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