[9002] | 1 | /*
|
---|
| 2 | * AbstractSearch.java
|
---|
| 3 | * Copyright (C) 2005 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
|
---|
[25764] | 22 | import java.util.ArrayList;
|
---|
| 23 |
|
---|
| 24 | import org.apache.log4j.Logger;
|
---|
[26042] | 25 | import org.greenstone.gsdl3.util.AbstractBasicDocument;
|
---|
| 26 | import org.greenstone.gsdl3.util.BasicDocument;
|
---|
[25764] | 27 | import org.greenstone.gsdl3.util.GSPath;
|
---|
[9002] | 28 | import org.greenstone.gsdl3.util.GSXML;
|
---|
[24858] | 29 | import org.w3c.dom.Element;
|
---|
[9002] | 30 | import org.w3c.dom.NodeList;
|
---|
| 31 |
|
---|
[24858] | 32 | /**
|
---|
| 33 | * Partially implements a generic search service
|
---|
| 34 | *
|
---|
[9002] | 35 | */
|
---|
| 36 |
|
---|
[24858] | 37 | public abstract class AbstractSearch extends ServiceRack
|
---|
[9002] | 38 | {
|
---|
[13124] | 39 |
|
---|
[24858] | 40 | static Logger logger = Logger.getLogger(org.greenstone.gsdl3.service.AbstractSearch.class.getName());
|
---|
[13124] | 41 |
|
---|
[24858] | 42 | // the search service
|
---|
| 43 | protected String QUERY_SERVICE = null; // set by subclass
|
---|
[9002] | 44 |
|
---|
[24858] | 45 | // compulsory params
|
---|
| 46 | protected static final String INDEX_PARAM = "index";
|
---|
| 47 | protected static final String QUERY_PARAM = "query";
|
---|
[24892] | 48 | protected static final String RAW_PARAM = "rawquery";
|
---|
[9002] | 49 |
|
---|
[24858] | 50 | // optional standard params - some of these have to be implemented
|
---|
| 51 | protected static final String MAXDOCS_PARAM = "maxDocs";
|
---|
| 52 | protected static final String HITS_PER_PAGE_PARAM = "hitsPerPage";
|
---|
| 53 | protected static final String START_PAGE_PARAM = "startPage";
|
---|
[13998] | 54 |
|
---|
[26042] | 55 | protected AbstractBasicDocument gs_doc = null;
|
---|
[9002] | 56 |
|
---|
[24858] | 57 | /** can more than one index be searched at the same time? */
|
---|
| 58 | protected boolean does_multi_index_search = false;
|
---|
| 59 | /** does this service support paging of results? */
|
---|
| 60 | protected boolean does_paging = false;
|
---|
| 61 | /** does this service support asking for a subset of results? */
|
---|
| 62 | protected boolean does_chunking = false;
|
---|
[25885] | 63 | /** does this service support faceting search results */
|
---|
| 64 | protected boolean does_faceting = false;
|
---|
[24858] | 65 | /**
|
---|
| 66 | * the default document type - use if all documents are the same type
|
---|
| 67 | */
|
---|
| 68 | protected String default_document_type = null;
|
---|
| 69 | /**
|
---|
| 70 | * the default index, or comma separated list if more than one is the
|
---|
| 71 | * default (with start and end commas, eg ,TI,SU,). Should be set by
|
---|
| 72 | * configure()
|
---|
| 73 | */
|
---|
| 74 | protected String default_index = "";
|
---|
[9002] | 75 |
|
---|
[24858] | 76 | protected String default_max_docs = "100";
|
---|
| 77 |
|
---|
| 78 | protected String default_hits_per_page = "10";
|
---|
| 79 |
|
---|
| 80 | public AbstractSearch()
|
---|
| 81 | {
|
---|
[10093] | 82 | }
|
---|
| 83 |
|
---|
[24858] | 84 | /**
|
---|
| 85 | * Sets up the short service info for service by QUERY_SERVICE (e.g.
|
---|
| 86 | * TextQuery or AudioQuery) If other services will be provided, should be
|
---|
| 87 | * added in the subclass configure also looks for search format info, and
|
---|
| 88 | * document format info
|
---|
| 89 | */
|
---|
| 90 | public boolean configure(Element info, Element extra_info)
|
---|
| 91 | {
|
---|
| 92 | if (!super.configure(info, extra_info))
|
---|
| 93 | {
|
---|
| 94 | return false;
|
---|
| 95 | }
|
---|
[9002] | 96 |
|
---|
[24858] | 97 | logger.info("Configuring AbstractSearch...");
|
---|
[24394] | 98 |
|
---|
[24858] | 99 | this.config_info = info;
|
---|
[9002] | 100 |
|
---|
[24858] | 101 | // set up short_service_info_
|
---|
| 102 | // => for now just has id and type. the name (lang dependent)
|
---|
| 103 | // will be added in if the list is requested.
|
---|
[24394] | 104 |
|
---|
[24858] | 105 | Element tq_service = this.doc.createElement(GSXML.SERVICE_ELEM);
|
---|
| 106 | tq_service.setAttribute(GSXML.TYPE_ATT, GSXML.SERVICE_TYPE_QUERY);
|
---|
| 107 | tq_service.setAttribute(GSXML.NAME_ATT, QUERY_SERVICE);
|
---|
| 108 | this.short_service_info.appendChild(tq_service);
|
---|
[13998] | 109 |
|
---|
[24858] | 110 | // add some format info to service map if there is any
|
---|
| 111 | // => lookin extra info first look in buildConfig
|
---|
[14183] | 112 |
|
---|
[24858] | 113 | Element format = (Element) GSXML.getChildByTagName(info, GSXML.FORMAT_ELEM);
|
---|
[24394] | 114 |
|
---|
[24858] | 115 | if (format == null)
|
---|
| 116 | {
|
---|
| 117 | String path = GSPath.appendLink(GSXML.SEARCH_ELEM, GSXML.FORMAT_ELEM);
|
---|
| 118 |
|
---|
| 119 | // Note by xiao: instead of retrieving the first 'format'
|
---|
| 120 | // element inside the 'search' element, we are trying to
|
---|
| 121 | // find the real format element which has at least one
|
---|
| 122 | // 'gsf:template' child element. (extra_info is
|
---|
| 123 | // collectionConfig.xml)
|
---|
| 124 |
|
---|
| 125 | //format = (Element) GSXML.getNodeByPath(extra_info, path);
|
---|
| 126 |
|
---|
| 127 | Element search_elem = (Element) GSXML.getChildByTagName(extra_info, GSXML.SEARCH_ELEM);
|
---|
| 128 | NodeList format_elems = null;
|
---|
| 129 | if (search_elem != null)
|
---|
| 130 | {
|
---|
| 131 | format_elems = search_elem.getElementsByTagName(GSXML.FORMAT_ELEM);
|
---|
| 132 | }
|
---|
| 133 | for (int i = 0; i < format_elems.getLength(); i++)
|
---|
| 134 | {
|
---|
| 135 | format = (Element) format_elems.item(i);
|
---|
| 136 | if (format.getElementsByTagName("gsf:template").getLength() != 0)
|
---|
| 137 | {
|
---|
| 138 | break;
|
---|
| 139 | }
|
---|
| 140 | }
|
---|
| 141 | }//end of if(format==null)
|
---|
| 142 | //
|
---|
| 143 | if (format != null)
|
---|
| 144 | {
|
---|
| 145 | this.format_info_map.put(QUERY_SERVICE, this.doc.importNode(format, true));
|
---|
[24394] | 146 | }
|
---|
[24858] | 147 |
|
---|
| 148 | // look for document display format - for documentType
|
---|
| 149 | String path = GSPath.appendLink(GSXML.DISPLAY_ELEM, GSXML.FORMAT_ELEM);
|
---|
| 150 | Element display_format = (Element) GSXML.getNodeByPath(extra_info, path);
|
---|
| 151 | if (display_format != null)
|
---|
| 152 | {
|
---|
| 153 | // check for docType option.
|
---|
| 154 | Element doc_type_opt = GSXML.getNamedElement(display_format, "gsf:option", GSXML.NAME_ATT, "documentType");
|
---|
| 155 | if (doc_type_opt != null)
|
---|
| 156 | {
|
---|
| 157 | String value = doc_type_opt.getAttribute(GSXML.VALUE_ATT);
|
---|
| 158 | if (!value.equals(""))
|
---|
| 159 | {
|
---|
| 160 | this.default_document_type = value;
|
---|
| 161 | }
|
---|
| 162 | }
|
---|
[9002] | 163 | }
|
---|
[24858] | 164 |
|
---|
| 165 | // Base line for document (might be overriden by sub-classes)
|
---|
[26042] | 166 | gs_doc = new BasicDocument(this.doc, this.default_document_type);
|
---|
[24858] | 167 |
|
---|
| 168 | return true;
|
---|
[9002] | 169 | }
|
---|
| 170 |
|
---|
[24858] | 171 | /**
|
---|
| 172 | * returns a basic description for QUERY_SERVICE. If a subclass provides
|
---|
| 173 | * other services they need to provide their own descriptions
|
---|
| 174 | */
|
---|
| 175 | protected Element getServiceDescription(String service, String lang, String subset)
|
---|
| 176 | {
|
---|
| 177 | if (!service.equals(QUERY_SERVICE))
|
---|
| 178 | {
|
---|
| 179 | return null;
|
---|
| 180 | }
|
---|
[24394] | 181 |
|
---|
[24858] | 182 | Element tq_service = this.doc.createElement(GSXML.SERVICE_ELEM);
|
---|
| 183 | tq_service.setAttribute(GSXML.TYPE_ATT, GSXML.SERVICE_TYPE_QUERY);
|
---|
| 184 | tq_service.setAttribute(GSXML.NAME_ATT, QUERY_SERVICE);
|
---|
| 185 | if (subset == null || subset.equals(GSXML.DISPLAY_TEXT_ELEM + GSXML.LIST_MODIFIER))
|
---|
| 186 | {
|
---|
| 187 | tq_service.appendChild(GSXML.createDisplayTextElement(this.doc, GSXML.DISPLAY_TEXT_NAME, getServiceName(QUERY_SERVICE, lang)));
|
---|
| 188 | tq_service.appendChild(GSXML.createDisplayTextElement(this.doc, GSXML.DISPLAY_TEXT_SUBMIT, getServiceSubmit(QUERY_SERVICE, lang)));
|
---|
| 189 | tq_service.appendChild(GSXML.createDisplayTextElement(this.doc, GSXML.DISPLAY_TEXT_DESCRIPTION, getServiceDescription(QUERY_SERVICE, lang)));
|
---|
| 190 | }
|
---|
| 191 | if (subset == null || subset.equals(GSXML.PARAM_ELEM + GSXML.LIST_MODIFIER))
|
---|
| 192 | {
|
---|
| 193 | Element param_list = this.doc.createElement(GSXML.PARAM_ELEM + GSXML.LIST_MODIFIER);
|
---|
| 194 | addCustomQueryParams(param_list, lang);
|
---|
| 195 | addStandardQueryParams(param_list, lang);
|
---|
| 196 | tq_service.appendChild(param_list);
|
---|
| 197 | }
|
---|
| 198 | return tq_service;
|
---|
| 199 |
|
---|
[9002] | 200 | }
|
---|
| 201 |
|
---|
[24858] | 202 | // perhaps these should be changed to search down the class hierarchy for
|
---|
| 203 | // values - do can just put the info in the resource bundle to use it
|
---|
| 204 | /** returns the default name for the TextQuery service */
|
---|
| 205 | protected String getServiceName(String service_id, String lang)
|
---|
| 206 | {
|
---|
| 207 | return getTextString(service_id + ".name", lang);
|
---|
[9002] | 208 | }
|
---|
[24858] | 209 |
|
---|
| 210 | /** returns the default description for the TextQuery service */
|
---|
| 211 | protected String getServiceDescription(String service_id, String lang)
|
---|
| 212 | {
|
---|
| 213 | return getTextString(service_id + ".description", lang);
|
---|
[9002] | 214 | }
|
---|
[9281] | 215 |
|
---|
[24858] | 216 | /** returns the default submit button text for the TextQuery service */
|
---|
| 217 | protected String getServiceSubmit(String service_id, String lang)
|
---|
| 218 | {
|
---|
| 219 | return getTextString(service_id + ".submit", lang);
|
---|
[9281] | 220 |
|
---|
[24858] | 221 | }
|
---|
[9281] | 222 |
|
---|
[24858] | 223 | /** adds the standard query params into the service description */
|
---|
| 224 | protected void addStandardQueryParams(Element param_list, String lang)
|
---|
| 225 | {
|
---|
| 226 | // this test is not so good. here we are using absence of default index
|
---|
| 227 | // to determine whether we have indexes or not. But in other places,
|
---|
| 228 | // absence of default index just means to use the first one as default.
|
---|
| 229 | if (!default_index.equals(""))
|
---|
| 230 | {
|
---|
| 231 | createParameter(INDEX_PARAM, param_list, lang);
|
---|
| 232 | }
|
---|
| 233 | if (does_chunking)
|
---|
| 234 | {
|
---|
| 235 | createParameter(MAXDOCS_PARAM, param_list, lang);
|
---|
| 236 | }
|
---|
| 237 | if (does_paging)
|
---|
| 238 | {
|
---|
| 239 | createParameter(HITS_PER_PAGE_PARAM, param_list, lang);
|
---|
| 240 | createParameter(START_PAGE_PARAM, param_list, lang);
|
---|
| 241 | }
|
---|
| 242 | createParameter(QUERY_PARAM, param_list, lang);
|
---|
[14002] | 243 | }
|
---|
[24858] | 244 |
|
---|
| 245 | /**
|
---|
| 246 | * adds any service specific query params into the service default
|
---|
| 247 | * implementation: add nothing. subclasses may need to override this to add
|
---|
| 248 | * in their specific parameters
|
---|
| 249 | */
|
---|
| 250 | protected void addCustomQueryParams(Element param_list, String lang)
|
---|
| 251 | {
|
---|
| 252 | // default behaviour, do nothing
|
---|
[9266] | 253 | }
|
---|
[24858] | 254 |
|
---|
| 255 | protected void createParameter(String name, Element param_list, String lang)
|
---|
| 256 | {
|
---|
| 257 | createParameter(name, param_list, lang, null);
|
---|
[9002] | 258 | }
|
---|
| 259 |
|
---|
[24858] | 260 | protected void createParameter(String name, Element param_list, String lang, String default_value)
|
---|
| 261 | {
|
---|
| 262 | // at this level, not interested in boolean return type
|
---|
| 263 | createParameterChain(name, param_list, lang, default_value);
|
---|
| 264 | }
|
---|
[9002] | 265 |
|
---|
[24858] | 266 | /**
|
---|
| 267 | * default implementations for the standard parameters plus some other
|
---|
| 268 | * common ones index, maxDocs, hitsPerPage, startPage
|
---|
| 269 | */
|
---|
[24394] | 270 |
|
---|
[24858] | 271 | protected boolean createParameterChain(String name, Element param_list, String lang, String default_value)
|
---|
| 272 | {
|
---|
| 273 | Element param = null;
|
---|
| 274 | String param_default = default_value;
|
---|
[24394] | 275 |
|
---|
[24892] | 276 | if (name.equals(QUERY_PARAM) || name.equals(RAW_PARAM))
|
---|
[24858] | 277 | {
|
---|
[24892] | 278 | param = GSXML.createParameterDescription(this.doc, name, getTextString("param." + name, lang), GSXML.PARAM_TYPE_STRING, param_default, null, null);
|
---|
[24858] | 279 | param_list.appendChild(param);
|
---|
| 280 | return true;
|
---|
| 281 | }
|
---|
| 282 | else if (name.equals(INDEX_PARAM))
|
---|
| 283 | {
|
---|
| 284 | // should we make these class fields?
|
---|
[25635] | 285 | ArrayList<String> index_ids = new ArrayList<String>();
|
---|
| 286 | ArrayList<String> index_names = new ArrayList<String>();
|
---|
[24858] | 287 | getIndexData(index_ids, index_names, lang);
|
---|
| 288 | String param_type = GSXML.PARAM_TYPE_ENUM_SINGLE;
|
---|
| 289 | if (does_multi_index_search)
|
---|
| 290 | {
|
---|
| 291 | param_type = GSXML.PARAM_TYPE_ENUM_MULTI;
|
---|
| 292 | }
|
---|
| 293 | if (param_default == null)
|
---|
| 294 | {
|
---|
| 295 | param_default = this.default_index;
|
---|
| 296 | }
|
---|
| 297 | param = GSXML.createParameterDescription2(this.doc, INDEX_PARAM, getTextString("param." + INDEX_PARAM, lang), param_type, param_default, index_ids, index_names);
|
---|
| 298 | param_list.appendChild(param);
|
---|
| 299 | return true;
|
---|
| 300 | }
|
---|
| 301 | else if (name.equals(MAXDOCS_PARAM))
|
---|
| 302 | {
|
---|
| 303 | if (param_default == null)
|
---|
| 304 | {
|
---|
| 305 | param_default = this.default_max_docs;
|
---|
| 306 | }
|
---|
[24394] | 307 |
|
---|
[24858] | 308 | param = GSXML.createParameterDescription(this.doc, name, getTextString("param." + name, lang), GSXML.PARAM_TYPE_INTEGER, param_default, null, null);
|
---|
| 309 | param_list.appendChild(param);
|
---|
| 310 | return true;
|
---|
| 311 | }
|
---|
| 312 | else if (name.equals(HITS_PER_PAGE_PARAM))
|
---|
| 313 | {
|
---|
| 314 | if (param_default == null)
|
---|
| 315 | {
|
---|
| 316 | param_default = this.default_hits_per_page;
|
---|
| 317 | }
|
---|
[9002] | 318 |
|
---|
[24858] | 319 | param = GSXML.createParameterDescription(this.doc, name, getTextString("param." + name, lang), GSXML.PARAM_TYPE_INTEGER, param_default, null, null);
|
---|
| 320 | param_list.appendChild(param);
|
---|
| 321 | return true;
|
---|
| 322 | }
|
---|
| 323 | else if (name.equals(START_PAGE_PARAM))
|
---|
| 324 | {
|
---|
| 325 | if (param_default == null)
|
---|
| 326 | {
|
---|
| 327 | param_default = "1";
|
---|
| 328 | }
|
---|
| 329 |
|
---|
| 330 | // start page - set to 1 for the search page
|
---|
| 331 | param = GSXML.createParameterDescription(this.doc, START_PAGE_PARAM, "", GSXML.PARAM_TYPE_INVISIBLE, param_default, null, null);
|
---|
| 332 | param_list.appendChild(param);
|
---|
| 333 | return true;
|
---|
| 334 | }
|
---|
| 335 |
|
---|
| 336 | // Get to there then none of the above params matched
|
---|
| 337 | // => return false so the chain can continue
|
---|
| 338 | return false;
|
---|
[13998] | 339 | }
|
---|
[24858] | 340 |
|
---|
| 341 | /**
|
---|
| 342 | * create an element to go into the search results list. A node element has
|
---|
| 343 | * the form <docNode nodeId='xxx' nodeType='leaf' docType='hierarchy'
|
---|
| 344 | * rank='0.23'/>
|
---|
| 345 | */
|
---|
| 346 | protected Element createDocNode(String node_id, String rank)
|
---|
| 347 | {
|
---|
| 348 | return this.gs_doc.createDocNode(node_id, rank);
|
---|
[14945] | 349 | }
|
---|
[24858] | 350 |
|
---|
| 351 | /**
|
---|
| 352 | * returns the document type of the doc that the specified node belongs to.
|
---|
| 353 | * should be one of GSXML.DOC_TYPE_SIMPLE, GSXML.DOC_TYPE_PAGED,
|
---|
| 354 | * GSXML.DOC_TYPE_HIERARCHY
|
---|
| 355 | */
|
---|
| 356 | protected String getDocType(String node_id)
|
---|
| 357 | {
|
---|
| 358 | return this.gs_doc.getDocType(node_id);
|
---|
[9002] | 359 | }
|
---|
[24394] | 360 |
|
---|
[24858] | 361 | /**
|
---|
| 362 | * returns the node type of the specified node. should be one of
|
---|
| 363 | * GSXML.NODE_TYPE_LEAF, GSXML.NODE_TYPE_INTERNAL, GSXML.NODE_TYPE_ROOT
|
---|
| 364 | */
|
---|
| 365 | protected String getNodeType(String node_id, String doc_type)
|
---|
| 366 | {
|
---|
| 367 | return this.gs_doc.getNodeType(node_id, doc_type);
|
---|
| 368 | }
|
---|
[24394] | 369 |
|
---|
[24858] | 370 | /** returns true if the node has child nodes */
|
---|
| 371 | protected boolean hasChildren(String node_id)
|
---|
| 372 | {
|
---|
| 373 | return this.gs_doc.hasChildren(node_id);
|
---|
| 374 | }
|
---|
[24394] | 375 |
|
---|
[24858] | 376 | /** returns true if the node has a parent */
|
---|
| 377 | protected boolean hasParent(String node_id)
|
---|
| 378 | {
|
---|
| 379 | return this.gs_doc.hasParent(node_id);
|
---|
| 380 | }
|
---|
[9002] | 381 |
|
---|
[24858] | 382 | /**
|
---|
| 383 | * get the details about the indexes available must be implemented by
|
---|
| 384 | * subclass there must be at least one index
|
---|
| 385 | */
|
---|
[25635] | 386 | abstract protected void getIndexData(ArrayList<String> index_ids, ArrayList<String> index_names, String lang);
|
---|
[24394] | 387 |
|
---|
[9266] | 388 | }
|
---|