Lucene.Net
3.0.3
Lucene.Net is a port of the Lucene search engine library, written in C# and targeted at .NET runtime users.
Main Page
Classes
Files
File List
File Members
All
Classes
Namespaces
Files
Functions
Variables
Typedefs
Enumerations
Properties
Pages
contrib
Analyzers
Compound
Hyphenation
HyphenationTree.cs
Go to the documentation of this file.
1
////*
2
// * Licensed to the Apache Software Foundation (ASF) under one or more
3
// * contributor license agreements. See the NOTICE file distributed with
4
// * this work for additional information regarding copyright ownership.
5
// * The ASF licenses this file to You under the Apache License, Version 2.0
6
// * (the "License"); you may not use this file except in compliance with
7
// * the License. You may obtain a copy of the License at
8
// *
9
// * http://www.apache.org/licenses/LICENSE-2.0
10
// *
11
// * Unless required by applicable law or agreed to in writing, software
12
// * distributed under the License is distributed on an "AS IS" BASIS,
13
// * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14
// * See the License for the specific language governing permissions and
15
// * limitations under the License.
16
// */
17
18
//using System;
19
//using System.Collections.Generic;
20
//using System.IO;
21
//using System.Linq;
22
//using System.Text;
23
//using Lucene.Net.Support;
24
25
//namespace Lucene.Net.Analysis.Compound.Hyphenation
26
//{
27
////*
28
// * This tree structure stores the hyphenation patterns in an efficient way for
29
// * fast lookup. It provides the provides the method to hyphenate a word.
30
// *
31
// * This class has been taken from the Apache FOP project (http://xmlgraphics.apache.org/fop/). They have been slightly modified.
32
// */
33
//[Serializable]
34
//public class HyphenationTree : TernaryTree, PatternConsumer
35
//{
36
37
// private static readonly long serialVersionUID = -7842107987915665573L;
38
39
// /*
40
// * value space: stores the interletter values
41
// */
42
// protected ByteVector vspace;
43
44
// /*
45
// * This map stores hyphenation exceptions
46
// */
47
// protected HashMap<String,ArrayList> stoplist;
48
49
// /*
50
// * This map stores the character classes
51
// */
52
// protected TernaryTree classmap;
53
54
// /*
55
// * Temporary map to store interletter values on pattern loading.
56
// */
57
// [NonSerialized]
58
// private TernaryTree ivalues;
59
60
// public HyphenationTree() {
61
// stoplist = new HashMap<string,ArrayList>(23); // usually a small table
62
// classmap = new TernaryTree();
63
// vspace = new ByteVector();
64
// vspace.Alloc(1); // this reserves index 0, which we don't use
65
// }
66
67
// /*
68
// * Packs the values by storing them in 4 bits, two values into a byte Values
69
// * range is from 0 to 9. We use zero as terminator, so we'll add 1 to the
70
// * value.
71
// *
72
// * @param values a string of digits from '0' to '9' representing the
73
// * interletter values.
74
// * @return the index into the vspace array where the packed values are stored.
75
// */
76
// protected int packValues(String values) {
77
// int i, n = values.Length;
78
// int m = (n & 1) == 1 ? (n >> 1) + 2 : (n >> 1) + 1;
79
// int offset = vspace.Alloc(m);
80
// byte[] va = vspace.GetArray();
81
// for (i = 0; i < n; i++) {
82
// int j = i >> 1;
83
// byte v = (byte) ((values[i] - '0' + 1) & 0x0f);
84
// if ((i & 1) == 1) {
85
// va[j + offset] = (byte) (va[j + offset] | v);
86
// } else {
87
// va[j + offset] = (byte) (v << 4); // big endian
88
// }
89
// }
90
// va[m - 1 + offset] = 0; // terminator
91
// return offset;
92
// }
93
94
// protected String unpackValues(int k) {
95
// StringBuilder buf = new StringBuilder();
96
// byte v = vspace.Get(k++);
97
// while (v != 0) {
98
// char c = (char) ((v >>> 4) - 1 + '0');
99
// buf.Append(c);
100
// c = (char) (v & 0x0f);
101
// if (c == 0) {
102
// break;
103
// }
104
// c = (char) (c - 1 + '0');
105
// buf.Append(c);
106
// v = vspace.Get(k++);
107
// }
108
// return buf.ToString();
109
// }
110
111
// /*
112
// * Read hyphenation patterns from an XML file.
113
// *
114
// * @param f the filename
115
// * @throws HyphenationException In case the parsing fails
116
// */
117
// public void loadPatterns(FileInfo f)
118
// {
119
// try {
120
// InputSource src = new InputSource(f.toURL().toExternalForm());
121
// loadPatterns(src);
122
// } catch (MalformedURLException e) {
123
// throw new HyphenationException("Error converting the File '" + f
124
// + "' to a URL: " + e.getMessage());
125
// }
126
// }
127
128
// /*
129
// * Read hyphenation patterns from an XML file.
130
// *
131
// * @param source the InputSource for the file
132
// * @throws HyphenationException In case the parsing fails
133
// */
134
// public void loadPatterns(InputSource source)
135
// {
136
// PatternParser pp = new PatternParser(this);
137
// ivalues = new TernaryTree();
138
139
// pp.parse(source);
140
141
// // patterns/values should be now in the tree
142
// // let's optimize a bit
143
// trimToSize();
144
// vspace.trimToSize();
145
// classmap.trimToSize();
146
147
// // get rid of the auxiliary map
148
// ivalues = null;
149
// }
150
151
// public String findPattern(String pat) {
152
// int k = super.find(pat);
153
// if (k >= 0) {
154
// return unpackValues(k);
155
// }
156
// return "";
157
// }
158
159
// /*
160
// * String compare, returns 0 if equal or t is a substring of s
161
// */
162
// protected int hstrcmp(char[] s, int si, char[] t, int ti) {
163
// for (; s[si] == t[ti]; si++, ti++) {
164
// if (s[si] == 0) {
165
// return 0;
166
// }
167
// }
168
// if (t[ti] == 0) {
169
// return 0;
170
// }
171
// return s[si] - t[ti];
172
// }
173
174
// protected byte[] getValues(int k) {
175
// StringBuilder buf = new StringBuilder();
176
// byte v = vspace.get(k++);
177
// while (v != 0) {
178
// char c = (char) ((v >>> 4) - 1);
179
// buf.append(c);
180
// c = (char) (v & 0x0f);
181
// if (c == 0) {
182
// break;
183
// }
184
// c = (char) (c - 1);
185
// buf.append(c);
186
// v = vspace.get(k++);
187
// }
188
// byte[] res = new byte[buf.length()];
189
// for (int i = 0; i < res.length; i++) {
190
// res[i] = (byte) buf.charAt(i);
191
// }
192
// return res;
193
// }
194
195
// /*
196
// * <p>
197
// * Search for all possible partial matches of word starting at index an update
198
// * interletter values. In other words, it does something like:
199
// * </p>
200
// * <code>
201
// * for(i=0; i<patterns.length; i++) {
202
// * if ( word.substring(index).startsWidth(patterns[i]) )
203
// * update_interletter_values(patterns[i]);
204
// * }
205
// * </code>
206
// * <p>
207
// * But it is done in an efficient way since the patterns are stored in a
208
// * ternary tree. In fact, this is the whole purpose of having the tree: doing
209
// * this search without having to test every single pattern. The number of
210
// * patterns for languages such as English range from 4000 to 10000. Thus,
211
// * doing thousands of string comparisons for each word to hyphenate would be
212
// * really slow without the tree. The tradeoff is memory, but using a ternary
213
// * tree instead of a trie, almost halves the the memory used by Lout or TeX.
214
// * It's also faster than using a hash table
215
// * </p>
216
// *
217
// * @param word null terminated word to match
218
// * @param index start index from word
219
// * @param il interletter values array to update
220
// */
221
// protected void searchPatterns(char[] word, int index, byte[] il) {
222
// byte[] values;
223
// int i = index;
224
// char p, q;
225
// char sp = word[i];
226
// p = root;
227
228
// while (p > 0 && p < sc.length) {
229
// if (sc[p] == 0xFFFF) {
230
// if (hstrcmp(word, i, kv.getArray(), lo[p]) == 0) {
231
// values = getValues(eq[p]); // data pointer is in eq[]
232
// int j = index;
233
// for (int k = 0; k < values.length; k++) {
234
// if (j < il.length && values[k] > il[j]) {
235
// il[j] = values[k];
236
// }
237
// j++;
238
// }
239
// }
240
// return;
241
// }
242
// int d = sp - sc[p];
243
// if (d == 0) {
244
// if (sp == 0) {
245
// break;
246
// }
247
// sp = word[++i];
248
// p = eq[p];
249
// q = p;
250
251
// // look for a pattern ending at this position by searching for
252
// // the null char ( splitchar == 0 )
253
// while (q > 0 && q < sc.length) {
254
// if (sc[q] == 0xFFFF) { // stop at compressed branch
255
// break;
256
// }
257
// if (sc[q] == 0) {
258
// values = getValues(eq[q]);
259
// int j = index;
260
// for (int k = 0; k < values.length; k++) {
261
// if (j < il.length && values[k] > il[j]) {
262
// il[j] = values[k];
263
// }
264
// j++;
265
// }
266
// break;
267
// } else {
268
// q = lo[q];
269
270
// /*
271
// * actually the code should be: q = sc[q] < 0 ? hi[q] : lo[q]; but
272
// * java chars are unsigned
273
// */
274
// }
275
// }
276
// } else {
277
// p = d < 0 ? lo[p] : hi[p];
278
// }
279
// }
280
// }
281
282
// /*
283
// * Hyphenate word and return a Hyphenation object.
284
// *
285
// * @param word the word to be hyphenated
286
// * @param remainCharCount Minimum number of characters allowed before the
287
// * hyphenation point.
288
// * @param pushCharCount Minimum number of characters allowed after the
289
// * hyphenation point.
290
// * @return a {@link Hyphenation Hyphenation} object representing the
291
// * hyphenated word or null if word is not hyphenated.
292
// */
293
// public Hyphenation hyphenate(String word, int remainCharCount,
294
// int pushCharCount) {
295
// char[] w = word.toCharArray();
296
// return hyphenate(w, 0, w.length, remainCharCount, pushCharCount);
297
// }
298
299
// /*
300
// * w = "****nnllllllnnn*****", where n is a non-letter, l is a letter, all n
301
// * may be absent, the first n is at offset, the first l is at offset +
302
// * iIgnoreAtBeginning; word = ".llllll.'\0'***", where all l in w are copied
303
// * into word. In the first part of the routine len = w.length, in the second
304
// * part of the routine len = word.length. Three indices are used: index(w),
305
// * the index in w, index(word), the index in word, letterindex(word), the
306
// * index in the letter part of word. The following relations exist: index(w) =
307
// * offset + i - 1 index(word) = i - iIgnoreAtBeginning letterindex(word) =
308
// * index(word) - 1 (see first loop). It follows that: index(w) - index(word) =
309
// * offset - 1 + iIgnoreAtBeginning index(w) = letterindex(word) + offset +
310
// * iIgnoreAtBeginning
311
// */
312
313
// /*
314
// * Hyphenate word and return an array of hyphenation points.
315
// *
316
// * @param w char array that contains the word
317
// * @param offset Offset to first character in word
318
// * @param len Length of word
319
// * @param remainCharCount Minimum number of characters allowed before the
320
// * hyphenation point.
321
// * @param pushCharCount Minimum number of characters allowed after the
322
// * hyphenation point.
323
// * @return a {@link Hyphenation Hyphenation} object representing the
324
// * hyphenated word or null if word is not hyphenated.
325
// */
326
// public Hyphenation hyphenate(char[] w, int offset, int len,
327
// int remainCharCount, int pushCharCount) {
328
// int i;
329
// char[] word = new char[len + 3];
330
331
// // normalize word
332
// char[] c = new char[2];
333
// int iIgnoreAtBeginning = 0;
334
// int iLength = len;
335
// boolean bEndOfLetters = false;
336
// for (i = 1; i <= len; i++) {
337
// c[0] = w[offset + i - 1];
338
// int nc = classmap.find(c, 0);
339
// if (nc < 0) { // found a non-letter character ...
340
// if (i == (1 + iIgnoreAtBeginning)) {
341
// // ... before any letter character
342
// iIgnoreAtBeginning++;
343
// } else {
344
// // ... after a letter character
345
// bEndOfLetters = true;
346
// }
347
// iLength--;
348
// } else {
349
// if (!bEndOfLetters) {
350
// word[i - iIgnoreAtBeginning] = (char) nc;
351
// } else {
352
// return null;
353
// }
354
// }
355
// }
356
// len = iLength;
357
// if (len < (remainCharCount + pushCharCount)) {
358
// // word is too short to be hyphenated
359
// return null;
360
// }
361
// int[] result = new int[len + 1];
362
// int k = 0;
363
364
// // check exception list first
365
// String sw = new String(word, 1, len);
366
// if (stoplist.containsKey(sw)) {
367
// // assume only simple hyphens (Hyphen.pre="-", Hyphen.post = Hyphen.no =
368
// // null)
369
// ArrayList hw = stoplist.get(sw);
370
// int j = 0;
371
// for (i = 0; i < hw.size(); i++) {
372
// Object o = hw.get(i);
373
// // j = index(sw) = letterindex(word)?
374
// // result[k] = corresponding index(w)
375
// if (o instanceof String) {
376
// j += ((String) o).length();
377
// if (j >= remainCharCount && j < (len - pushCharCount)) {
378
// result[k++] = j + iIgnoreAtBeginning;
379
// }
380
// }
381
// }
382
// } else {
383
// // use algorithm to get hyphenation points
384
// word[0] = '.'; // word start marker
385
// word[len + 1] = '.'; // word end marker
386
// word[len + 2] = 0; // null terminated
387
// byte[] il = new byte[len + 3]; // initialized to zero
388
// for (i = 0; i < len + 1; i++) {
389
// searchPatterns(word, i, il);
390
// }
391
392
// // hyphenation points are located where interletter value is odd
393
// // i is letterindex(word),
394
// // i + 1 is index(word),
395
// // result[k] = corresponding index(w)
396
// for (i = 0; i < len; i++) {
397
// if (((il[i + 1] & 1) == 1) && i >= remainCharCount
398
// && i <= (len - pushCharCount)) {
399
// result[k++] = i + iIgnoreAtBeginning;
400
// }
401
// }
402
// }
403
404
// if (k > 0) {
405
// // trim result array
406
// int[] res = new int[k+2];
407
// System.arraycopy(result, 0, res, 1, k);
408
// // We add the synthetical hyphenation points
409
// // at the beginning and end of the word
410
// res[0]=0;
411
// res[k+1]=len;
412
// return new Hyphenation(res);
413
// } else {
414
// return null;
415
// }
416
// }
417
418
// /*
419
// * Add a character class to the tree. It is used by
420
// * {@link PatternParser PatternParser} as callback to add character classes.
421
// * Character classes define the valid word characters for hyphenation. If a
422
// * word contains a character not defined in any of the classes, it is not
423
// * hyphenated. It also defines a way to normalize the characters in order to
424
// * compare them with the stored patterns. Usually pattern files use only lower
425
// * case characters, in this case a class for letter 'a', for example, should
426
// * be defined as "aA", the first character being the normalization char.
427
// */
428
// public void AddClass(String chargroup) {
429
// if (chargroup.Length > 0) {
430
// char equivChar = chargroup[0];
431
// char[] key = new char[2];
432
// key[1] = (char)0;
433
// for (int i = 0; i < chargroup.Length; i++) {
434
// key[0] = chargroup[i];
435
// classmap.Insert(key, 0, equivChar);
436
// }
437
// }
438
// }
439
440
// /*
441
// * Add an exception to the tree. It is used by
442
// * {@link PatternParser PatternParser} class as callback to store the
443
// * hyphenation exceptions.
444
// *
445
// * @param word normalized word
446
// * @param hyphenatedword a vector of alternating strings and
447
// * {@link Hyphen hyphen} objects.
448
// */
449
// public void AddException(String word, ArrayList hyphenatedword) {
450
// stoplist.Add(word, hyphenatedword);
451
// }
452
453
// /*
454
// * Add a pattern to the tree. Mainly, to be used by
455
// * {@link PatternParser PatternParser} class as callback to add a pattern to
456
// * the tree.
457
// *
458
// * @param pattern the hyphenation pattern
459
// * @param ivalue interletter weight values indicating the desirability and
460
// * priority of hyphenating at a given point within the pattern. It
461
// * should contain only digit characters. (i.e. '0' to '9').
462
// */
463
// public void AddPattern(String pattern, String ivalue) {
464
// int k = ivalues.Find(ivalue);
465
// if (k <= 0) {
466
// k = packValues(ivalue);
467
// ivalues.Insert(ivalue, (char)k);
468
// }
469
// Insert(pattern, (char) k);
470
// }
471
472
// public override void PrintStats() {
473
// Console.WriteLine("Value space size = " + vspace.Length());
474
// base.PrintStats();
475
476
// }
477
//}
478
479
//}
480
Generated on Thu Jan 3 2013 02:34:08 for Lucene.Net by
1.8.3