1 | /**********************************************************************
|
---|
2 | *
|
---|
3 | * OIDtools.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 | * $Id: OIDtools.cpp 533 1999-09-07 04:57:01Z sjboddie $
|
---|
25 | *
|
---|
26 | *********************************************************************/
|
---|
27 |
|
---|
28 | /*
|
---|
29 | $Log$
|
---|
30 | Revision 1.16 1999/09/07 04:56:51 sjboddie
|
---|
31 | added GPL notice
|
---|
32 |
|
---|
33 | Revision 1.15 1999/09/02 00:22:04 rjmcnab
|
---|
34 | added extra check in get_info
|
---|
35 |
|
---|
36 | Revision 1.14 1999/08/25 04:45:25 sjboddie
|
---|
37 | changed FilterRequest_t::docSet into an array
|
---|
38 |
|
---|
39 | Revision 1.13 1999/08/10 22:44:06 sjboddie
|
---|
40 | altered the get_contents function so it's now passed metadata
|
---|
41 | rather than having it hard-coded
|
---|
42 |
|
---|
43 | Revision 1.12 1999/07/30 02:26:44 sjboddie
|
---|
44 | added overloaded get_info function for getting info on multiple OIDs
|
---|
45 |
|
---|
46 | Revision 1.11 1999/07/20 02:59:03 sjboddie
|
---|
47 | get_children now takes a getParents argument
|
---|
48 |
|
---|
49 | Revision 1.10 1999/07/07 05:47:41 sjboddie
|
---|
50 | changed around the way browsetools works
|
---|
51 |
|
---|
52 | Revision 1.9 1999/06/16 23:51:53 sjboddie
|
---|
53 | added a strip_suffix function
|
---|
54 |
|
---|
55 | Revision 1.8 1999/06/16 03:11:25 sjboddie
|
---|
56 | get_info() now takes a getParents argument
|
---|
57 |
|
---|
58 | Revision 1.7 1999/06/16 02:05:23 sjboddie
|
---|
59 | just changed a comment that was confusing me
|
---|
60 |
|
---|
61 | Revision 1.6 1999/05/10 03:40:25 sjboddie
|
---|
62 | lots of changes - slowly getting document action sorted out
|
---|
63 |
|
---|
64 | Revision 1.5 1999/04/30 01:59:37 sjboddie
|
---|
65 | lots of stuff - getting documentaction working (documentaction replaces
|
---|
66 | old browseaction)
|
---|
67 |
|
---|
68 | Revision 1.4 1999/03/29 02:14:25 sjboddie
|
---|
69 |
|
---|
70 | More changes to browseaction
|
---|
71 |
|
---|
72 | Revision 1.3 1999/03/25 03:13:42 sjboddie
|
---|
73 |
|
---|
74 | More library functions for dealing with OIDs. Many of them just
|
---|
75 | return dummy data at present
|
---|
76 |
|
---|
77 | Revision 1.2 1999/03/05 03:53:53 sjboddie
|
---|
78 |
|
---|
79 | fixed some bugs
|
---|
80 |
|
---|
81 | Revision 1.1 1999/03/04 22:38:20 sjboddie
|
---|
82 |
|
---|
83 | Added subjectbrowseaction. - Doesn't do anything yet.
|
---|
84 |
|
---|
85 | */
|
---|
86 |
|
---|
87 | #include "OIDtools.h"
|
---|
88 |
|
---|
89 |
|
---|
90 | // returns (in top) the top level of OID (i.e. everything
|
---|
91 | // up until the first dot)
|
---|
92 | void get_top (const text_t &OID, text_t &top) {
|
---|
93 |
|
---|
94 | top.clear();
|
---|
95 | if (OID.empty()) return;
|
---|
96 |
|
---|
97 | text_t::const_iterator begin = OID.begin();
|
---|
98 | text_t::const_iterator end = OID.end();
|
---|
99 |
|
---|
100 | top.appendrange (begin, findchar(begin, end, '.'));
|
---|
101 | }
|
---|
102 |
|
---|
103 |
|
---|
104 | // checks if OID is top level (i.e. contains no dots)
|
---|
105 | bool is_top (const text_t &OID) {
|
---|
106 |
|
---|
107 | if (OID.empty()) return true;
|
---|
108 |
|
---|
109 | text_t::const_iterator here = OID.begin();
|
---|
110 | text_t::const_iterator end = OID.end();
|
---|
111 | here = findchar (here, end, '.');
|
---|
112 |
|
---|
113 | if (here == end) return true;
|
---|
114 | return false;
|
---|
115 | }
|
---|
116 |
|
---|
117 |
|
---|
118 | // get_parents_array loads the parents array with all the parents of the
|
---|
119 | // document or classification specified by OID (not including OID itself)
|
---|
120 | void get_parents_array (const text_t &OID, text_tarray &parents) {
|
---|
121 |
|
---|
122 | parents.erase (parents.begin(), parents.end());
|
---|
123 |
|
---|
124 | text_t::const_iterator here = OID.begin ();
|
---|
125 | text_t::const_iterator end = OID.end ();
|
---|
126 | text_t thisparent;
|
---|
127 |
|
---|
128 | while (here != end) {
|
---|
129 | if (*here == '.') parents.push_back(thisparent);
|
---|
130 | thisparent.push_back(*here);
|
---|
131 | here ++;
|
---|
132 | }
|
---|
133 | }
|
---|
134 |
|
---|
135 |
|
---|
136 | // get_info does a protocol call and returns (in response) the metadata
|
---|
137 | // associated with OID. The metadata array should be loaded with whatever
|
---|
138 | // metadata fields are to be requested.
|
---|
139 |
|
---|
140 | bool get_info (const text_t &OID, const text_t &collection,
|
---|
141 | const text_tarray &metadata, bool getParents,
|
---|
142 | recptproto *collectproto, FilterResponse_t &response,
|
---|
143 | ostream &logout) {
|
---|
144 |
|
---|
145 | response.clear();
|
---|
146 |
|
---|
147 | comerror_t err;
|
---|
148 | FilterRequest_t request;
|
---|
149 |
|
---|
150 | request.filterName = "NullFilter";
|
---|
151 | request.filterResultOptions = FRmetadata;
|
---|
152 | request.getParents = getParents;
|
---|
153 | request.fields = metadata;
|
---|
154 | request.docSet.push_back (OID);
|
---|
155 |
|
---|
156 | collectproto->filter (collection, request, response, err, logout);
|
---|
157 | if (err != noError) {
|
---|
158 | outconvertclass text_t2ascii;
|
---|
159 | logout << text_t2ascii
|
---|
160 | << "Error: call to filter failed for " << OID
|
---|
161 | << " in OIDtools::get_info ("
|
---|
162 | << get_comerror_string (err) << ")\n";
|
---|
163 | return false;
|
---|
164 | }
|
---|
165 |
|
---|
166 | // check the result to make sure we have the right amount of metadata
|
---|
167 | if (!response.docInfo.empty() &&
|
---|
168 | response.docInfo[0].metadata.size() != metadata.size()) return false;
|
---|
169 |
|
---|
170 | return true;
|
---|
171 | }
|
---|
172 |
|
---|
173 | bool get_info (const text_tarray &OIDs, const text_t &collection,
|
---|
174 | const text_tarray &metadata, bool getParents,
|
---|
175 | recptproto *collectproto, FilterResponse_t &response,
|
---|
176 | ostream &logout) {
|
---|
177 |
|
---|
178 | response.clear();
|
---|
179 | if (OIDs.empty()) return true;
|
---|
180 |
|
---|
181 | comerror_t err;
|
---|
182 | FilterRequest_t request;
|
---|
183 |
|
---|
184 | request.filterName = "NullFilter";
|
---|
185 | request.filterResultOptions = FRmetadata;
|
---|
186 | request.getParents = getParents;
|
---|
187 | request.fields = metadata;
|
---|
188 |
|
---|
189 | request.docSet = OIDs;
|
---|
190 |
|
---|
191 | collectproto->filter (collection, request, response, err, logout);
|
---|
192 | if (err != noError) {
|
---|
193 | outconvertclass text_t2ascii;
|
---|
194 | logout << text_t2ascii
|
---|
195 | << "Error: call to filter failed in OIDtools::get_info ("
|
---|
196 | << get_comerror_string (err) << ")\n";
|
---|
197 | return false;
|
---|
198 | }
|
---|
199 |
|
---|
200 | // check the result to make sure we have the right amount of metadata
|
---|
201 | if (!response.docInfo.empty() &&
|
---|
202 | response.docInfo[0].metadata.size() != metadata.size()) return false;
|
---|
203 |
|
---|
204 | return true;
|
---|
205 | }
|
---|
206 |
|
---|
207 | // has_children returns true if OID has children
|
---|
208 | bool has_children (const text_t &OID, const text_t &collection,
|
---|
209 | recptproto *collectproto, ostream &logout) {
|
---|
210 |
|
---|
211 | FilterResponse_t response;
|
---|
212 | text_tarray metadata;
|
---|
213 | metadata.push_back ("haschildren");
|
---|
214 |
|
---|
215 | if (get_info (OID, collection, metadata, false, collectproto, response, logout)) {
|
---|
216 | if (response.docInfo[0].metadata[0].values[0] == "1")
|
---|
217 | return true;
|
---|
218 | }
|
---|
219 | return false;
|
---|
220 | }
|
---|
221 |
|
---|
222 |
|
---|
223 | // get_children does a protocol call and returns (in response) the OIDs and
|
---|
224 | // metadata of all the children of OID. The metadata array should be loaded
|
---|
225 | // with whatever metadata fields are to be requested.
|
---|
226 |
|
---|
227 | bool get_children (const text_t &OID, const text_t &collection,
|
---|
228 | const text_tarray &metadata, bool getParents,
|
---|
229 | recptproto *collectproto, FilterResponse_t &response,
|
---|
230 | ostream &logout) {
|
---|
231 |
|
---|
232 | response.clear();
|
---|
233 |
|
---|
234 | comerror_t err;
|
---|
235 | FilterRequest_t request;
|
---|
236 | OptionValue_t option;
|
---|
237 |
|
---|
238 | option.name = "ParentNode";
|
---|
239 | option.value = OID;
|
---|
240 | request.filterOptions.push_back (option);
|
---|
241 | request.filterName = "BrowseFilter";
|
---|
242 | request.filterResultOptions = FROID | FRmetadata;
|
---|
243 | request.fields = metadata;
|
---|
244 | request.getParents = getParents;
|
---|
245 |
|
---|
246 | collectproto->filter (collection, request, response, err, logout);
|
---|
247 |
|
---|
248 | if (err != noError) {
|
---|
249 | outconvertclass text_t2ascii;
|
---|
250 | logout << text_t2ascii
|
---|
251 | << "Error: call to filter failed for " << OID
|
---|
252 | << " in OIDtools::get_children ("
|
---|
253 | << get_comerror_string (err) << ")\n";
|
---|
254 | return false;
|
---|
255 | }
|
---|
256 | return true;
|
---|
257 | }
|
---|
258 |
|
---|
259 | // get_parent returns the parent of the document or classification
|
---|
260 | // specified by OID
|
---|
261 | text_t get_parent (text_t OID) {
|
---|
262 |
|
---|
263 | if (OID.empty() || is_top (OID)) return "";
|
---|
264 |
|
---|
265 | text_t::const_iterator begin = OID.begin();
|
---|
266 | text_t::const_iterator here = (OID.end() - 1);
|
---|
267 |
|
---|
268 | while (here >= begin) {
|
---|
269 | OID.pop_back();
|
---|
270 | if (*here == '.') break;
|
---|
271 | here --;
|
---|
272 | }
|
---|
273 | return OID;
|
---|
274 | }
|
---|
275 |
|
---|
276 | // takes an OID like ".2 and replaces the " with parent
|
---|
277 | void translate_parent (text_t &OID, const text_t &parent) {
|
---|
278 |
|
---|
279 | text_t::const_iterator here = OID.begin();
|
---|
280 | text_t::const_iterator end = OID.end();
|
---|
281 | text_t temp;
|
---|
282 |
|
---|
283 | while (here != end) {
|
---|
284 | if (*here == '"') temp += parent;
|
---|
285 | else temp.push_back (*here);
|
---|
286 | here ++;
|
---|
287 | }
|
---|
288 | OID = temp;
|
---|
289 | }
|
---|
290 |
|
---|
291 | // shrink_parent does the opposite to translate_parent
|
---|
292 | void shrink_parent (text_t &OID) {
|
---|
293 |
|
---|
294 | text_tarray tmp;
|
---|
295 | splitchar (OID.begin(), OID.end(), '.', tmp);
|
---|
296 | OID = "\"." + tmp.back();
|
---|
297 | }
|
---|
298 |
|
---|
299 | // checks if OID uses ".fc", ".lc", ".pr", ".ns",
|
---|
300 | // or ".ps" syntax (first child, last child, parent,
|
---|
301 | // next sibling, previous sibling)
|
---|
302 | bool needs_translating (const text_t &OID) {
|
---|
303 |
|
---|
304 | if (OID.size() < 4) return false;
|
---|
305 |
|
---|
306 | text_t tail = substr (OID.end()-3, OID.end());
|
---|
307 | if (tail == ".fc" || tail == ".lc" || tail == ".pr" ||
|
---|
308 | tail == ".ns" || tail == ".ps") return true;
|
---|
309 |
|
---|
310 | return false;
|
---|
311 | }
|
---|
312 |
|
---|
313 | // strips the ".fc", ".lc", ".pr", ".ns",
|
---|
314 | // or ".ps" suffix from the end of OID
|
---|
315 | void strip_suffix (text_t &OID) {
|
---|
316 |
|
---|
317 | text_t tail = substr (OID.end()-3, OID.end());
|
---|
318 | while (tail == ".fc" || tail == ".lc" || tail == ".pr" ||
|
---|
319 | tail == ".ns" || tail == ".ps") {
|
---|
320 | OID.erase (OID.end()-3, OID.end());
|
---|
321 | tail = substr (OID.end()-3, OID.end());
|
---|
322 | }
|
---|
323 | }
|
---|
324 |
|
---|
325 | void recurse_contents (const ResultDocInfo_t section, const bool &classify,
|
---|
326 | int &totalcols, const text_t &collection,
|
---|
327 | const text_tarray &metadata, recptproto *collectproto,
|
---|
328 | FilterResponse_t &response, ostream &logout) {
|
---|
329 |
|
---|
330 | int metasize = section.metadata.size();
|
---|
331 |
|
---|
332 | int haschildren = section.metadata[metasize-2].values[0].getint();
|
---|
333 | const text_t &doctype = section.metadata[metasize-1].values[0];
|
---|
334 | int cols;
|
---|
335 |
|
---|
336 | if ((haschildren == 1) && ((!classify) || (doctype == "classify"))) {
|
---|
337 | text_t parent = response.docInfo.back().OID;
|
---|
338 | int parentcols = countchar (parent.begin(), parent.end(), '.');
|
---|
339 | FilterResponse_t tmp;
|
---|
340 | bool getParents = false;
|
---|
341 | get_children (section.OID, collection, metadata, getParents, collectproto, tmp, logout);
|
---|
342 | ResultDocInfo_tarray::const_iterator thisdoc = tmp.docInfo.begin();
|
---|
343 | ResultDocInfo_tarray::const_iterator lastdoc = tmp.docInfo.end();
|
---|
344 | while (thisdoc != lastdoc) {
|
---|
345 | if (((*thisdoc).metadata[metasize-1].values[0] != "classify") && (classify))
|
---|
346 | cols = parentcols + 1;
|
---|
347 | else
|
---|
348 | cols = countchar ((*thisdoc).OID.begin(), (*thisdoc).OID.end(), '.');
|
---|
349 | if (cols > totalcols) totalcols = cols;
|
---|
350 | response.docInfo.push_back (*thisdoc);
|
---|
351 | recurse_contents (*thisdoc, classify, totalcols, collection,
|
---|
352 | metadata, collectproto, response, logout);
|
---|
353 | thisdoc ++;
|
---|
354 | }
|
---|
355 | }
|
---|
356 | }
|
---|
357 |
|
---|
358 | // get_contents returns OIDs and metadata of all contents
|
---|
359 | // below (and including) OID.
|
---|
360 | void get_contents (const text_t &topOID, const text_t &classifytype,
|
---|
361 | text_tarray &metadata, int &totalcols,
|
---|
362 | const text_t &collection, recptproto *collectproto,
|
---|
363 | FilterResponse_t &response, ostream &logout) {
|
---|
364 |
|
---|
365 | if (topOID.empty()) return;
|
---|
366 | bool classify = false;
|
---|
367 | response.clear();
|
---|
368 |
|
---|
369 | metadata.push_back("haschildren");
|
---|
370 | metadata.push_back("doctype");
|
---|
371 |
|
---|
372 | // we don't want to recurse all the way down through each document
|
---|
373 | // if we're expanding top level contents
|
---|
374 | if (classifytype == "classify") classify = true;
|
---|
375 |
|
---|
376 | // update totalcols
|
---|
377 | totalcols = countchar (topOID.begin(), topOID.end(), '.');
|
---|
378 |
|
---|
379 | // get topOIDs info
|
---|
380 | if (get_info (topOID, collection, metadata, false, collectproto, response, logout))
|
---|
381 | recurse_contents (response.docInfo[0], classify, totalcols, collection,
|
---|
382 | metadata, collectproto, response, logout);
|
---|
383 | }
|
---|
384 |
|
---|
385 | // is_child_of returns true if OID2 is a child of OID1
|
---|
386 | bool is_child_of(const text_t &OID1, const text_t &OID2) {
|
---|
387 |
|
---|
388 | text_t parent = get_parent(OID2);
|
---|
389 |
|
---|
390 | while (!parent.empty()) {
|
---|
391 | if (parent == OID1) return true;
|
---|
392 | parent = get_parent(parent);
|
---|
393 | }
|
---|
394 | return false;
|
---|
395 | }
|
---|