc")) {
|
---|
276 | return doc_type;
|
---|
277 | }
|
---|
278 |
|
---|
279 | String top_id = OID.getTop(node_id);
|
---|
280 | boolean is_top = (top_id.equals(node_id) ? true : false);
|
---|
281 |
|
---|
282 | String children = info.getInfo("contains");
|
---|
283 | boolean is_leaf = (children.equals("") ? true : false);
|
---|
284 |
|
---|
285 | if (is_top && is_leaf) { // a single section document
|
---|
286 | return GSXML.DOC_TYPE_SIMPLE;
|
---|
287 | }
|
---|
288 |
|
---|
289 | // now we just check the top node
|
---|
290 | if (!is_top) { // we need to look at the top info
|
---|
291 | info = this.coll_db.getInfo(top_id);
|
---|
292 | }
|
---|
293 | if (info == null) {
|
---|
294 | return GSXML.DOC_TYPE_HIERARCHY;
|
---|
295 | }
|
---|
296 |
|
---|
297 | String childtype = info.getInfo("childtype");
|
---|
298 | if (childtype.equals("Paged")) {
|
---|
299 | return GSXML.DOC_TYPE_PAGED;
|
---|
300 | }
|
---|
301 | return GSXML.DOC_TYPE_HIERARCHY;
|
---|
302 | }
|
---|
303 |
|
---|
304 | /** returns the content of a node
|
---|
305 | * should return a nodeContent element:
|
---|
306 | * <nodeContent>text content or other elements</nodeContent>
|
---|
307 | */
|
---|
308 | abstract protected Element getNodeContent(String doc_id, String lang) throws GSException;
|
---|
309 |
|
---|
310 | protected String getMetadata(String node_id, DBInfo info,
|
---|
311 | String metadata, String lang) {
|
---|
312 | boolean multiple = false;
|
---|
313 | String relation = "";
|
---|
314 | String separator = ", ";
|
---|
315 | int pos = metadata.indexOf(GSConstants.META_RELATION_SEP);
|
---|
316 | if (pos ==-1) {
|
---|
317 | Vector values = info.getMultiInfo(metadata);
|
---|
318 | if (values !=null){
|
---|
319 | // just a plain meta entry eg dc.Title
|
---|
320 | StringBuffer result = new StringBuffer();
|
---|
321 | boolean first = true;
|
---|
322 | for (int i=0; i<values.size(); i++) {
|
---|
323 | if (first) {
|
---|
324 | first = false;
|
---|
325 | } else {
|
---|
326 | result.append(separator);
|
---|
327 | }
|
---|
328 | result.append(this.macro_resolver.resolve((String)values.elementAt(i), lang, MacroResolver.SCOPE_META, node_id));
|
---|
329 | }
|
---|
330 | return result.toString();
|
---|
331 | }
|
---|
332 | else{
|
---|
333 | String result = info.getInfo(metadata);
|
---|
334 | return this.macro_resolver.resolve(result, lang, MacroResolver.SCOPE_META, node_id);
|
---|
335 | }
|
---|
336 | }
|
---|
337 |
|
---|
338 | String temp = metadata.substring(0, pos);
|
---|
339 | metadata = metadata.substring(pos+1);
|
---|
340 | // check for all on the front
|
---|
341 | if (temp.equals("all")) {
|
---|
342 | multiple=true;
|
---|
343 | pos = metadata.indexOf(GSConstants.META_RELATION_SEP);
|
---|
344 | if (pos ==-1) {
|
---|
345 | temp = "";
|
---|
346 | } else {
|
---|
347 | temp = metadata.substring(0, pos);
|
---|
348 | metadata = metadata.substring(pos+1);
|
---|
349 | }
|
---|
350 | }
|
---|
351 |
|
---|
352 | // now check for relational info
|
---|
353 | if (temp.equals("parent") || temp.equals("root") || temp.equals( "ancestors")) { // "current" "siblings" "children" "descendants"
|
---|
354 | relation = temp;
|
---|
355 | pos = metadata.indexOf(GSConstants.META_RELATION_SEP);
|
---|
356 | if (pos == -1) {
|
---|
357 | temp = "";
|
---|
358 | } else {
|
---|
359 | temp = metadata.substring(0, pos);
|
---|
360 | metadata = metadata.substring(pos+1);
|
---|
361 | }
|
---|
362 | }
|
---|
363 |
|
---|
364 | // now look for separator info
|
---|
365 | if (temp.startsWith(GSConstants.META_SEPARATOR_SEP) && temp.endsWith(GSConstants.META_SEPARATOR_SEP)) {
|
---|
366 | separator = temp.substring(1, temp.length()-1);
|
---|
367 |
|
---|
368 | }
|
---|
369 |
|
---|
370 | String relation_id = node_id;
|
---|
371 | if (relation.equals("parent") || relation.equals("ancestors")) {
|
---|
372 | relation_id = OID.getParent(node_id);
|
---|
373 | // parent or ancestor does not include self
|
---|
374 | if (relation_id.equals(node_id)){
|
---|
375 | return "";
|
---|
376 | }
|
---|
377 | } else if (relation.equals("root")) {
|
---|
378 | relation_id = OID.getTop(node_id);
|
---|
379 | }
|
---|
380 |
|
---|
381 | // now we either have a single node, or we have ancestors
|
---|
382 | DBInfo relation_info;
|
---|
383 | if (relation_id.equals(node_id)) {
|
---|
384 | relation_info = info;
|
---|
385 | } else {
|
---|
386 | relation_info = this.coll_db.getInfo(relation_id);
|
---|
387 | }
|
---|
388 | if (relation_info == null) {
|
---|
389 | return "";
|
---|
390 | }
|
---|
391 |
|
---|
392 | StringBuffer result = new StringBuffer();
|
---|
393 |
|
---|
394 | if (!multiple) {
|
---|
395 | result.append(this.macro_resolver.resolve(relation_info.getInfo(metadata), lang, MacroResolver.SCOPE_META, relation_id));
|
---|
396 | } else {
|
---|
397 | // we have multiple meta
|
---|
398 | Vector values = relation_info.getMultiInfo(metadata);
|
---|
399 | if (values != null) {
|
---|
400 | boolean first = true;
|
---|
401 | for (int i=0; i<values.size(); i++) {
|
---|
402 | if (first) {
|
---|
403 | first = false;
|
---|
404 | } else {
|
---|
405 | result.append(separator);
|
---|
406 | }
|
---|
407 | result.append(this.macro_resolver.resolve((String)values.elementAt(i), lang, MacroResolver.SCOPE_META, relation_id));
|
---|
408 | }
|
---|
409 | }
|
---|
410 | logger.info(result);
|
---|
411 | }
|
---|
412 | // if not ancestors, then this is all we do
|
---|
413 | if (!relation.equals("ancestors")) {
|
---|
414 | return result.toString();
|
---|
415 | }
|
---|
416 |
|
---|
417 | // now do the ancestors
|
---|
418 | String current_id = relation_id;
|
---|
419 | relation_id = OID.getParent(current_id);
|
---|
420 | while (!relation_id.equals(current_id)) {
|
---|
421 | relation_info = this.coll_db.getInfo(relation_id);
|
---|
422 | if (relation_info == null) return result.toString();
|
---|
423 | if (!multiple) {
|
---|
424 | result.insert(0, separator);
|
---|
425 | result.insert(0, this.macro_resolver.resolve(relation_info.getInfo(metadata), lang, MacroResolver.SCOPE_META, relation_id));
|
---|
426 | } else {
|
---|
427 | Vector values = relation_info.getMultiInfo(metadata);
|
---|
428 | if (values != null) {
|
---|
429 | for (int i=values.size()-1; i>=0; i--) {
|
---|
430 | result.insert(0, separator);
|
---|
431 | result.insert(0, this.macro_resolver.resolve((String)values.elementAt(i), lang, MacroResolver.SCOPE_META, relation_id));
|
---|
432 | }
|
---|
433 | }
|
---|
434 |
|
---|
435 | }
|
---|
436 | current_id = relation_id;
|
---|
437 | relation_id = OID.getParent(current_id);
|
---|
438 | }
|
---|
439 | return result.toString();
|
---|
440 | }
|
---|
441 |
|
---|
442 |
|
---|
443 | /** needs to get info from collection database - if the calling code gets it already it may pay to pass it in instead */
|
---|
444 | protected String resolveTextMacros(String doc_content, String doc_id, String lang)
|
---|
445 | {
|
---|
446 | // resolve any collection specific macros
|
---|
447 | doc_content = macro_resolver.resolve(doc_content, lang, MacroResolver.SCOPE_TEXT, doc_id);
|
---|
448 | return doc_content;
|
---|
449 | }
|
---|
450 |
|
---|
451 | protected Element getInfo(String doc_id, String info_type) {
|
---|
452 |
|
---|
453 | String value="";
|
---|
454 | if (info_type.equals(INFO_NUM_SIBS)) {
|
---|
455 | String parent_id = OID.getParent(doc_id);
|
---|
456 | if (parent_id.equals(doc_id)) {
|
---|
457 | value="0";
|
---|
458 | } else {
|
---|
459 | value = String.valueOf(getNumChildren(parent_id));
|
---|
460 | }
|
---|
461 | } else if (info_type.equals(INFO_NUM_CHILDREN)) {
|
---|
462 | value = String.valueOf(getNumChildren(doc_id));
|
---|
463 | } else if (info_type.equals(INFO_SIB_POS)) {
|
---|
464 | String parent_id = OID.getParent(doc_id);
|
---|
465 | if (parent_id.equals(doc_id)) {
|
---|
466 | value="-1";
|
---|
467 | } else {
|
---|
468 | DBInfo info = this.coll_db.getInfo(parent_id);
|
---|
469 | if (info==null) {
|
---|
470 | value ="-1";
|
---|
471 | } else {
|
---|
472 | String contains = info.getInfo("contains");
|
---|
473 | contains = contains.replaceAll("\"", parent_id);
|
---|
474 | String [] children = contains.split(";");
|
---|
475 | for (int i=0;i<children.length;i++) {
|
---|
476 | String child_id = children[i];
|
---|
477 | if (child_id.equals(doc_id)) {
|
---|
478 | value = String.valueOf(i+1); // make it from 1 to length
|
---|
479 | break;
|
---|
480 | }
|
---|
481 | }
|
---|
482 | }
|
---|
483 | }
|
---|
484 | } else {
|
---|
485 | return null;
|
---|
486 | }
|
---|
487 | Element info_elem = this.doc.createElement("info");
|
---|
488 | info_elem.setAttribute(GSXML.NAME_ATT, info_type);
|
---|
489 | info_elem.setAttribute(GSXML.VALUE_ATT, value);
|
---|
490 | return info_elem;
|
---|
491 | }
|
---|
492 |
|
---|
493 | protected String getHrefOID(String href_url){
|
---|
494 | return this.coll_db.docnum2OID(href_url);
|
---|
495 | }
|
---|
496 |
|
---|
497 | }
|
---|