1 | /**********************************************************************
|
---|
2 | *
|
---|
3 | * filter.cpp --
|
---|
4 | * Copyright (C) 1999 The New Zealand Digital Library Project
|
---|
5 | *
|
---|
6 | * A component of the Greenstone digital library software
|
---|
7 | * from the New Zealand Digital Library Project at the
|
---|
8 | * University of Waikato, New Zealand.
|
---|
9 | *
|
---|
10 | * This program is free software; you can redistribute it and/or modify
|
---|
11 | * it under the terms of the GNU General Public License as published by
|
---|
12 | * the Free Software Foundation; either version 2 of the License, or
|
---|
13 | * (at your option) any later version.
|
---|
14 | *
|
---|
15 | * This program is distributed in the hope that it will be useful,
|
---|
16 | * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
17 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
18 | * GNU General Public License for more details.
|
---|
19 | *
|
---|
20 | * You should have received a copy of the GNU General Public License
|
---|
21 | * along with this program; if not, write to the Free Software
|
---|
22 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
23 | *
|
---|
24 | *********************************************************************/
|
---|
25 |
|
---|
26 | #include "filter.h"
|
---|
27 | #include "fileutil.h"
|
---|
28 | #include <assert.h>
|
---|
29 |
|
---|
30 | #include <iostream>
|
---|
31 | using namespace std;
|
---|
32 |
|
---|
33 |
|
---|
34 | // default constructor does nothing
|
---|
35 | filterclass::filterclass () {
|
---|
36 | }
|
---|
37 |
|
---|
38 | // default destructor does nothing
|
---|
39 | filterclass::~filterclass () {
|
---|
40 | }
|
---|
41 |
|
---|
42 | // configure should be called once for each configuration line
|
---|
43 | // default configures the default filter options
|
---|
44 | void filterclass::configure (const text_t &key, const text_tarray &cfgline) {
|
---|
45 | if (cfgline.size() >= 1) {
|
---|
46 | const text_t &value = cfgline[0];
|
---|
47 |
|
---|
48 | if (key == "collection") collection = value;
|
---|
49 | else if (key == "collectdir") collectdir = value;
|
---|
50 | else if (key == "gsdlhome") gsdlhome = value;
|
---|
51 | else if (key == "collecthome") collecthome = value;
|
---|
52 | else if (key == "gdbmhome") dbhome = value;
|
---|
53 | else if ((key == "filteroptdefault") && (cfgline.size() == 2)) {
|
---|
54 | // see if this filter has an option with this name
|
---|
55 | FilterOption_tmap::iterator thisfilteroption =
|
---|
56 | filterOptions.find(cfgline[0]);
|
---|
57 | if (thisfilteroption != filterOptions.end())
|
---|
58 | (*thisfilteroption).second.defaultValue = cfgline[1];
|
---|
59 | }
|
---|
60 | }
|
---|
61 | }
|
---|
62 |
|
---|
63 | text_t filterclass::resolve_db_filename(const text_t& idx,
|
---|
64 | const text_t& file_ext)
|
---|
65 | {
|
---|
66 | text_t resolved_filename;
|
---|
67 |
|
---|
68 | if (gsdlhome==dbhome) {
|
---|
69 | // dbhome has defaulted to gsdlhome which we take to means the
|
---|
70 | // database has been specifically moved out of gsdlhome area.
|
---|
71 | // => it should be whereever collecthome is set to
|
---|
72 | resolved_filename = filename_cat(collecthome, collection, "index", "text", idx);
|
---|
73 | }
|
---|
74 | else {
|
---|
75 | // dbhome is explicitly set to something other than gsdlhome
|
---|
76 | // => use dbhome
|
---|
77 | resolved_filename = filename_cat(dbhome, "collect", collection, "index", "text", idx);
|
---|
78 | }
|
---|
79 |
|
---|
80 | resolved_filename += file_ext;
|
---|
81 |
|
---|
82 | return resolved_filename;
|
---|
83 | }
|
---|
84 |
|
---|
85 |
|
---|
86 | // init should be called after all the configuration is done but
|
---|
87 | // before any other methods are called
|
---|
88 | // default checks all the filter option defaults
|
---|
89 | bool filterclass::init (ostream &/*logout*/) {
|
---|
90 | // check all the filter defaults
|
---|
91 | FilterOption_tmap::iterator filteroption_here = filterOptions.begin();
|
---|
92 | FilterOption_tmap::iterator filteroption_end = filterOptions.end();
|
---|
93 | while (filteroption_here != filteroption_end) {
|
---|
94 | (*filteroption_here).second.check_defaultValue ();
|
---|
95 |
|
---|
96 | ++filteroption_here;
|
---|
97 | }
|
---|
98 |
|
---|
99 | if (collecthome.empty()) collecthome = filename_cat(gsdlhome,"collect");
|
---|
100 | if (dbhome.empty()) dbhome = gsdlhome;
|
---|
101 |
|
---|
102 | // get the collection directory name
|
---|
103 | if (collectdir.empty()) {
|
---|
104 | collectdir = filename_cat (collecthome, collection);
|
---|
105 | }
|
---|
106 |
|
---|
107 | return true;
|
---|
108 | }
|
---|
109 |
|
---|
110 | // returns the name of this filter
|
---|
111 | // default returns "NullFilter"
|
---|
112 | text_t filterclass::get_filter_name () {
|
---|
113 | return "NullFilter";
|
---|
114 | }
|
---|
115 |
|
---|
116 | // returns the current filter options
|
---|
117 | void filterclass::get_filteroptions (InfoFilterOptionsResponse_t &response,
|
---|
118 | comerror_t &err, ostream &/*logout*/) {
|
---|
119 | response.clear();
|
---|
120 | response.filterOptions = filterOptions;
|
---|
121 | err = noError;
|
---|
122 | }
|
---|
123 |
|
---|
124 | // default returns nothing
|
---|
125 | void filterclass::filter (const FilterRequest_t &request,
|
---|
126 | FilterResponse_t &response,
|
---|
127 | comerror_t &err, ostream &/*logout*/) {
|
---|
128 | ResultDocInfo_t resultdoc;
|
---|
129 |
|
---|
130 | response.clear();
|
---|
131 |
|
---|
132 | if ((request.filterResultOptions & FROID) ||
|
---|
133 | (request.filterResultOptions & FRmetadata)) {
|
---|
134 | // copy the OIDs from the request to the response
|
---|
135 | text_tarray::const_iterator here = request.docSet.begin();
|
---|
136 | text_tarray::const_iterator end = request.docSet.end();
|
---|
137 | while (here != end) {
|
---|
138 | resultdoc.OID = (*here);
|
---|
139 | response.docInfo.push_back(resultdoc);
|
---|
140 |
|
---|
141 | ++here;
|
---|
142 | }
|
---|
143 | }
|
---|
144 |
|
---|
145 | response.numDocs = response.docInfo.size();
|
---|
146 | response.isApprox = Exact;
|
---|
147 | err = noError;
|
---|
148 | }
|
---|
149 |
|
---|
150 |
|
---|
151 | bool operator==(const filterptr &x, const filterptr &y) {
|
---|
152 | return (x.f == y.f);
|
---|
153 | }
|
---|
154 |
|
---|
155 | bool operator<(const filterptr &x, const filterptr &y) {
|
---|
156 | return (x.f < y.f);
|
---|
157 | }
|
---|
158 |
|
---|
159 |
|
---|
160 | // thefilter remains the property of the calling code but
|
---|
161 | // should not be deleted until it is removed from this list.
|
---|
162 | void filtermapclass::addfilter (filterclass *thefilter) {
|
---|
163 | // can't add a null filter
|
---|
164 | assert (thefilter != NULL);
|
---|
165 | if (thefilter == NULL) return;
|
---|
166 |
|
---|
167 | // can't add an filter with no name
|
---|
168 | assert (!(thefilter->get_filter_name()).empty());
|
---|
169 | if ((thefilter->get_filter_name()).empty()) return;
|
---|
170 |
|
---|
171 | filterptr fptr;
|
---|
172 | fptr.f = thefilter;
|
---|
173 | filterptrs[thefilter->get_filter_name()] = fptr;
|
---|
174 | }
|
---|
175 |
|
---|
176 | // getfilter will return NULL if the filter could not be found
|
---|
177 | filterclass *filtermapclass::getfilter (const text_t &key) {
|
---|
178 | // can't find an filter with no name
|
---|
179 | assert (!key.empty());
|
---|
180 | if (key.empty()) return NULL;
|
---|
181 |
|
---|
182 | iterator here = filterptrs.find (key);
|
---|
183 | if (here == filterptrs.end()) return NULL;
|
---|
184 |
|
---|
185 | return (*here).second.f;
|
---|
186 | }
|
---|
187 |
|
---|
188 |
|
---|
189 |
|
---|
190 |
|
---|
191 | // some useful functions for dealing with document sets
|
---|
192 |
|
---|
193 | // returns -1 if t1 is a child of t2
|
---|
194 | // returns 0 if t1 and t2 are not parent-child related
|
---|
195 | // returns 1 if t1 is a parent of t2
|
---|
196 | int child_compare (const text_t &t1, const text_t &t2) {
|
---|
197 | text_t::const_iterator t1_here = t1.begin();
|
---|
198 | text_t::const_iterator t1_end = t1.end();
|
---|
199 | text_t::const_iterator t2_here = t2.begin();
|
---|
200 | text_t::const_iterator t2_end = t2.end();
|
---|
201 |
|
---|
202 | while ((t1_here != t1_end) && (t2_here != t2_end)) {
|
---|
203 | if (*t1_here != *t2_here) return 0; // unrelated
|
---|
204 | ++t1_here;
|
---|
205 | ++t2_here;
|
---|
206 | }
|
---|
207 |
|
---|
208 | if ((t1_here == t1_end) && (t2_here == t2_end)) return 0; // equal
|
---|
209 | if (t1_here != t1_end) {
|
---|
210 | if (*t1_here == '.') return -1; // t1 is child
|
---|
211 | else return 0; // unrelated
|
---|
212 | }
|
---|
213 |
|
---|
214 | if (t2_here != t2_end) {
|
---|
215 | if (*t2_here == '.') return 1; // t2 is child
|
---|
216 | else return 0; // unrelated
|
---|
217 | }
|
---|
218 |
|
---|
219 | return 0; // shouldn't get here...
|
---|
220 | }
|
---|
221 |
|
---|
222 | // intersect places the result in set1
|
---|
223 | void intersect (text_tset &set1, const text_tset &set2) {
|
---|
224 | text_tset resultset;
|
---|
225 | int childcomp = 0;
|
---|
226 |
|
---|
227 | text_tset::const_iterator set1_here = set1.begin();
|
---|
228 | text_tset::const_iterator set1_end = set1.end();
|
---|
229 | text_tset::const_iterator set2_here = set2.begin();
|
---|
230 | text_tset::const_iterator set2_end = set2.end();
|
---|
231 | while ((set1_here != set1_end) && (set2_here != set2_end)) {
|
---|
232 | if (*set1_here == *set2_here) {
|
---|
233 | // equal
|
---|
234 | resultset.insert (*set1_here);
|
---|
235 | ++set1_here;
|
---|
236 | ++set2_here;
|
---|
237 |
|
---|
238 | } else if ((childcomp=child_compare(*set1_here, *set2_here)) != 0) {
|
---|
239 | if (childcomp < 0) {
|
---|
240 | // set1_here is child
|
---|
241 | resultset.insert (*set1_here);
|
---|
242 | ++set1_here;
|
---|
243 | } else {
|
---|
244 | // set2_here is child
|
---|
245 | resultset.insert (*set2_here);
|
---|
246 | ++set2_here;
|
---|
247 | }
|
---|
248 |
|
---|
249 | } else if (*set1_here < *set2_here) {
|
---|
250 | // set1 is less
|
---|
251 | ++set1_here;
|
---|
252 |
|
---|
253 | } else {
|
---|
254 | // set2 is less
|
---|
255 | ++set2_here;
|
---|
256 | }
|
---|
257 | }
|
---|
258 |
|
---|
259 | set1 = resultset;
|
---|
260 | }
|
---|
261 |
|
---|
262 | void intersect (text_tarray &set1, const text_tset &set2) {
|
---|
263 | text_tarray resultset;
|
---|
264 |
|
---|
265 | text_tarray::const_iterator set1_here = set1.begin();
|
---|
266 | text_tarray::const_iterator set1_end = set1.end();
|
---|
267 |
|
---|
268 | while (set1_here != set1_end) {
|
---|
269 | if (in_set (set2, *set1_here))
|
---|
270 | resultset.push_back (*set1_here);
|
---|
271 | ++set1_here;
|
---|
272 | }
|
---|
273 | set1 = resultset;
|
---|
274 | }
|
---|
275 |
|
---|
276 | void intersect (text_tarray &set1, const text_tarray &set2) {
|
---|
277 | text_tarray resultset;
|
---|
278 |
|
---|
279 | text_tarray::const_iterator set1_here = set1.begin();
|
---|
280 | text_tarray::const_iterator set1_end = set1.end();
|
---|
281 |
|
---|
282 | while (set1_here != set1_end) {
|
---|
283 | if (in_set (set2, *set1_here))
|
---|
284 | resultset.push_back (*set1_here);
|
---|
285 | ++set1_here;
|
---|
286 | }
|
---|
287 | set1 = resultset;
|
---|
288 | }
|
---|
289 |
|
---|
290 | // tests to see if el is in set
|
---|
291 | bool in_set (const text_tset &set1, const text_t &el) {
|
---|
292 | text_t::const_iterator here = el.begin();
|
---|
293 | text_t::const_iterator end = el.end();
|
---|
294 | text_t tryel, tryel_add;
|
---|
295 | bool first = true;
|
---|
296 |
|
---|
297 | // the element is in the set if any of its parents are
|
---|
298 | // in the set
|
---|
299 | do {
|
---|
300 | // get next possible element to try
|
---|
301 | here = getdelimitstr (here, end, '.', tryel_add);
|
---|
302 | if (!first) tryel += ".";
|
---|
303 | first = false;
|
---|
304 | tryel += tryel_add;
|
---|
305 |
|
---|
306 | // see if this element is in the set
|
---|
307 | if (set1.find(tryel) != set1.end()) return true;
|
---|
308 | } while (here != end);
|
---|
309 |
|
---|
310 | return false;
|
---|
311 | }
|
---|
312 |
|
---|
313 | bool in_set (const text_tarray &set1, const text_t &el) {
|
---|
314 | text_t::const_iterator here = el.begin();
|
---|
315 | text_t::const_iterator end = el.end();
|
---|
316 | text_t tryel, tryel_add;
|
---|
317 | bool first = true;
|
---|
318 |
|
---|
319 | // the element is in the set if any of its parents are
|
---|
320 | // in the set
|
---|
321 | do {
|
---|
322 | // get next possible element to try
|
---|
323 | here = getdelimitstr (here, end, '.', tryel_add);
|
---|
324 | if (!first) tryel += ".";
|
---|
325 | first = false;
|
---|
326 | tryel += tryel_add;
|
---|
327 |
|
---|
328 | // see if this element is in the set
|
---|
329 | text_tarray::const_iterator h = set1.begin();
|
---|
330 | text_tarray::const_iterator e = set1.end();
|
---|
331 | while (h != e) {
|
---|
332 | if (*h == tryel) return true;
|
---|
333 | ++h;
|
---|
334 | }
|
---|
335 | } while (here != end);
|
---|
336 |
|
---|
337 | return false;
|
---|
338 | }
|
---|