1 | /* |
---|
2 | * This program is free software; you can redistribute it and/or modify |
---|
3 | * it under the terms of the GNU General Public License as published by |
---|
4 | * the Free Software Foundation; either version 2 of the License, or |
---|
5 | * (at your option) any later version. |
---|
6 | * |
---|
7 | * This program is distributed in the hope that it will be useful, |
---|
8 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
---|
9 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
---|
10 | * GNU General Public License for more details. |
---|
11 | * |
---|
12 | * You should have received a copy of the GNU General Public License |
---|
13 | * along with this program; if not, write to the Free Software |
---|
14 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. |
---|
15 | */ |
---|
16 | |
---|
17 | /* |
---|
18 | * AprioriItemSet.java |
---|
19 | * Copyright (C) 2004 University of Waikato, Hamilton, New Zealand |
---|
20 | * |
---|
21 | */ |
---|
22 | |
---|
23 | package weka.associations; |
---|
24 | |
---|
25 | import weka.core.ContingencyTables; |
---|
26 | import weka.core.FastVector; |
---|
27 | import weka.core.Instances; |
---|
28 | import weka.core.RevisionHandler; |
---|
29 | import weka.core.RevisionUtils; |
---|
30 | |
---|
31 | import java.io.Serializable; |
---|
32 | import java.util.Enumeration; |
---|
33 | import java.util.Hashtable; |
---|
34 | |
---|
35 | |
---|
36 | /** |
---|
37 | * Class for storing a set of items. Item sets are stored in a lexicographic |
---|
38 | * order, which is determined by the header information of the set of instances |
---|
39 | * used for generating the set of items. All methods in this class assume that |
---|
40 | * item sets are stored in lexicographic order. |
---|
41 | * The class provides methods that are used in the Apriori algorithm to construct |
---|
42 | * association rules. |
---|
43 | * |
---|
44 | * @author Eibe Frank (eibe@cs.waikato.ac.nz) |
---|
45 | * @author Stefan Mutter (mutter@cs.waikato.ac.nz) |
---|
46 | * @version $Revision: 5130 $ |
---|
47 | */ |
---|
48 | public class AprioriItemSet |
---|
49 | extends ItemSet |
---|
50 | implements Serializable, RevisionHandler { |
---|
51 | |
---|
52 | /** for serialization */ |
---|
53 | static final long serialVersionUID = 7684467755712672058L; |
---|
54 | |
---|
55 | /** |
---|
56 | * Constructor |
---|
57 | * |
---|
58 | * @param totalTrans the total number of transactions in the data |
---|
59 | */ |
---|
60 | public AprioriItemSet(int totalTrans) { |
---|
61 | super(totalTrans); |
---|
62 | } |
---|
63 | |
---|
64 | |
---|
65 | /** |
---|
66 | * Outputs the confidence for a rule. |
---|
67 | * |
---|
68 | * @param premise the premise of the rule |
---|
69 | * @param consequence the consequence of the rule |
---|
70 | * @return the confidence on the training data |
---|
71 | */ |
---|
72 | public static double confidenceForRule(AprioriItemSet premise, |
---|
73 | AprioriItemSet consequence) { |
---|
74 | |
---|
75 | return (double)consequence.m_counter/(double)premise.m_counter; |
---|
76 | } |
---|
77 | |
---|
78 | /** |
---|
79 | * Outputs the lift for a rule. Lift is defined as:<br> |
---|
80 | * confidence / prob(consequence) |
---|
81 | * |
---|
82 | * @param premise the premise of the rule |
---|
83 | * @param consequence the consequence of the rule |
---|
84 | * @param consequenceCount how many times the consequence occurs independent |
---|
85 | * of the premise |
---|
86 | * @return the lift on the training data |
---|
87 | */ |
---|
88 | public double liftForRule(AprioriItemSet premise, |
---|
89 | AprioriItemSet consequence, |
---|
90 | int consequenceCount) { |
---|
91 | double confidence = confidenceForRule(premise, consequence); |
---|
92 | |
---|
93 | return confidence / ((double)consequenceCount / |
---|
94 | (double)m_totalTransactions); |
---|
95 | } |
---|
96 | |
---|
97 | /** |
---|
98 | * Outputs the leverage for a rule. Leverage is defined as: <br> |
---|
99 | * prob(premise & consequence) - (prob(premise) * prob(consequence)) |
---|
100 | * |
---|
101 | * @param premise the premise of the rule |
---|
102 | * @param consequence the consequence of the rule |
---|
103 | * @param premiseCount how many times the premise occurs independent |
---|
104 | * of the consequent |
---|
105 | * @param consequenceCount how many times the consequence occurs independent |
---|
106 | * of the premise |
---|
107 | * @return the leverage on the training data |
---|
108 | */ |
---|
109 | public double leverageForRule(AprioriItemSet premise, |
---|
110 | AprioriItemSet consequence, |
---|
111 | int premiseCount, |
---|
112 | int consequenceCount) { |
---|
113 | double coverageForItemSet = (double)consequence.m_counter / |
---|
114 | (double)m_totalTransactions; |
---|
115 | double expectedCoverageIfIndependent = |
---|
116 | ((double)premiseCount / (double)m_totalTransactions) * |
---|
117 | ((double)consequenceCount / (double)m_totalTransactions); |
---|
118 | double lev = coverageForItemSet - expectedCoverageIfIndependent; |
---|
119 | return lev; |
---|
120 | } |
---|
121 | |
---|
122 | /** |
---|
123 | * Outputs the conviction for a rule. Conviction is defined as: <br> |
---|
124 | * prob(premise) * prob(!consequence) / prob(premise & !consequence) |
---|
125 | * |
---|
126 | * @param premise the premise of the rule |
---|
127 | * @param consequence the consequence of the rule |
---|
128 | * @param premiseCount how many times the premise occurs independent |
---|
129 | * of the consequent |
---|
130 | * @param consequenceCount how many times the consequence occurs independent |
---|
131 | * of the premise |
---|
132 | * @return the conviction on the training data |
---|
133 | */ |
---|
134 | public double convictionForRule(AprioriItemSet premise, |
---|
135 | AprioriItemSet consequence, |
---|
136 | int premiseCount, |
---|
137 | int consequenceCount) { |
---|
138 | double num = |
---|
139 | (double)premiseCount * (double)(m_totalTransactions - consequenceCount) / |
---|
140 | (double)m_totalTransactions; |
---|
141 | double denom = |
---|
142 | ((premiseCount - consequence.m_counter)+1); |
---|
143 | |
---|
144 | if (num < 0 || denom < 0) { |
---|
145 | System.err.println("*** "+num+" "+denom); |
---|
146 | System.err.println("premis count: "+premiseCount+" consequence count "+consequenceCount+" total trans "+m_totalTransactions); |
---|
147 | } |
---|
148 | return num / denom; |
---|
149 | } |
---|
150 | |
---|
151 | |
---|
152 | |
---|
153 | /** |
---|
154 | * Generates all rules for an item set. |
---|
155 | * |
---|
156 | * @param minConfidence the minimum confidence the rules have to have |
---|
157 | * @param hashtables containing all(!) previously generated |
---|
158 | * item sets |
---|
159 | * @param numItemsInSet the size of the item set for which the rules |
---|
160 | * are to be generated |
---|
161 | * @return all the rules with minimum confidence for the given item set |
---|
162 | */ |
---|
163 | public FastVector[] generateRules(double minConfidence, |
---|
164 | FastVector hashtables, |
---|
165 | int numItemsInSet) { |
---|
166 | |
---|
167 | FastVector premises = new FastVector(),consequences = new FastVector(), |
---|
168 | conf = new FastVector(); |
---|
169 | FastVector[] rules = new FastVector[3], moreResults; |
---|
170 | AprioriItemSet premise, consequence; |
---|
171 | Hashtable hashtable = (Hashtable)hashtables.elementAt(numItemsInSet - 2); |
---|
172 | |
---|
173 | // Generate all rules with one item in the consequence. |
---|
174 | for (int i = 0; i < m_items.length; i++) |
---|
175 | if (m_items[i] != -1) { |
---|
176 | premise = new AprioriItemSet(m_totalTransactions); |
---|
177 | consequence = new AprioriItemSet(m_totalTransactions); |
---|
178 | premise.m_items = new int[m_items.length]; |
---|
179 | consequence.m_items = new int[m_items.length]; |
---|
180 | consequence.m_counter = m_counter; |
---|
181 | |
---|
182 | for (int j = 0; j < m_items.length; j++) |
---|
183 | consequence.m_items[j] = -1; |
---|
184 | System.arraycopy(m_items, 0, premise.m_items, 0, m_items.length); |
---|
185 | premise.m_items[i] = -1; |
---|
186 | |
---|
187 | consequence.m_items[i] = m_items[i]; |
---|
188 | premise.m_counter = ((Integer)hashtable.get(premise)).intValue(); |
---|
189 | premises.addElement(premise); |
---|
190 | consequences.addElement(consequence); |
---|
191 | conf.addElement(new Double(confidenceForRule(premise, consequence))); |
---|
192 | } |
---|
193 | rules[0] = premises; |
---|
194 | rules[1] = consequences; |
---|
195 | rules[2] = conf; |
---|
196 | pruneRules(rules, minConfidence); |
---|
197 | |
---|
198 | // Generate all the other rules |
---|
199 | moreResults = moreComplexRules(rules, numItemsInSet, 1, minConfidence, |
---|
200 | hashtables); |
---|
201 | if (moreResults != null) |
---|
202 | for (int i = 0; i < moreResults[0].size(); i++) { |
---|
203 | rules[0].addElement(moreResults[0].elementAt(i)); |
---|
204 | rules[1].addElement(moreResults[1].elementAt(i)); |
---|
205 | rules[2].addElement(moreResults[2].elementAt(i)); |
---|
206 | } |
---|
207 | return rules; |
---|
208 | } |
---|
209 | |
---|
210 | |
---|
211 | |
---|
212 | /** |
---|
213 | * Generates all significant rules for an item set. |
---|
214 | * |
---|
215 | * @param minMetric the minimum metric (confidence, lift, leverage, |
---|
216 | * improvement) the rules have to have |
---|
217 | * @param metricType (confidence=0, lift, leverage, improvement) |
---|
218 | * @param hashtables containing all(!) previously generated |
---|
219 | * item sets |
---|
220 | * @param numItemsInSet the size of the item set for which the rules |
---|
221 | * are to be generated |
---|
222 | * @param numTransactions |
---|
223 | * @param significanceLevel the significance level for testing the rules |
---|
224 | * @return all the rules with minimum metric for the given item set |
---|
225 | * @exception Exception if something goes wrong |
---|
226 | */ |
---|
227 | public final FastVector[] generateRulesBruteForce(double minMetric, |
---|
228 | int metricType, |
---|
229 | FastVector hashtables, |
---|
230 | int numItemsInSet, |
---|
231 | int numTransactions, |
---|
232 | double significanceLevel) |
---|
233 | throws Exception { |
---|
234 | |
---|
235 | FastVector premises = new FastVector(),consequences = new FastVector(), |
---|
236 | conf = new FastVector(), lift = new FastVector(), lev = new FastVector(), |
---|
237 | conv = new FastVector(); |
---|
238 | FastVector[] rules = new FastVector[6]; |
---|
239 | AprioriItemSet premise, consequence; |
---|
240 | Hashtable hashtableForPremise, hashtableForConsequence; |
---|
241 | int numItemsInPremise, help, max, consequenceUnconditionedCounter; |
---|
242 | double[][] contingencyTable = new double[2][2]; |
---|
243 | double metric, chiSquared; |
---|
244 | |
---|
245 | // Generate all possible rules for this item set and test their |
---|
246 | // significance. |
---|
247 | max = (int)Math.pow(2, numItemsInSet); |
---|
248 | for (int j = 1; j < max; j++) { |
---|
249 | numItemsInPremise = 0; |
---|
250 | help = j; |
---|
251 | while (help > 0) { |
---|
252 | if (help % 2 == 1) |
---|
253 | numItemsInPremise++; |
---|
254 | help /= 2; |
---|
255 | } |
---|
256 | if (numItemsInPremise < numItemsInSet) { |
---|
257 | hashtableForPremise = |
---|
258 | (Hashtable)hashtables.elementAt(numItemsInPremise-1); |
---|
259 | hashtableForConsequence = |
---|
260 | (Hashtable)hashtables.elementAt(numItemsInSet-numItemsInPremise-1); |
---|
261 | premise = new AprioriItemSet(m_totalTransactions); |
---|
262 | consequence = new AprioriItemSet(m_totalTransactions); |
---|
263 | premise.m_items = new int[m_items.length]; |
---|
264 | |
---|
265 | consequence.m_items = new int[m_items.length]; |
---|
266 | consequence.m_counter = m_counter; |
---|
267 | help = j; |
---|
268 | for (int i = 0; i < m_items.length; i++) |
---|
269 | if (m_items[i] != -1) { |
---|
270 | if (help % 2 == 1) { |
---|
271 | premise.m_items[i] = m_items[i]; |
---|
272 | consequence.m_items[i] = -1; |
---|
273 | } else { |
---|
274 | premise.m_items[i] = -1; |
---|
275 | consequence.m_items[i] = m_items[i]; |
---|
276 | } |
---|
277 | help /= 2; |
---|
278 | } else { |
---|
279 | premise.m_items[i] = -1; |
---|
280 | consequence.m_items[i] = -1; |
---|
281 | } |
---|
282 | premise.m_counter = ((Integer)hashtableForPremise.get(premise)).intValue(); |
---|
283 | consequenceUnconditionedCounter = |
---|
284 | ((Integer)hashtableForConsequence.get(consequence)).intValue(); |
---|
285 | |
---|
286 | if (metricType == 0) { |
---|
287 | contingencyTable[0][0] = (double)(consequence.m_counter); |
---|
288 | contingencyTable[0][1] = (double)(premise.m_counter - consequence.m_counter); |
---|
289 | contingencyTable[1][0] = (double)(consequenceUnconditionedCounter - |
---|
290 | consequence.m_counter); |
---|
291 | contingencyTable[1][1] = (double)(numTransactions - premise.m_counter - |
---|
292 | consequenceUnconditionedCounter + |
---|
293 | consequence.m_counter); |
---|
294 | chiSquared = ContingencyTables.chiSquared(contingencyTable, false); |
---|
295 | |
---|
296 | metric = confidenceForRule(premise, consequence); |
---|
297 | |
---|
298 | if ((!(metric < minMetric)) && |
---|
299 | (!(chiSquared > significanceLevel))) { |
---|
300 | premises.addElement(premise); |
---|
301 | consequences.addElement(consequence); |
---|
302 | conf.addElement(new Double(metric)); |
---|
303 | lift.addElement(new Double(liftForRule(premise, consequence, |
---|
304 | consequenceUnconditionedCounter))); |
---|
305 | lev.addElement(new Double(leverageForRule(premise, consequence, |
---|
306 | premise.m_counter, |
---|
307 | consequenceUnconditionedCounter))); |
---|
308 | conv.addElement(new Double(convictionForRule(premise, consequence, |
---|
309 | premise.m_counter, |
---|
310 | consequenceUnconditionedCounter))); |
---|
311 | } |
---|
312 | } else { |
---|
313 | double tempConf = confidenceForRule(premise, consequence); |
---|
314 | double tempLift = liftForRule(premise, consequence, |
---|
315 | consequenceUnconditionedCounter); |
---|
316 | double tempLev = leverageForRule(premise, consequence, |
---|
317 | premise.m_counter, |
---|
318 | consequenceUnconditionedCounter); |
---|
319 | double tempConv = convictionForRule(premise, consequence, |
---|
320 | premise.m_counter, |
---|
321 | consequenceUnconditionedCounter); |
---|
322 | switch(metricType) { |
---|
323 | case 1: |
---|
324 | metric = tempLift; |
---|
325 | break; |
---|
326 | case 2: |
---|
327 | metric = tempLev; |
---|
328 | break; |
---|
329 | case 3: |
---|
330 | metric = tempConv; |
---|
331 | break; |
---|
332 | default: |
---|
333 | throw new Exception("ItemSet: Unknown metric type!"); |
---|
334 | } |
---|
335 | if (!(metric < minMetric)) { |
---|
336 | premises.addElement(premise); |
---|
337 | consequences.addElement(consequence); |
---|
338 | conf.addElement(new Double(tempConf)); |
---|
339 | lift.addElement(new Double(tempLift)); |
---|
340 | lev.addElement(new Double(tempLev)); |
---|
341 | conv.addElement(new Double(tempConv)); |
---|
342 | } |
---|
343 | } |
---|
344 | } |
---|
345 | } |
---|
346 | rules[0] = premises; |
---|
347 | rules[1] = consequences; |
---|
348 | rules[2] = conf; |
---|
349 | rules[3] = lift; |
---|
350 | rules[4] = lev; |
---|
351 | rules[5] = conv; |
---|
352 | return rules; |
---|
353 | } |
---|
354 | |
---|
355 | /** |
---|
356 | * Subtracts an item set from another one. |
---|
357 | * |
---|
358 | * @param toSubtract the item set to be subtracted from this one. |
---|
359 | * @return an item set that only contains items form this item sets that |
---|
360 | * are not contained by toSubtract |
---|
361 | */ |
---|
362 | public final AprioriItemSet subtract(AprioriItemSet toSubtract) { |
---|
363 | |
---|
364 | AprioriItemSet result = new AprioriItemSet(m_totalTransactions); |
---|
365 | |
---|
366 | result.m_items = new int[m_items.length]; |
---|
367 | |
---|
368 | for (int i = 0; i < m_items.length; i++) |
---|
369 | if (toSubtract.m_items[i] == -1) |
---|
370 | result.m_items[i] = m_items[i]; |
---|
371 | else |
---|
372 | result.m_items[i] = -1; |
---|
373 | result.m_counter = 0; |
---|
374 | return result; |
---|
375 | } |
---|
376 | |
---|
377 | |
---|
378 | /** |
---|
379 | * Generates rules with more than one item in the consequence. |
---|
380 | * |
---|
381 | * @param rules all the rules having (k-1)-item sets as consequences |
---|
382 | * @param numItemsInSet the size of the item set for which the rules |
---|
383 | * are to be generated |
---|
384 | * @param numItemsInConsequence the value of (k-1) |
---|
385 | * @param minConfidence the minimum confidence a rule has to have |
---|
386 | * @param hashtables the hashtables containing all(!) previously generated |
---|
387 | * item sets |
---|
388 | * @return all the rules having (k)-item sets as consequences |
---|
389 | */ |
---|
390 | private final FastVector[] moreComplexRules(FastVector[] rules, |
---|
391 | int numItemsInSet, |
---|
392 | int numItemsInConsequence, |
---|
393 | double minConfidence, |
---|
394 | FastVector hashtables) { |
---|
395 | |
---|
396 | AprioriItemSet newPremise; |
---|
397 | FastVector[] result, moreResults; |
---|
398 | FastVector newConsequences, newPremises = new FastVector(), |
---|
399 | newConf = new FastVector(); |
---|
400 | Hashtable hashtable; |
---|
401 | |
---|
402 | if (numItemsInSet > numItemsInConsequence + 1) { |
---|
403 | hashtable = |
---|
404 | (Hashtable)hashtables.elementAt(numItemsInSet - numItemsInConsequence - 2); |
---|
405 | newConsequences = mergeAllItemSets(rules[1], |
---|
406 | numItemsInConsequence - 1, |
---|
407 | m_totalTransactions); |
---|
408 | Enumeration enu = newConsequences.elements(); |
---|
409 | while (enu.hasMoreElements()) { |
---|
410 | AprioriItemSet current = (AprioriItemSet)enu.nextElement(); |
---|
411 | current.m_counter = m_counter; |
---|
412 | newPremise = subtract(current); |
---|
413 | newPremise.m_counter = ((Integer)hashtable.get(newPremise)).intValue(); |
---|
414 | newPremises.addElement(newPremise); |
---|
415 | newConf.addElement(new Double(confidenceForRule(newPremise, current))); |
---|
416 | } |
---|
417 | result = new FastVector[3]; |
---|
418 | result[0] = newPremises; |
---|
419 | result[1] = newConsequences; |
---|
420 | result[2] = newConf; |
---|
421 | pruneRules(result, minConfidence); |
---|
422 | moreResults = moreComplexRules(result,numItemsInSet,numItemsInConsequence+1, |
---|
423 | minConfidence, hashtables); |
---|
424 | if (moreResults != null) |
---|
425 | for (int i = 0; i < moreResults[0].size(); i++) { |
---|
426 | result[0].addElement(moreResults[0].elementAt(i)); |
---|
427 | result[1].addElement(moreResults[1].elementAt(i)); |
---|
428 | result[2].addElement(moreResults[2].elementAt(i)); |
---|
429 | } |
---|
430 | return result; |
---|
431 | } else |
---|
432 | return null; |
---|
433 | } |
---|
434 | |
---|
435 | |
---|
436 | /** |
---|
437 | * Returns the contents of an item set as a string. |
---|
438 | * |
---|
439 | * @param instances contains the relevant header information |
---|
440 | * @return string describing the item set |
---|
441 | */ |
---|
442 | public final String toString(Instances instances) { |
---|
443 | |
---|
444 | return super.toString(instances); |
---|
445 | } |
---|
446 | |
---|
447 | /** |
---|
448 | * Converts the header info of the given set of instances into a set |
---|
449 | * of item sets (singletons). The ordering of values in the header file |
---|
450 | * determines the lexicographic order. |
---|
451 | * |
---|
452 | * @param instances the set of instances whose header info is to be used |
---|
453 | * @return a set of item sets, each containing a single item |
---|
454 | * @exception Exception if singletons can't be generated successfully |
---|
455 | */ |
---|
456 | public static FastVector singletons(Instances instances, |
---|
457 | boolean treatZeroAsMissing) throws Exception { |
---|
458 | |
---|
459 | FastVector setOfItemSets = new FastVector(); |
---|
460 | ItemSet current; |
---|
461 | |
---|
462 | for (int i = 0; i < instances.numAttributes(); i++) { |
---|
463 | if (instances.attribute(i).isNumeric()) |
---|
464 | throw new Exception("Can't handle numeric attributes!"); |
---|
465 | int j = (treatZeroAsMissing) ? 1 : 0; |
---|
466 | for (; j < instances.attribute(i).numValues(); j++) { |
---|
467 | current = new AprioriItemSet(instances.numInstances()); |
---|
468 | current.setTreatZeroAsMissing(treatZeroAsMissing); |
---|
469 | current.m_items = new int[instances.numAttributes()]; |
---|
470 | for (int k = 0; k < instances.numAttributes(); k++) |
---|
471 | current.m_items[k] = -1; |
---|
472 | current.m_items[i] = j; |
---|
473 | setOfItemSets.addElement(current); |
---|
474 | } |
---|
475 | } |
---|
476 | return setOfItemSets; |
---|
477 | } |
---|
478 | |
---|
479 | /** |
---|
480 | * Merges all item sets in the set of (k-1)-item sets |
---|
481 | * to create the (k)-item sets and updates the counters. |
---|
482 | * |
---|
483 | * @param itemSets the set of (k-1)-item sets |
---|
484 | * @param size the value of (k-1) |
---|
485 | * @param totalTrans the total number of transactions in the data |
---|
486 | * @return the generated (k)-item sets |
---|
487 | */ |
---|
488 | public static FastVector mergeAllItemSets(FastVector itemSets, int size, |
---|
489 | int totalTrans) { |
---|
490 | |
---|
491 | FastVector newVector = new FastVector(); |
---|
492 | ItemSet result; |
---|
493 | int numFound, k; |
---|
494 | |
---|
495 | for (int i = 0; i < itemSets.size(); i++) { |
---|
496 | ItemSet first = (ItemSet)itemSets.elementAt(i); |
---|
497 | out: |
---|
498 | for (int j = i+1; j < itemSets.size(); j++) { |
---|
499 | ItemSet second = (ItemSet)itemSets.elementAt(j); |
---|
500 | result = new AprioriItemSet(totalTrans); |
---|
501 | result.m_items = new int[first.m_items.length]; |
---|
502 | |
---|
503 | // Find and copy common prefix of size 'size' |
---|
504 | numFound = 0; |
---|
505 | k = 0; |
---|
506 | while (numFound < size) { |
---|
507 | if (first.m_items[k] == second.m_items[k]) { |
---|
508 | if (first.m_items[k] != -1) |
---|
509 | numFound++; |
---|
510 | result.m_items[k] = first.m_items[k]; |
---|
511 | } else |
---|
512 | break out; |
---|
513 | k++; |
---|
514 | } |
---|
515 | |
---|
516 | // Check difference |
---|
517 | while (k < first.m_items.length) { |
---|
518 | if ((first.m_items[k] != -1) && (second.m_items[k] != -1)) |
---|
519 | break; |
---|
520 | else { |
---|
521 | if (first.m_items[k] != -1) |
---|
522 | result.m_items[k] = first.m_items[k]; |
---|
523 | else |
---|
524 | result.m_items[k] = second.m_items[k]; |
---|
525 | } |
---|
526 | k++; |
---|
527 | } |
---|
528 | if (k == first.m_items.length) { |
---|
529 | result.m_counter = 0; |
---|
530 | newVector.addElement(result); |
---|
531 | } |
---|
532 | } |
---|
533 | } |
---|
534 | return newVector; |
---|
535 | } |
---|
536 | |
---|
537 | /** |
---|
538 | * Returns the revision string. |
---|
539 | * |
---|
540 | * @return the revision |
---|
541 | */ |
---|
542 | public String getRevision() { |
---|
543 | return RevisionUtils.extract("$Revision: 5130 $"); |
---|
544 | } |
---|
545 | } |
---|