[2491] | 1 | /**********************************************************************
|
---|
| 2 | *
|
---|
| 3 | * phindaction.cpp --
|
---|
| 4 | *
|
---|
| 5 | * Copyright 2001 Gordon W. Paynter
|
---|
| 6 | * Copyright 2001 The New Zealand Digital Library Project
|
---|
| 7 | *
|
---|
| 8 | * A component of the Greenstone digital library software
|
---|
| 9 | * from the New Zealand Digital Library Project at the
|
---|
| 10 | * University of Waikato, New Zealand.
|
---|
| 11 | *
|
---|
| 12 | * This program is free software; you can redistribute it and/or modify
|
---|
| 13 | * it under the terms of the GNU General Public License as published by
|
---|
| 14 | * the Free Software Foundation; either version 2 of the License, or
|
---|
| 15 | * (at your option) any later version.
|
---|
| 16 | *
|
---|
| 17 | * This program is distributed in the hope that it will be useful,
|
---|
| 18 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 19 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
| 20 | * GNU General Public License for more details.
|
---|
| 21 | *
|
---|
| 22 | * You should have received a copy of the GNU General Public License
|
---|
| 23 | * along with this program; if not, write to the Free Software
|
---|
| 24 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
| 25 | *
|
---|
| 26 | *********************************************************************/
|
---|
| 27 |
|
---|
[7371] | 28 | #include "gsdl_modules_cfg.h"
|
---|
| 29 | #ifdef GSDL_USE_PHIND_ACTION
|
---|
| 30 |
|
---|
[2542] | 31 | // Note that this action uses mgpp to retrieve phind info, calling MGQuery
|
---|
| 32 | // etc. directly, not through the protocol. This breaks our receptionist -
|
---|
| 33 | // collection server separation and should be fixed some day I guess.
|
---|
| 34 |
|
---|
[2491] | 35 | #include "phindaction.h"
|
---|
| 36 | #include "fileutil.h"
|
---|
| 37 |
|
---|
| 38 | phindaction::phindaction () {
|
---|
| 39 |
|
---|
| 40 | cgiarginfo arg_ainfo;
|
---|
| 41 |
|
---|
| 42 | arg_ainfo.shortname = "pc";
|
---|
| 43 | arg_ainfo.longname = "phind classifier";
|
---|
| 44 | arg_ainfo.multiplechar = true;
|
---|
| 45 | arg_ainfo.defaultstatus = cgiarginfo::weak;
|
---|
[7384] | 46 | arg_ainfo.argdefault = g_EmptyText;
|
---|
[2491] | 47 | arg_ainfo.savedarginfo = cgiarginfo::mustnot;
|
---|
| 48 | argsinfo.addarginfo (NULL, arg_ainfo);
|
---|
| 49 |
|
---|
| 50 | arg_ainfo.shortname = "pxml";
|
---|
| 51 | arg_ainfo.longname = "phind XML mode";
|
---|
| 52 | arg_ainfo.multiplechar = false;
|
---|
| 53 | arg_ainfo.defaultstatus = cgiarginfo::weak;
|
---|
| 54 | arg_ainfo.argdefault = "0";
|
---|
| 55 | arg_ainfo.savedarginfo = cgiarginfo::mustnot;
|
---|
| 56 | argsinfo.addarginfo (NULL, arg_ainfo);
|
---|
| 57 |
|
---|
| 58 | arg_ainfo.shortname = "ppnum";
|
---|
| 59 | arg_ainfo.longname = "phind phrase number";
|
---|
| 60 | arg_ainfo.multiplechar = true;
|
---|
| 61 | arg_ainfo.defaultstatus = cgiarginfo::weak;
|
---|
| 62 | arg_ainfo.argdefault = "0";
|
---|
| 63 | arg_ainfo.savedarginfo = cgiarginfo::mustnot;
|
---|
| 64 | argsinfo.addarginfo (NULL, arg_ainfo);
|
---|
| 65 |
|
---|
| 66 | arg_ainfo.shortname = "pptext";
|
---|
| 67 | arg_ainfo.longname = "phind phrase text";
|
---|
| 68 | arg_ainfo.multiplechar = true;
|
---|
| 69 | arg_ainfo.defaultstatus = cgiarginfo::weak;
|
---|
[7384] | 70 | arg_ainfo.argdefault = g_EmptyText;
|
---|
[2491] | 71 | arg_ainfo.savedarginfo = cgiarginfo::mustnot;
|
---|
| 72 | argsinfo.addarginfo (NULL, arg_ainfo);
|
---|
| 73 |
|
---|
| 74 | arg_ainfo.shortname = "pfe";
|
---|
| 75 | arg_ainfo.longname = "phind first_e";
|
---|
| 76 | arg_ainfo.multiplechar = true;
|
---|
| 77 | arg_ainfo.defaultstatus = cgiarginfo::weak;
|
---|
| 78 | arg_ainfo.argdefault = "0";
|
---|
| 79 | arg_ainfo.savedarginfo = cgiarginfo::mustnot;
|
---|
| 80 | argsinfo.addarginfo (NULL, arg_ainfo);
|
---|
| 81 |
|
---|
| 82 | arg_ainfo.shortname = "ple";
|
---|
| 83 | arg_ainfo.longname = "phind last_e";
|
---|
| 84 | arg_ainfo.multiplechar = true;
|
---|
| 85 | arg_ainfo.defaultstatus = cgiarginfo::weak;
|
---|
| 86 | arg_ainfo.argdefault = "10";
|
---|
| 87 | arg_ainfo.savedarginfo = cgiarginfo::mustnot;
|
---|
| 88 | argsinfo.addarginfo (NULL, arg_ainfo);
|
---|
| 89 |
|
---|
| 90 | arg_ainfo.shortname = "pfl";
|
---|
| 91 | arg_ainfo.longname = "phind first_l";
|
---|
| 92 | arg_ainfo.multiplechar = true;
|
---|
| 93 | arg_ainfo.defaultstatus = cgiarginfo::weak;
|
---|
| 94 | arg_ainfo.argdefault = "0";
|
---|
| 95 | arg_ainfo.savedarginfo = cgiarginfo::mustnot;
|
---|
| 96 | argsinfo.addarginfo (NULL, arg_ainfo);
|
---|
| 97 |
|
---|
| 98 | arg_ainfo.shortname = "pll";
|
---|
| 99 | arg_ainfo.longname = "phind last_l";
|
---|
| 100 | arg_ainfo.multiplechar = true;
|
---|
| 101 | arg_ainfo.defaultstatus = cgiarginfo::weak;
|
---|
| 102 | arg_ainfo.argdefault = "10";
|
---|
| 103 | arg_ainfo.savedarginfo = cgiarginfo::mustnot;
|
---|
| 104 | argsinfo.addarginfo (NULL, arg_ainfo);
|
---|
| 105 |
|
---|
| 106 | arg_ainfo.shortname = "pfd";
|
---|
| 107 | arg_ainfo.longname = "phind first_d";
|
---|
| 108 | arg_ainfo.multiplechar = true;
|
---|
| 109 | arg_ainfo.defaultstatus = cgiarginfo::weak;
|
---|
| 110 | arg_ainfo.argdefault = "0";
|
---|
| 111 | arg_ainfo.savedarginfo = cgiarginfo::mustnot;
|
---|
| 112 | argsinfo.addarginfo (NULL, arg_ainfo);
|
---|
| 113 |
|
---|
| 114 | arg_ainfo.shortname = "pld";
|
---|
| 115 | arg_ainfo.longname = "phind last_d";
|
---|
| 116 | arg_ainfo.multiplechar = true;
|
---|
| 117 | arg_ainfo.defaultstatus = cgiarginfo::weak;
|
---|
| 118 | arg_ainfo.argdefault = "10";
|
---|
| 119 | arg_ainfo.savedarginfo = cgiarginfo::mustnot;
|
---|
| 120 | argsinfo.addarginfo (NULL, arg_ainfo);
|
---|
| 121 | }
|
---|
| 122 |
|
---|
| 123 | phindaction::~phindaction () {
|
---|
| 124 | }
|
---|
| 125 |
|
---|
[6787] | 126 | void phindaction::get_cgihead_info (cgiargsclass &args, recptprotolistclass * /*protos*/,
|
---|
[2491] | 127 | response_t &response,text_t &response_data,
|
---|
| 128 | ostream &/*logout*/) {
|
---|
| 129 | response = content;
|
---|
[6787] | 130 | if (args["pxml"] == "1") {
|
---|
| 131 | response_data = "text/xml";
|
---|
| 132 | } else {
|
---|
| 133 | response_data = "text/html";
|
---|
| 134 | }
|
---|
[2491] | 135 | }
|
---|
| 136 |
|
---|
| 137 | bool phindaction::do_action (cgiargsclass &args, recptprotolistclass *protos,
|
---|
| 138 | browsermapclass * /*browsers*/, displayclass &disp,
|
---|
| 139 | outconvertclass &outconvert, ostream &textout,
|
---|
| 140 | ostream &logout) {
|
---|
| 141 |
|
---|
| 142 | unsigned long count_l, count_e, count_d;
|
---|
| 143 | unsigned long phrase = args["ppnum"].getulong();
|
---|
| 144 | text_t &word = args["pptext"];
|
---|
| 145 | unsigned long first_e = args["pfe"].getulong();
|
---|
| 146 | unsigned long last_e = args["ple"].getulong();
|
---|
| 147 | unsigned long first_l = args["pfl"].getulong();
|
---|
| 148 | unsigned long last_l = args["pll"].getulong();
|
---|
| 149 | unsigned long first_d = args["pfd"].getulong();
|
---|
| 150 | unsigned long last_d = args["pld"].getulong();
|
---|
| 151 | bool XMLmode = false;
|
---|
| 152 | if (args["pxml"] == "1") XMLmode = true;
|
---|
| 153 |
|
---|
[2502] | 154 | // must have a valid collection server
|
---|
| 155 | recptproto *collectproto = protos->getrecptproto (args["c"], logout);
|
---|
| 156 | if (collectproto == NULL) {
|
---|
| 157 | output_error("phindaction: ERROR: collection not set", textout,
|
---|
| 158 | outconvert, disp, logout, XMLmode);
|
---|
| 159 | return true;
|
---|
| 160 | }
|
---|
[2491] | 161 |
|
---|
| 162 | // the frequency and occurances of the phrase
|
---|
| 163 | unsigned long tf;
|
---|
| 164 | vector <unsigned long> el, linkdest, docNums, docfreq;
|
---|
| 165 | vector <UCArray> linktype;
|
---|
| 166 |
|
---|
| 167 | // the number of occurances to display
|
---|
| 168 | unsigned long ef, lf, df;
|
---|
| 169 |
|
---|
[16310] | 170 | text_t basepath = filename_cat(collecthome, args["c"],
|
---|
[2491] | 171 | "index", "phind" + args["pc"]);
|
---|
| 172 |
|
---|
| 173 | // If we don't know the phrase number, look it up
|
---|
| 174 | if (phrase == 0) {
|
---|
| 175 |
|
---|
| 176 | if (word.empty()) {
|
---|
[2502] | 177 | output_error("phindaction: ERROR: no phrase number or word", textout,
|
---|
| 178 | outconvert, disp, logout, XMLmode);
|
---|
[2491] | 179 | return true;
|
---|
| 180 | }
|
---|
| 181 |
|
---|
| 182 | DocNumArray result;
|
---|
[7734] | 183 | /** In order to prevent browser crashing problems, any method which
|
---|
| 184 | * previously suffered a silent fatal error, now instead returns false
|
---|
| 185 | * to indicate a fatal error has occured. We can then dispatch an
|
---|
| 186 | * appropriate error tag to the Phind applet (rather than leave it
|
---|
| 187 | * whiling away the milliseconds until the end of existence - or at
|
---|
| 188 | * least your browser - in an infinite loop!)
|
---|
| 189 | * DLConsulting 12-07-2004
|
---|
| 190 | */
|
---|
| 191 | if(!find_phrase_number_from_word(basepath, word, result)) {
|
---|
| 192 | output_error("phindaction: Fatal Error! Couldn't load index information in find_phrase_number_from_word()",
|
---|
| 193 | textout, outconvert, disp, logout, XMLmode);
|
---|
| 194 | return true;
|
---|
| 195 | }
|
---|
[2491] | 196 |
|
---|
| 197 | if (result.empty()) {
|
---|
[2502] | 198 | output_error("phindaction: The search term does not occur in the collection",
|
---|
| 199 | textout, outconvert, disp, logout, XMLmode);
|
---|
[2491] | 200 | return true;
|
---|
| 201 | } else {
|
---|
| 202 | phrase = result[0];
|
---|
| 203 | }
|
---|
| 204 | }
|
---|
| 205 |
|
---|
| 206 | // Create a TextData object to read the phrase data (pdata)
|
---|
| 207 | TextData textdata;
|
---|
| 208 |
|
---|
| 209 | text_t fullpath = filename_cat(basepath, "pdata");
|
---|
| 210 | char *fullpathc = fullpath.getcstr();
|
---|
| 211 | #if defined __WIN32__
|
---|
| 212 | char *base = "";
|
---|
| 213 | #else
|
---|
| 214 | char *base = "/";
|
---|
| 215 | #endif
|
---|
| 216 |
|
---|
| 217 | if (!textdata.LoadData (base, fullpathc)) {
|
---|
| 218 | // FatalError (1, "Couldn't load text information for \"%s\"", fullpathc);
|
---|
[7734] | 219 | //exit(0);
|
---|
| 220 | /** We must return something to the client, whether this error is fatal or
|
---|
| 221 | * no, otherwise we risk sending their browser into an infinite loop!
|
---|
| 222 | * DLConsulting 12-07-2004
|
---|
| 223 | */
|
---|
| 224 | output_error("phindaction: Fatal Error! Couldn't load text information for collection",
|
---|
| 225 | textout, outconvert, disp, logout, XMLmode);
|
---|
| 226 | return true;
|
---|
[2491] | 227 | }
|
---|
| 228 |
|
---|
[7384] | 229 | delete []fullpathc;
|
---|
[2491] | 230 |
|
---|
[7734] | 231 | /** Another previously silent method can now cry out.
|
---|
| 232 | * DLConsulting 12-07-2004
|
---|
| 233 | */
|
---|
| 234 | if(!get_phrase_all_data(textdata, phrase, word, tf, ef, lf, df, el,
|
---|
| 235 | linkdest, linktype, docNums, docfreq)) {
|
---|
| 236 | output_error(
|
---|
| 237 | "phindaction: Fatal Error! Couldn't parse phrase in get_phrase_all_data()",
|
---|
| 238 | textout, outconvert, disp, logout, XMLmode);
|
---|
| 239 | return true;
|
---|
| 240 | }
|
---|
[2491] | 241 |
|
---|
| 242 | // Output the header
|
---|
| 243 | if (XMLmode) {
|
---|
[6787] | 244 | textout << "<phinddata id=\"" << phrase
|
---|
[2491] | 245 | << "\" text=\"" << word
|
---|
| 246 | << "\" tf=\"" << tf
|
---|
| 247 | << "\" ef=\"" << ef
|
---|
| 248 | << "\" df=\"" << df
|
---|
| 249 | << "\" lf=\"" << lf
|
---|
| 250 | << "\">\n";
|
---|
| 251 | } else {
|
---|
[6787] | 252 | textout << "<html><head><title>" << word << "</title></head>\n"
|
---|
[2491] | 253 | << "<body><center>\n"
|
---|
| 254 | << "<p><h1>" << word << "</h1>\n"
|
---|
| 255 | << "<p><b>"<< word << "</b> occurs "
|
---|
| 256 | << tf << " times in " << df << " documents\n";
|
---|
| 257 | }
|
---|
| 258 |
|
---|
| 259 | // Output the thesaurus links
|
---|
| 260 | if ((lf > 0) && (first_l < last_l)) {
|
---|
| 261 |
|
---|
| 262 | // figure out the number of phrases to output
|
---|
| 263 | if (last_l > lf) {
|
---|
| 264 | last_l = lf;
|
---|
| 265 | }
|
---|
| 266 | count_l = last_l - first_l;
|
---|
| 267 |
|
---|
| 268 | if (XMLmode) {
|
---|
| 269 | textout << "<thesauruslist length=\"" << lf
|
---|
| 270 | << "\" start=\"" << first_l
|
---|
| 271 | << "\" end=\"" << last_l << "\">\n";
|
---|
[7734] | 272 | /** DLConsulting 12-07-2004 */
|
---|
| 273 | if(!print_thesaurus_links(args["c"], XMLmode, textdata, linkdest, linktype,
|
---|
| 274 | first_l, last_l, disp, outconvert, textout)) {
|
---|
| 275 | output_error(
|
---|
| 276 | "phindaction: Fatal Error! Couldn't get phrase in get_phrase_freq_data()",
|
---|
| 277 | textout, outconvert, disp, logout, XMLmode);
|
---|
| 278 | return true;
|
---|
| 279 | }
|
---|
[2491] | 280 | textout << "</thesauruslist>\n";
|
---|
| 281 | }
|
---|
| 282 |
|
---|
| 283 | // output links as HTML
|
---|
| 284 | else {
|
---|
| 285 | if (count_l == lf) {
|
---|
| 286 | textout << "<p><b> " << count_l << " thesaurus links</b>\n";
|
---|
| 287 | } else {
|
---|
| 288 | textout << "<p><b>" << count_l << " of " << lf << " thesaurus links</b>\n";
|
---|
| 289 | }
|
---|
| 290 |
|
---|
| 291 | textout << "<p><table border=1><tr><th>type</th><th>topic</th><th>freq</th><th>docs</th></tr>\n";
|
---|
[7734] | 292 | /** DLConsulting 12-07-2004 */
|
---|
| 293 | if(!print_thesaurus_links(args["c"], XMLmode, textdata, linkdest, linktype,
|
---|
| 294 | first_l, last_l, disp, outconvert, textout)) {
|
---|
| 295 | output_error(
|
---|
| 296 | "phindaction: Fatal Error! Couldn't get phrase in get_phrase_freq_data()",
|
---|
| 297 | textout, outconvert, disp, logout, XMLmode);
|
---|
| 298 | return true;
|
---|
| 299 | }
|
---|
[2491] | 300 | textout << "</table>\n";
|
---|
| 301 |
|
---|
| 302 | if (last_l < lf) {
|
---|
| 303 | if ((last_l + 10) < lf) {
|
---|
| 304 | textout << outconvert << disp
|
---|
| 305 | << "<br><a href=\"_gwcgi_?"
|
---|
| 306 | << "c=" << args["c"]
|
---|
| 307 | << "&ppnum=" << phrase
|
---|
| 308 | << "&pfe=" << first_e
|
---|
| 309 | << "&ple=" << last_e
|
---|
| 310 | << "&pfd=" << first_d
|
---|
| 311 | << "&pld=" << last_d
|
---|
| 312 | << "&pfl=" << first_l
|
---|
| 313 | << "&pll=" << (last_l + 10)
|
---|
| 314 | << "\">Get more thesaurus links</a>\n";
|
---|
| 315 | }
|
---|
| 316 | textout << outconvert << disp
|
---|
| 317 | << "<br><a href=\"_gwcgi_?"
|
---|
| 318 | << "c=" << args["c"]
|
---|
| 319 | << "&ppnum=" << phrase
|
---|
| 320 | << "&pfe=" << first_e
|
---|
| 321 | << "&ple=" << last_e
|
---|
| 322 | << "&pfd=" << first_d
|
---|
| 323 | << "&pld=" << last_d
|
---|
| 324 | << "&pfl=" << first_l
|
---|
| 325 | << "&pll=" << lf
|
---|
| 326 | << "\">Get every thesaurus link</a>\n" ;
|
---|
| 327 | }
|
---|
| 328 | }
|
---|
| 329 | }
|
---|
| 330 |
|
---|
| 331 | // Output the expansions
|
---|
| 332 | if ((ef > 0) && (first_e < last_e)) {
|
---|
| 333 |
|
---|
| 334 | // figure out the number of phrases to output
|
---|
| 335 | if (last_e > el.size()) {
|
---|
| 336 | last_e = el.size();
|
---|
| 337 | }
|
---|
| 338 | count_e = last_e - first_e;
|
---|
| 339 |
|
---|
| 340 | // output expansions as XML
|
---|
| 341 | if (XMLmode) {
|
---|
| 342 | textout << "<expansionlist length=\"" << ef
|
---|
| 343 | << "\" start=\"" << first_e
|
---|
| 344 | << "\" end=\"" << last_e << "\">" << endl;
|
---|
| 345 |
|
---|
| 346 | print_expansions(args["c"], XMLmode, word, textdata, el, first_e,
|
---|
| 347 | last_e, disp, outconvert, textout);
|
---|
| 348 |
|
---|
| 349 | textout << "</expansionlist>\n";
|
---|
| 350 | }
|
---|
| 351 |
|
---|
| 352 | // output expansions as HTML
|
---|
| 353 | else {
|
---|
| 354 | if (count_e == el.size()) {
|
---|
| 355 | textout << "<p><b> " << count_e << " expansions</b>\n";
|
---|
| 356 | } else {
|
---|
| 357 | textout << "<p><b>" << count_e << " of " << ef << " expansions</b>\n";
|
---|
| 358 | }
|
---|
| 359 |
|
---|
| 360 | textout << "<p><table border=1><tr><th colspan=3>phrase</th><th>freq</th><th>docs</th></tr>\n";
|
---|
| 361 | print_expansions(args["c"], XMLmode, word, textdata, el, first_e,
|
---|
| 362 | last_e, disp, outconvert, textout);
|
---|
| 363 | textout << "</table>\n";
|
---|
| 364 |
|
---|
| 365 | if (last_e < ef) {
|
---|
| 366 | if ((last_e + 10) < ef) {
|
---|
| 367 | textout << outconvert << disp
|
---|
| 368 | << "<br><a href=\"_gwcgi_?"
|
---|
| 369 | << "c=" << args["c"]
|
---|
| 370 | << "&ppnum=" << phrase
|
---|
| 371 | << "&pfe=" << first_e
|
---|
| 372 | << "&ple=" << (last_e + 10)
|
---|
| 373 | << "&pfd=" << first_d
|
---|
| 374 | << "&pld=" << last_d
|
---|
| 375 | << "&pfl=" << first_l
|
---|
| 376 | << "&pll=" << last_l
|
---|
| 377 | << "\">Get more expansions</a>\n";
|
---|
| 378 | }
|
---|
| 379 | textout << outconvert << disp
|
---|
| 380 | << "<br><a href=\"_gwcgi_?"
|
---|
| 381 | << "c=" << args["c"]
|
---|
| 382 | << "&ppnum=" << phrase
|
---|
| 383 | << "&pfe=" << first_e
|
---|
| 384 | << "&ple=" << ef
|
---|
| 385 | << "&pfd=" << first_d
|
---|
| 386 | << "&pld=" << last_d
|
---|
| 387 | << "&pfl=" << first_l
|
---|
| 388 | << "&pll=" << last_l
|
---|
| 389 | << "\">Get every expansion</a>\n";
|
---|
| 390 | }
|
---|
| 391 | }
|
---|
| 392 | }
|
---|
| 393 |
|
---|
| 394 | // Output the document occurances
|
---|
| 395 | if ((df > 0) && (first_d < last_d)) {
|
---|
| 396 |
|
---|
| 397 | // figure out the phrases to output
|
---|
| 398 | if (last_d > docNums.size()) {
|
---|
| 399 | last_d = docNums.size();
|
---|
| 400 | }
|
---|
| 401 | count_d = last_d - first_d;
|
---|
| 402 |
|
---|
| 403 | // output document list as XML
|
---|
| 404 | if (XMLmode) {
|
---|
| 405 | textout << "<documentlist length=\"" << df
|
---|
| 406 | << "\" start=\"" << first_d
|
---|
| 407 | << "\" end=\"" << last_d << "\">\n";
|
---|
| 408 |
|
---|
[7734] | 409 | if(!print_documents(XMLmode, basepath, args["c"], docNums, docfreq,
|
---|
| 410 | first_d, last_d, disp, outconvert, textout)) {
|
---|
| 411 | output_error(
|
---|
| 412 | "phindaction: Fatal Error! Couldn't load text information in print_documents() or get_document_all_data()",
|
---|
| 413 | textout, outconvert, disp, logout, XMLmode);
|
---|
| 414 | return true;
|
---|
| 415 | }
|
---|
[2491] | 416 |
|
---|
| 417 | textout << "</documentlist>\n";
|
---|
| 418 | }
|
---|
| 419 |
|
---|
| 420 | // output document list as HTML
|
---|
| 421 | else {
|
---|
| 422 |
|
---|
| 423 | if (count_d == docNums.size()) {
|
---|
| 424 | textout << "<p><b> " << count_d << " documents</b>\n";
|
---|
| 425 | } else {
|
---|
| 426 | textout << "<p><b>" << count_d << " of " << df << " documents</b>\n";
|
---|
| 427 | }
|
---|
| 428 |
|
---|
| 429 | textout << "<p><table border=1><tr><th align=left>document</th><th>freq</th></tr>\n";
|
---|
[7734] | 430 | if(!print_documents(XMLmode, basepath, args["c"], docNums, docfreq,
|
---|
| 431 | first_d, last_d, disp, outconvert, textout)) {
|
---|
| 432 | output_error(
|
---|
| 433 | "phindaction: Fatal Error! Couldn't load text information in print_documents()",
|
---|
| 434 | textout, outconvert, disp, logout, XMLmode);
|
---|
| 435 | return true;
|
---|
| 436 | }
|
---|
[2491] | 437 | textout << "</table>\n";
|
---|
| 438 |
|
---|
| 439 | if (last_d < df) {
|
---|
| 440 | if ((last_d + 10) < df) {
|
---|
| 441 | textout << outconvert << disp
|
---|
| 442 | << "<br><a href=\"_gwcgi_?"
|
---|
| 443 | << "c=" << args["c"]
|
---|
| 444 | << "&ppnum=" << phrase
|
---|
| 445 | << "&pfe=" << first_e
|
---|
| 446 | << "&ple=" << last_e
|
---|
| 447 | << "&pfd=" << first_d
|
---|
| 448 | << "&pld=" << (last_d + 10)
|
---|
| 449 | << "&pfl=" << first_l
|
---|
| 450 | << "&pll=" << last_l
|
---|
| 451 | << "\">Get more documents</a>\n";
|
---|
| 452 | }
|
---|
| 453 | textout << outconvert << disp
|
---|
| 454 | << "<br><a href=\"_gwcgi_?"
|
---|
| 455 | << "c=" << args["c"]
|
---|
| 456 | << "&ppnum=" << phrase
|
---|
| 457 | << "&pfe=" << first_e
|
---|
| 458 | << "&ple=" << last_e
|
---|
| 459 | << "&pfd=" << first_d
|
---|
| 460 | << "&pld=" << df
|
---|
| 461 | << "&pfl=" << first_l
|
---|
| 462 | << "&pll=" << last_l
|
---|
| 463 | << "\">Get every document</a>\n";
|
---|
| 464 | }
|
---|
| 465 | }
|
---|
| 466 | }
|
---|
| 467 |
|
---|
| 468 | // Close the document
|
---|
| 469 | if (XMLmode) {
|
---|
| 470 | textout << "</phinddata>\n";
|
---|
| 471 | } else {
|
---|
| 472 | textout << "</center></body></html>\n";
|
---|
| 473 | }
|
---|
| 474 |
|
---|
| 475 | textdata.UnloadData ();
|
---|
| 476 |
|
---|
| 477 | return true;
|
---|
| 478 | }
|
---|
| 479 |
|
---|
| 480 | // Find the phrase number of a word in the index file
|
---|
[7734] | 481 | bool phindaction::find_phrase_number_from_word(const text_t &basepath,
|
---|
[2491] | 482 | const text_t &query,
|
---|
| 483 | DocNumArray &result) {
|
---|
| 484 |
|
---|
| 485 | // Open the index file for searching
|
---|
| 486 | IndexData indexData;
|
---|
| 487 |
|
---|
| 488 | text_t fullpath = filename_cat(basepath, "pword");
|
---|
| 489 | char *fullpathc = fullpath.getcstr();
|
---|
| 490 | #if defined __WIN32__
|
---|
| 491 | char *base = "";
|
---|
| 492 | #else
|
---|
| 493 | char *base = "/";
|
---|
| 494 | #endif
|
---|
| 495 |
|
---|
| 496 | if (!indexData.LoadData (base, fullpathc)) {
|
---|
| 497 | // FatalError (1, "Couldn't load index information for \"%s\"", fullpathc);
|
---|
[7734] | 498 | //exit(0);
|
---|
| 499 | /** Don't handle fatal errors here anymore.
|
---|
| 500 | * DLConsulting 12-07-2004
|
---|
| 501 | */
|
---|
| 502 | return false; // Indicates something very bad has happened
|
---|
[2491] | 503 | }
|
---|
| 504 |
|
---|
[7384] | 505 | delete []fullpathc;
|
---|
[2491] | 506 |
|
---|
| 507 | // set up the query object
|
---|
| 508 | QueryInfo queryInfo;
|
---|
[7415] | 509 | SetCStr (queryInfo.docLevel, "Document", 8);
|
---|
[2491] | 510 | queryInfo.maxDocs = 5;
|
---|
| 511 | queryInfo.sortByRank = true;
|
---|
| 512 | queryInfo.exactWeights = false;
|
---|
| 513 | queryInfo.needRankInfo = true;
|
---|
| 514 | queryInfo.needTermFreqs = true;
|
---|
| 515 |
|
---|
| 516 | // mode 1 = casefolded, unstemmed search
|
---|
| 517 | UCArray ucquery;
|
---|
| 518 | toUCArray(query, ucquery);
|
---|
[12315] | 519 | QueryNode *queryTree = ParseQuery(ucquery, 1, 1, 4);
|
---|
[2491] | 520 |
|
---|
| 521 | // perform the query
|
---|
| 522 | ExtQueryResult queryResult;
|
---|
| 523 | MGQuery (indexData, queryInfo, queryTree, queryResult);
|
---|
| 524 | // cout << "-- word lookup result -- " << endl << queryResult << endl ;
|
---|
| 525 |
|
---|
| 526 | result.clear();
|
---|
| 527 | result = queryResult.docs;
|
---|
| 528 |
|
---|
| 529 | // delete the query
|
---|
| 530 | if (queryTree != NULL) delete queryTree;
|
---|
[2542] | 531 |
|
---|
| 532 | indexData.UnloadData();
|
---|
[7734] | 533 |
|
---|
| 534 | /** This method now returns a boolean, so...
|
---|
| 535 | * DLConsulting 12-07-2004
|
---|
| 536 | */
|
---|
| 537 | return true; // Indicates that what happened is all good, baby.
|
---|
[2491] | 538 | }
|
---|
| 539 |
|
---|
| 540 | // Get all the data about a phrase
|
---|
| 541 | //
|
---|
| 542 | // The phrase is stored in textData as record phrase.
|
---|
| 543 | // We retrieve:
|
---|
| 544 | // word - the text of the phrase
|
---|
| 545 | // tf - the total frequency of the phrase
|
---|
| 546 | // ef - the expansion frequency of the phrase
|
---|
| 547 | // lf - the thesaurus link frequency of the phrase
|
---|
| 548 | // df - the document frequency of the phrase
|
---|
| 549 | // el - the list of phrases that are expansions of phrase
|
---|
| 550 | // ll - the list of phrases that are thesaurus links
|
---|
| 551 | // dl - the list of documents that contain phrase
|
---|
[7734] | 552 | bool phindaction::get_phrase_all_data(TextData &textdata, unsigned long phrase,
|
---|
[2491] | 553 | text_t &word, unsigned long &tf, unsigned long &ef,
|
---|
| 554 | unsigned long &lf, unsigned long &df,
|
---|
| 555 | vector <unsigned long> &el,
|
---|
| 556 | vector <unsigned long> &linkdest,
|
---|
| 557 | vector <UCArray> &linktype,
|
---|
| 558 | vector <unsigned long> &docnum,
|
---|
| 559 | vector <unsigned long> &docfrq) {
|
---|
| 560 | UCArray text;
|
---|
| 561 | UCArray docLevel;
|
---|
[7415] | 562 | SetCStr(docLevel, "Document", 8);
|
---|
[2491] | 563 |
|
---|
| 564 | // Look the word up in the textData
|
---|
| 565 | if (!GetDocText (textdata, docLevel, phrase, text)) {
|
---|
| 566 | // FatalError (1, "Error while trying to get phrase %u", phrase);
|
---|
[7734] | 567 | //exit(0);
|
---|
| 568 | return false; // Something very bad has happened.
|
---|
[2491] | 569 | }
|
---|
| 570 |
|
---|
| 571 | // Ignore everything up to the first colon
|
---|
| 572 | UCArray::iterator next = text.begin();
|
---|
| 573 | while (*next++ != ':');
|
---|
| 574 |
|
---|
| 575 | // ignore training carriage returns
|
---|
| 576 | while (text.back() == '\n') {
|
---|
| 577 | text.pop_back();
|
---|
| 578 | }
|
---|
| 579 |
|
---|
| 580 | // Get the word
|
---|
| 581 | word.clear();
|
---|
[9620] | 582 | for (; *next != ':'; ++next) {
|
---|
[2491] | 583 | word.push_back(*next);
|
---|
| 584 | }
|
---|
| 585 |
|
---|
| 586 | // Get total frequency
|
---|
| 587 | tf = 0;
|
---|
[9620] | 588 | for (++next; *next != ':'; ++next) {
|
---|
[2491] | 589 | tf *= 10;
|
---|
| 590 | tf += (*next - '0');
|
---|
| 591 | }
|
---|
| 592 |
|
---|
| 593 | // Get expansion frequency
|
---|
| 594 | ef = 0;
|
---|
[9620] | 595 | for (++next; *next != ':'; ++next) {
|
---|
[2491] | 596 | ef *= 10;
|
---|
| 597 | ef += (*next - '0');
|
---|
| 598 | }
|
---|
| 599 |
|
---|
| 600 | // Get document frequency
|
---|
| 601 | df = 0;
|
---|
[9620] | 602 | for (++next; *next != ':'; ++next) {
|
---|
[2491] | 603 | df *= 10;
|
---|
| 604 | df += (*next - '0');
|
---|
| 605 | }
|
---|
| 606 |
|
---|
| 607 | // Get expansion list
|
---|
| 608 | el.clear();
|
---|
| 609 | unsigned long e = 0;
|
---|
[9620] | 610 | for (++next; *next != ':'; ++next) {
|
---|
[2491] | 611 | if (*next == ',') {
|
---|
| 612 | el.push_back(e);
|
---|
| 613 | e = 0;
|
---|
| 614 | } else {
|
---|
| 615 | e *= 10;
|
---|
| 616 | e += (*next - '0');
|
---|
| 617 | }
|
---|
| 618 | }
|
---|
| 619 |
|
---|
| 620 | // Get document list & the document frequency list
|
---|
| 621 | docnum.clear();
|
---|
| 622 | docfrq.clear();
|
---|
| 623 | bool readnum = false;
|
---|
| 624 | unsigned long d = 0;
|
---|
[9620] | 625 | for (++next; *next != ':'; ++next) {
|
---|
[2491] | 626 | if (*next == ',') {
|
---|
| 627 | docnum.push_back(d);
|
---|
| 628 | readnum = true;
|
---|
| 629 | d = 0;
|
---|
| 630 | } else if (*next == ';') {
|
---|
| 631 | if (readnum) {
|
---|
| 632 | docfrq.push_back(d);
|
---|
| 633 | } else {
|
---|
| 634 | docnum.push_back(d);
|
---|
| 635 | docfrq.push_back(1);
|
---|
| 636 | }
|
---|
| 637 | readnum = false;
|
---|
| 638 | d = 0;
|
---|
| 639 | } else {
|
---|
| 640 | d *= 10;
|
---|
| 641 | d += (*next - '0');
|
---|
| 642 | }
|
---|
| 643 | }
|
---|
| 644 |
|
---|
| 645 | // Get thesaurus link frequency & link list
|
---|
| 646 | text.push_back(':');
|
---|
| 647 | text.push_back(':');
|
---|
| 648 |
|
---|
| 649 | // link frequency
|
---|
| 650 | lf = 0;
|
---|
[9620] | 651 | for (++next; *next != ':'; ++next) {
|
---|
[2491] | 652 | lf *= 10;
|
---|
| 653 | lf += (*next - '0');
|
---|
| 654 | }
|
---|
| 655 |
|
---|
| 656 | // two lists of link data
|
---|
| 657 | linkdest.clear();
|
---|
| 658 | linktype.clear();
|
---|
| 659 |
|
---|
| 660 | UCArray thistype;
|
---|
| 661 | thistype.clear();
|
---|
| 662 | bool typedone = false;
|
---|
| 663 | unsigned long l = 0;
|
---|
[9620] | 664 | for (++next; *next != ':'; ++next) {
|
---|
[2491] | 665 |
|
---|
| 666 | if (!typedone) {
|
---|
| 667 | // first read the link type, a charactor string
|
---|
| 668 | if (*next == ',') {
|
---|
| 669 | typedone = true;
|
---|
| 670 | } else {
|
---|
| 671 | thistype.push_back(*next);
|
---|
| 672 | }
|
---|
| 673 | } else {
|
---|
| 674 | // having read the link type, read the list of link destinations
|
---|
| 675 | if (*next == ',') {
|
---|
| 676 | linkdest.push_back(l);
|
---|
| 677 | linktype.push_back(thistype);
|
---|
| 678 | l = 0;
|
---|
| 679 | } else if (*next == ';') {
|
---|
| 680 | linkdest.push_back(l);
|
---|
| 681 | linktype.push_back(thistype);
|
---|
| 682 | l = 0;
|
---|
| 683 | thistype.clear();
|
---|
| 684 | typedone = false;
|
---|
| 685 | } else {
|
---|
| 686 | l *= 10;
|
---|
| 687 | l += (*next - '0');
|
---|
| 688 | }
|
---|
| 689 | }
|
---|
| 690 | }
|
---|
[7734] | 691 |
|
---|
| 692 | return true; // Indicates that what happened is all good, baby.
|
---|
[2491] | 693 | }
|
---|
| 694 |
|
---|
[7734] | 695 | bool phindaction::print_thesaurus_links(const text_t &collection, bool XMLmode,
|
---|
[2491] | 696 | TextData &textdata, vector <unsigned long> &linkdest,
|
---|
| 697 | vector <UCArray> &linktype, unsigned long first,
|
---|
| 698 | unsigned long last, displayclass &disp,
|
---|
| 699 | outconvertclass &outconvert, ostream &textout) {
|
---|
| 700 |
|
---|
| 701 | // information describing each link in the list
|
---|
| 702 | unsigned long phrase, tf, ef, df;
|
---|
| 703 | UCArray type, text;
|
---|
| 704 |
|
---|
[9620] | 705 | for (unsigned long l = first; l < last; ++l) {
|
---|
[2491] | 706 |
|
---|
| 707 | // get the phrase data
|
---|
| 708 | phrase = linkdest[l];
|
---|
| 709 | type = linktype[l];
|
---|
[7734] | 710 |
|
---|
| 711 | /** DLConsulting 12-07-2004 */
|
---|
| 712 | if(!get_phrase_freq_data(textdata, phrase, text, tf, ef, df)) {
|
---|
| 713 | return false;
|
---|
| 714 | }
|
---|
[2491] | 715 |
|
---|
| 716 | if (XMLmode) {
|
---|
| 717 | textout << "<thesaurus num=\"" << l
|
---|
| 718 | << "\" id=\"" << phrase
|
---|
| 719 | << "\" tf=\"" << tf
|
---|
| 720 | << "\" df=\"" << df
|
---|
| 721 | << "\" type=\"" << type
|
---|
| 722 | << "\" text=\"" << text
|
---|
| 723 | << "\"/>\n";
|
---|
| 724 | } else {
|
---|
| 725 | textout << "<tr valign=top><td>" << type << "</td><td>";
|
---|
| 726 | textout << outconvert << disp
|
---|
| 727 | << "<a href=\"_gwcgi_?c=" << collection;
|
---|
| 728 | textout << "&ppnum=" << phrase << "\">" << text << "</a>"
|
---|
| 729 | << "</td><td>" << tf << "</td><td>" << df << "</td></tr>\n";
|
---|
| 730 | }
|
---|
| 731 | }
|
---|
[7734] | 732 |
|
---|
| 733 | /** DLConsulting 12-07-2004 */
|
---|
| 734 | return true;
|
---|
[2491] | 735 | }
|
---|
| 736 |
|
---|
| 737 | // Get the frequency data about a phrase
|
---|
| 738 | //
|
---|
| 739 | // The phrase is stored in textData as record phrase.
|
---|
| 740 | // We retrieve:
|
---|
| 741 | // word - the text of the phrase
|
---|
| 742 | // tf - the total frequency of the phrase
|
---|
| 743 | // ef - the expansion frequency of the phrase
|
---|
| 744 | // df - the document frequency of the phrase
|
---|
[7734] | 745 | /**
|
---|
| 746 | * Returns:
|
---|
| 747 | * false if the method suffered a fatal error, true otherwise
|
---|
| 748 | */
|
---|
| 749 | bool phindaction::get_phrase_freq_data(TextData &textdata, unsigned long phrase,
|
---|
[2491] | 750 | UCArray &word, unsigned long &tf,
|
---|
| 751 | unsigned long &ef, unsigned long &df) {
|
---|
| 752 |
|
---|
| 753 | UCArray text;
|
---|
| 754 | UCArray docLevel;
|
---|
[7415] | 755 | SetCStr(docLevel, "Document", 8);
|
---|
[2491] | 756 |
|
---|
| 757 | // Look the word up in the textData
|
---|
| 758 | if (!GetDocText (textdata, docLevel, phrase, text)) {
|
---|
| 759 | // FatalError (1, "Error while trying to get phrase %u", phrase);
|
---|
[7734] | 760 | //exit(0);
|
---|
| 761 | /** DLConsulting 12-07-2004 */
|
---|
| 762 | return false;
|
---|
[2491] | 763 | }
|
---|
| 764 |
|
---|
| 765 | // Ignore everything up to the first colon
|
---|
| 766 | UCArray::iterator next = text.begin();
|
---|
| 767 | while (*next++ != ':');
|
---|
| 768 |
|
---|
| 769 | // Get the word
|
---|
| 770 | word.clear();
|
---|
[9620] | 771 | for (; *next != ':'; ++next) {
|
---|
[2491] | 772 | word.push_back(*next);
|
---|
| 773 | }
|
---|
| 774 |
|
---|
| 775 | // Get total frequency
|
---|
| 776 | tf = 0;
|
---|
[9620] | 777 | for (++next; *next != ':'; ++next) {
|
---|
[2491] | 778 | tf *= 10;
|
---|
| 779 | tf += (*next - '0');
|
---|
| 780 | }
|
---|
| 781 |
|
---|
| 782 | // Get expansion frequency
|
---|
| 783 | ef = 0;
|
---|
[9620] | 784 | for (++next; *next != ':'; ++next) {
|
---|
[2491] | 785 | ef *= 10;
|
---|
| 786 | ef += (*next - '0');
|
---|
| 787 | }
|
---|
| 788 |
|
---|
| 789 | // Get document frequency
|
---|
| 790 | df = 0;
|
---|
[9620] | 791 | for (++next; *next != ':'; ++next) {
|
---|
[2491] | 792 | df *= 10;
|
---|
| 793 | df += (*next - '0');
|
---|
| 794 | }
|
---|
[7734] | 795 |
|
---|
| 796 | /** DLConsulting 12-07-2004 */
|
---|
| 797 | return true;
|
---|
[2491] | 798 | }
|
---|
| 799 |
|
---|
| 800 | // Print a list of expansions
|
---|
| 801 | //
|
---|
| 802 | // Given the textData and a list of phrase numbers, print out each of the
|
---|
| 803 | // expansions.
|
---|
| 804 | void phindaction::print_expansions(const text_t &collection, bool XMLmode,
|
---|
| 805 | const text_t &body, TextData &textdata,
|
---|
| 806 | const vector <unsigned long> &elist,
|
---|
| 807 | unsigned long first, unsigned long last,
|
---|
| 808 | displayclass &disp, outconvertclass &outconvert,
|
---|
| 809 | ostream &textout) {
|
---|
| 810 |
|
---|
| 811 | UCArray word;
|
---|
| 812 | unsigned long phrase, tf, df, ef;
|
---|
| 813 |
|
---|
| 814 | UCArray suffix, prefix, ucbody;
|
---|
| 815 |
|
---|
| 816 | toUCArray(body, ucbody);
|
---|
| 817 |
|
---|
[9620] | 818 | for (unsigned long e = first; e < last; ++e) {
|
---|
[2491] | 819 |
|
---|
| 820 | phrase = elist[e];
|
---|
| 821 | get_phrase_freq_data(textdata, phrase, word, tf, ef, df);
|
---|
| 822 |
|
---|
| 823 | split_phrase(word, ucbody, prefix, suffix);
|
---|
| 824 |
|
---|
| 825 | if (XMLmode) {
|
---|
| 826 | // body is always the same as the text of the phrase, so no need to send it
|
---|
| 827 | textout << "<expansion num=\"" << e
|
---|
| 828 | << "\" id=\"" << phrase
|
---|
| 829 | << "\" tf=\"" << tf
|
---|
| 830 | << "\" df=\"" << df;
|
---|
| 831 | if (!prefix.empty()) {
|
---|
| 832 | textout << "\" prefix=\"" << prefix;
|
---|
| 833 | }
|
---|
| 834 | if (!suffix.empty()) {
|
---|
| 835 | textout << "\" suffix=\"" << suffix;
|
---|
| 836 | }
|
---|
| 837 | textout << "\"/>\n";
|
---|
| 838 | } else {
|
---|
| 839 | textout << outconvert << disp
|
---|
| 840 | << "<tr valign=top><td align=right><a href=\"_gwcgi_?"
|
---|
| 841 | << "c=" << collection << "&ppnum=" << phrase << "\">";
|
---|
| 842 | textout << prefix << "</a></td>";
|
---|
| 843 | textout <<outconvert << disp
|
---|
| 844 | << "<td align=center><a href=\"_gwcgi_?"
|
---|
| 845 | << "c=" << collection << "&ppnum=" << phrase << "\">"
|
---|
| 846 | << body << "</a></td>"
|
---|
| 847 | << "<td align=left><a href=\"_gwcgi_?"
|
---|
| 848 | << "c=" << collection << "&ppnum=" << phrase << "\">";
|
---|
| 849 | textout << suffix << "</a></td>"
|
---|
| 850 | << "<td>" << tf << "</td><td>" << df << "</td></tr>\n";
|
---|
| 851 | }
|
---|
| 852 | }
|
---|
| 853 | }
|
---|
| 854 |
|
---|
| 855 | // split an expansion into prefix and suffix
|
---|
| 856 | void phindaction::split_phrase(const UCArray &word, const UCArray &body,
|
---|
| 857 | UCArray &prefix, UCArray &suffix) {
|
---|
| 858 |
|
---|
| 859 | prefix.clear();
|
---|
| 860 | suffix.clear();
|
---|
| 861 |
|
---|
| 862 | bool readingPrefix = true;
|
---|
| 863 | UCArray::const_iterator here = word.begin();
|
---|
| 864 | UCArray::const_iterator end = word.end();
|
---|
| 865 |
|
---|
| 866 | while (here != end) {
|
---|
| 867 |
|
---|
| 868 | // if we've not read all the prefix, add the next char to the prefix
|
---|
| 869 | if (readingPrefix) {
|
---|
| 870 | if (phrase_match(body, here, end)) {
|
---|
| 871 | readingPrefix = false;
|
---|
| 872 | // trim whitespace from end of prefix & start of suffix
|
---|
| 873 | if (!prefix.empty()) {
|
---|
| 874 | prefix.pop_back();
|
---|
| 875 | }
|
---|
| 876 | if ((here != end) && (*here == ' ')) {
|
---|
[7417] | 877 | ++here;
|
---|
[2491] | 878 | }
|
---|
| 879 | } else {
|
---|
| 880 | prefix.push_back(*here);
|
---|
[7417] | 881 | ++here;
|
---|
[2491] | 882 | }
|
---|
| 883 | }
|
---|
| 884 | // if we've finished with the prefix, update the suffix
|
---|
| 885 | else {
|
---|
| 886 | suffix.push_back(*here);
|
---|
[7417] | 887 | ++here;
|
---|
[2491] | 888 | }
|
---|
| 889 | }
|
---|
| 890 | }
|
---|
| 891 |
|
---|
| 892 | // phrase_match
|
---|
| 893 | //
|
---|
| 894 | // compare two strings, one represented as an UCArray, the other as two
|
---|
| 895 | // UCArray iterators.
|
---|
| 896 | //
|
---|
| 897 | // Return true if the UCArray is the same as the phrase the iterators point
|
---|
| 898 | // to for the length of the UCArray.
|
---|
| 899 | bool phindaction::phrase_match(const UCArray &text, UCArray::const_iterator &here,
|
---|
| 900 | UCArray::const_iterator end) {
|
---|
| 901 |
|
---|
| 902 | UCArray::const_iterator one_here = text.begin();
|
---|
| 903 | UCArray::const_iterator one_end = text.end();
|
---|
| 904 | UCArray::const_iterator two_here = here;
|
---|
| 905 |
|
---|
| 906 | // iterate over the length of the first string, comparing each element to
|
---|
| 907 | // the corresponding element in the second string.
|
---|
| 908 | while (one_here != one_end) {
|
---|
| 909 |
|
---|
| 910 | if (two_here == end) {
|
---|
| 911 | return false;
|
---|
| 912 | } else if (*one_here != *two_here) {
|
---|
| 913 | return false;
|
---|
| 914 | }
|
---|
[7417] | 915 | ++one_here;
|
---|
| 916 | ++two_here;
|
---|
[2491] | 917 | }
|
---|
| 918 |
|
---|
| 919 | here = two_here;
|
---|
| 920 | return true;
|
---|
| 921 | }
|
---|
| 922 |
|
---|
[7734] | 923 | bool phindaction::print_documents(bool XMLmode, const text_t &basepath,
|
---|
[2491] | 924 | const text_t &collection,
|
---|
| 925 | const vector <unsigned long> &docNums,
|
---|
| 926 | const vector <unsigned long> &docFreq,
|
---|
| 927 | unsigned long first, unsigned long last,
|
---|
| 928 | displayclass &disp, outconvertclass &outconvert,
|
---|
| 929 | ostream &textout) {
|
---|
| 930 |
|
---|
| 931 | // Create a TextData object to read the document data
|
---|
| 932 | TextData docdata;
|
---|
| 933 |
|
---|
| 934 | text_t fullpath = filename_cat(basepath, "docs");
|
---|
| 935 | char *fullpathc = fullpath.getcstr();
|
---|
| 936 | #if defined __WIN32__
|
---|
| 937 | char *base = "";
|
---|
| 938 | #else
|
---|
| 939 | char *base = "/";
|
---|
| 940 | #endif
|
---|
| 941 |
|
---|
| 942 | if (!docdata.LoadData (base, fullpathc)) {
|
---|
| 943 | // FatalError (1, "Couldn't load text information for \"%s\"", fullpathc);
|
---|
[7734] | 944 | //exit(0);
|
---|
| 945 | /** DLConsulting 12-07-2004 */
|
---|
| 946 | return false;
|
---|
[2491] | 947 | }
|
---|
| 948 |
|
---|
[7384] | 949 | delete []fullpathc;
|
---|
[2491] | 950 |
|
---|
| 951 | UCArray title, hash;
|
---|
| 952 | unsigned long freq, doc;
|
---|
| 953 |
|
---|
[9620] | 954 | for (unsigned long d = first; d < last; ++d) {
|
---|
[2491] | 955 | doc = docNums[d];
|
---|
| 956 | freq = docFreq[d];
|
---|
| 957 |
|
---|
[7734] | 958 | /** DLConsulting 13-07-2004 */
|
---|
| 959 | if(!get_document_all_data(docdata, doc, title, hash)) {
|
---|
| 960 | return false;
|
---|
| 961 | }
|
---|
[2491] | 962 |
|
---|
| 963 | if (XMLmode) {
|
---|
| 964 | textout << "<document num=\"" << d
|
---|
| 965 | << "\" hash=\"" << hash
|
---|
| 966 | << "\" freq=\"" << freq
|
---|
| 967 | << "\" title=\"" << title << "\"/>\n";
|
---|
| 968 | } else {
|
---|
| 969 | textout << outconvert << disp
|
---|
| 970 | << "<tr valign=top><td><a href=\"_gwcgi_?"
|
---|
| 971 | << "c=" << collection;
|
---|
| 972 | textout << "&a=d&d=" << hash << "\">" << title << "</a>"
|
---|
| 973 | << "</td><td>" << freq << "</td></tr>\n";
|
---|
| 974 | }
|
---|
| 975 | }
|
---|
[2542] | 976 |
|
---|
| 977 | docdata.UnloadData();
|
---|
[7734] | 978 |
|
---|
| 979 | /** DLConsulting 12-07-2004 */
|
---|
| 980 | return true;
|
---|
[2491] | 981 | }
|
---|
| 982 |
|
---|
| 983 | // Get all the data about a docment
|
---|
| 984 | //
|
---|
| 985 | // The document's details are stored in docData as record docNum.
|
---|
| 986 | // We retrieve:
|
---|
| 987 | // title - the document's title
|
---|
| 988 | // hash - the document's unique OID
|
---|
[7734] | 989 | /** Returns:
|
---|
| 990 | * false if a fatal error occured, true otherwise
|
---|
| 991 | * DLConsulting 12-07-2004
|
---|
| 992 | */
|
---|
| 993 | bool phindaction::get_document_all_data(TextData &docdata, unsigned long docNum,
|
---|
[2491] | 994 | UCArray &title, UCArray &hash) {
|
---|
| 995 |
|
---|
| 996 | UCArray text;
|
---|
| 997 | UCArray docLevel;
|
---|
[7415] | 998 | SetCStr(docLevel, "Document", 8);
|
---|
[2491] | 999 |
|
---|
| 1000 | // Look the word up in the textData
|
---|
| 1001 | if (!GetDocText (docdata, docLevel, docNum, text)) {
|
---|
| 1002 | // FatalError (1, "Error while trying to get document %u", docNum);
|
---|
[7734] | 1003 | //exit(0);
|
---|
| 1004 | /** DLConsulting 13-07-2004 */
|
---|
| 1005 | return false;
|
---|
[2491] | 1006 | }
|
---|
| 1007 |
|
---|
| 1008 | // Ignore everything up to the first colon
|
---|
| 1009 | UCArray::iterator next = text.begin();
|
---|
| 1010 | while (*next++ != '\t');
|
---|
| 1011 |
|
---|
| 1012 | // Get the document OID (hash)
|
---|
| 1013 | hash.clear();
|
---|
[9620] | 1014 | for (; *next != '\t'; ++next) {
|
---|
[2491] | 1015 | hash.push_back(*next);
|
---|
| 1016 | }
|
---|
| 1017 |
|
---|
| 1018 | // Get the title
|
---|
| 1019 | text.push_back('\n');
|
---|
| 1020 | title.clear();
|
---|
[9620] | 1021 | for (++next; *next != '\n'; ++next) {
|
---|
[2491] | 1022 | title.push_back(*next);
|
---|
| 1023 | }
|
---|
[7734] | 1024 |
|
---|
| 1025 | /** DLConsulting 13-07-2004 */
|
---|
| 1026 | return true;
|
---|
[2491] | 1027 | }
|
---|
| 1028 |
|
---|
| 1029 | void phindaction::toUCArray(const text_t &in, UCArray &out) {
|
---|
| 1030 | out.clear();
|
---|
[7417] | 1031 | if (out.capacity() < in.size() + 1) {
|
---|
| 1032 | out.reserve(in.size() + 1);
|
---|
| 1033 | }
|
---|
[2491] | 1034 | text_t::const_iterator here = in.begin();
|
---|
| 1035 | text_t::const_iterator end = in.end();
|
---|
| 1036 | while (here != end) {
|
---|
| 1037 | out.push_back((unsigned char) *here);
|
---|
[7417] | 1038 | ++here;
|
---|
[2491] | 1039 | }
|
---|
| 1040 | }
|
---|
| 1041 |
|
---|
[2502] | 1042 | void phindaction::output_error (const text_t &message, ostream &textout,
|
---|
| 1043 | outconvertclass &outconvert,
|
---|
| 1044 | displayclass & disp, ostream &logout,
|
---|
| 1045 | bool XMLmode) {
|
---|
| 1046 |
|
---|
| 1047 | logout << outconvert << message << "\n";
|
---|
| 1048 | if (XMLmode) {
|
---|
| 1049 | textout << outconvert
|
---|
| 1050 | << "<phinddata>\n"
|
---|
| 1051 | << "<phinderror>" << message << "</phinderror>\n"
|
---|
| 1052 | << "</phinddata>\n";
|
---|
| 1053 | } else {
|
---|
| 1054 | textout << outconvert << disp
|
---|
| 1055 | << "_header_\n"
|
---|
| 1056 | << message
|
---|
| 1057 | << "_footer_\n";
|
---|
| 1058 | }
|
---|
| 1059 | }
|
---|
[7371] | 1060 |
|
---|
| 1061 | #endif //GSDL_USE_PHIND_ACTION
|
---|
[7734] | 1062 |
|
---|