Mercurial > hg > truffle

comparison test/compiler/7070134/Stemmer.java @ 3840:4e761e7e6e12

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
7070134: Hotspot crashes with sigsegv from PorterStemmer Summary: Do not move data nodes which are attached to a predicate test to a dominating test. Reviewed-by: never
author kvn
date Tue, 26 Jul 2011 19:35:23 -0700
parents
children 3a97daec1b34
comparison
equal deleted inserted replaced
3839:3d42f82cd811 3840:4e761e7e6e12
1 /**
2 * @test
3 * @bug 7070134
4 * @summary Hotspot crashes with sigsegv from PorterStemmer
5 *
6 * @run shell Test7070134.sh
7 */
8
9 /*
10
11 Porter stemmer in Java. The original paper is in
12
13 Porter, 1980, An algorithm for suffix stripping, Program, Vol. 14,
14 no. 3, pp 130-137,
15
16 See also http://www.tartarus.org/~martin/PorterStemmer
17
18 History:
19
20 Release 1
21
22 Bug 1 (reported by Gonzalo Parra 16/10/99) fixed as marked below.
23 The words 'aed', 'eed', 'oed' leave k at 'a' for step 3, and b[k-1]
24 is then out outside the bounds of b.
25
26 Release 2
27
28 Similarly,
29
30 Bug 2 (reported by Steve Dyrdahl 22/2/00) fixed as marked below.
31 'ion' by itself leaves j = -1 in the test for 'ion' in step 5, and
32 b[j] is then outside the bounds of b.
33
34 Release 3
35
36 Considerably revised 4/9/00 in the light of many helpful suggestions
37 from Brian Goetz of Quiotix Corporation (brian@quiotix.com).
38
39 Release 4
40
41 */
42
43 import java.io.*;
44
45 /**
46 * Stemmer, implementing the Porter Stemming Algorithm
47 *
48 * The Stemmer class transforms a word into its root form. The input
49 * word can be provided a character at time (by calling add()), or at once
50 * by calling one of the various stem(something) methods.
51 */
52
53 class Stemmer
54 { private char[] b;
55 private int i, /* offset into b */
56 i_end, /* offset to end of stemmed word */
57 j, k;
58 private static final int INC = 50;
59 /* unit of size whereby b is increased */
60 public Stemmer()
61 { b = new char[INC];
62 i = 0;
63 i_end = 0;
64 }
65
66 /**
67 * Add a character to the word being stemmed. When you are finished
68 * adding characters, you can call stem(void) to stem the word.
69 */
70
71 public void add(char ch)
72 { if (i == b.length)
73 { char[] new_b = new char[i+INC];
74 for (int c = 0; c < i; c++) new_b[c] = b[c];
75 b = new_b;
76 }
77 b[i++] = ch;
78 }
79
80
81 /** Adds wLen characters to the word being stemmed contained in a portion
82 * of a char[] array. This is like repeated calls of add(char ch), but
83 * faster.
84 */
85
86 public void add(char[] w, int wLen)
87 { if (i+wLen >= b.length)
88 { char[] new_b = new char[i+wLen+INC];
89 for (int c = 0; c < i; c++) new_b[c] = b[c];
90 b = new_b;
91 }
92 for (int c = 0; c < wLen; c++) b[i++] = w[c];
93 }
94
95 /**
96 * After a word has been stemmed, it can be retrieved by toString(),
97 * or a reference to the internal buffer can be retrieved by getResultBuffer
98 * and getResultLength (which is generally more efficient.)
99 */
100 public String toString() { return new String(b,0,i_end); }
101
102 /**
103 * Returns the length of the word resulting from the stemming process.
104 */
105 public int getResultLength() { return i_end; }
106
107 /**
108 * Returns a reference to a character buffer containing the results of
109 * the stemming process. You also need to consult getResultLength()
110 * to determine the length of the result.
111 */
112 public char[] getResultBuffer() { return b; }
113
114 /* cons(i) is true <=> b[i] is a consonant. */
115
116 private final boolean cons(int i)
117 { switch (b[i])
118 { case 'a': case 'e': case 'i': case 'o': case 'u': return false;
119 case 'y': return (i==0) ? true : !cons(i-1);
120 default: return true;
121 }
122 }
123
124 /* m() measures the number of consonant sequences between 0 and j. if c is
125 a consonant sequence and v a vowel sequence, and <..> indicates arbitrary
126 presence,
127
128 <c><v> gives 0
129 <c>vc<v> gives 1
130 <c>vcvc<v> gives 2
131 <c>vcvcvc<v> gives 3
132 ....
133 */
134
135 private final int m()
136 { int n = 0;
137 int i = 0;
138 while(true)
139 { if (i > j) return n;
140 if (! cons(i)) break; i++;
141 }
142 i++;
143 while(true)
144 { while(true)
145 { if (i > j) return n;
146 if (cons(i)) break;
147 i++;
148 }
149 i++;
150 n++;
151 while(true)
152 { if (i > j) return n;
153 if (! cons(i)) break;
154 i++;
155 }
156 i++;
157 }
158 }
159
160 /* vowelinstem() is true <=> 0,...j contains a vowel */
161
162 private final boolean vowelinstem()
163 { int i; for (i = 0; i <= j; i++) if (! cons(i)) return true;
164 return false;
165 }
166
167 /* doublec(j) is true <=> j,(j-1) contain a double consonant. */
168
169 private final boolean doublec(int j)
170 { if (j < 1) return false;
171 if (b[j] != b[j-1]) return false;
172 return cons(j);
173 }
174
175 /* cvc(i) is true <=> i-2,i-1,i has the form consonant - vowel - consonant
176 and also if the second c is not w,x or y. this is used when trying to
177 restore an e at the end of a short word. e.g.
178
179 cav(e), lov(e), hop(e), crim(e), but
180 snow, box, tray.
181
182 */
183
184 private final boolean cvc(int i)
185 { if (i < 2 || !cons(i) || cons(i-1) || !cons(i-2)) return false;
186 { int ch = b[i];
187 if (ch == 'w' || ch == 'x' || ch == 'y') return false;
188 }
189 return true;
190 }
191
192 private final boolean ends(String s)
193 { int l = s.length();
194 int o = k-l+1;
195 if (o < 0) return false;
196 for (int i = 0; i < l; i++) if (b[o+i] != s.charAt(i)) return false;
197 j = k-l;
198 return true;
199 }
200
201 /* setto(s) sets (j+1),...k to the characters in the string s, readjusting
202 k. */
203
204 private final void setto(String s)
205 { int l = s.length();
206 int o = j+1;
207 for (int i = 0; i < l; i++) b[o+i] = s.charAt(i);
208 k = j+l;
209 }
210
211 /* r(s) is used further down. */
212
213 private final void r(String s) { if (m() > 0) setto(s); }
214
215 /* step1() gets rid of plurals and -ed or -ing. e.g.
216
217 caresses -> caress
218 ponies -> poni
219 ties -> ti
220 caress -> caress
221 cats -> cat
222
223 feed -> feed
224 agreed -> agree
225 disabled -> disable
226
227 matting -> mat
228 mating -> mate
229 meeting -> meet
230 milling -> mill
231 messing -> mess
232
233 meetings -> meet
234
235 */
236
237 private final void step1()
238 { if (b[k] == 's')
239 { if (ends("sses")) k -= 2; else
240 if (ends("ies")) setto("i"); else
241 if (b[k-1] != 's') k--;
242 }
243 if (ends("eed")) { if (m() > 0) k--; } else
244 if ((ends("ed") || ends("ing")) && vowelinstem())
245 { k = j;
246 if (ends("at")) setto("ate"); else
247 if (ends("bl")) setto("ble"); else
248 if (ends("iz")) setto("ize"); else
249 if (doublec(k))
250 { k--;
251 { int ch = b[k];
252 if (ch == 'l' || ch == 's' || ch == 'z') k++;
253 }
254 }
255 else if (m() == 1 && cvc(k)) setto("e");
256 }
257 }
258
259 /* step2() turns terminal y to i when there is another vowel in the stem. */
260
261 private final void step2() { if (ends("y") && vowelinstem()) b[k] = 'i'; }
262
263 /* step3() maps double suffices to single ones. so -ization ( = -ize plus
264 -ation) maps to -ize etc. note that the string before the suffix must give
265 m() > 0. */
266
267 private final void step3() { if (k == 0) return; /* For Bug 1 */ switch (b[k-1])
268 {
269 case 'a': if (ends("ational")) { r("ate"); break; }
270 if (ends("tional")) { r("tion"); break; }
271 break;
272 case 'c': if (ends("enci")) { r("ence"); break; }
273 if (ends("anci")) { r("ance"); break; }
274 break;
275 case 'e': if (ends("izer")) { r("ize"); break; }
276 break;
277 case 'l': if (ends("bli")) { r("ble"); break; }
278 if (ends("alli")) { r("al"); break; }
279 if (ends("entli")) { r("ent"); break; }
280 if (ends("eli")) { r("e"); break; }
281 if (ends("ousli")) { r("ous"); break; }
282 break;
283 case 'o': if (ends("ization")) { r("ize"); break; }
284 if (ends("ation")) { r("ate"); break; }
285 if (ends("ator")) { r("ate"); break; }
286 break;
287 case 's': if (ends("alism")) { r("al"); break; }
288 if (ends("iveness")) { r("ive"); break; }
289 if (ends("fulness")) { r("ful"); break; }
290 if (ends("ousness")) { r("ous"); break; }
291 break;
292 case 't': if (ends("aliti")) { r("al"); break; }
293 if (ends("iviti")) { r("ive"); break; }
294 if (ends("biliti")) { r("ble"); break; }
295 break;
296 case 'g': if (ends("logi")) { r("log"); break; }
297 } }
298
299 /* step4() deals with -ic-, -full, -ness etc. similar strategy to step3. */
300
301 private final void step4() { switch (b[k])
302 {
303 case 'e': if (ends("icate")) { r("ic"); break; }
304 if (ends("ative")) { r(""); break; }
305 if (ends("alize")) { r("al"); break; }
306 break;
307 case 'i': if (ends("iciti")) { r("ic"); break; }
308 break;
309 case 'l': if (ends("ical")) { r("ic"); break; }
310 if (ends("ful")) { r(""); break; }
311 break;
312 case 's': if (ends("ness")) { r(""); break; }
313 break;
314 } }
315
316 /* step5() takes off -ant, -ence etc., in context <c>vcvc<v>. */
317
318 private final void step5()
319 { if (k == 0) return; /* for Bug 1 */ switch (b[k-1])
320 { case 'a': if (ends("al")) break; return;
321 case 'c': if (ends("ance")) break;
322 if (ends("ence")) break; return;
323 case 'e': if (ends("er")) break; return;
324 case 'i': if (ends("ic")) break; return;
325 case 'l': if (ends("able")) break;
326 if (ends("ible")) break; return;
327 case 'n': if (ends("ant")) break;
328 if (ends("ement")) break;
329 if (ends("ment")) break;
330 /* element etc. not stripped before the m */
331 if (ends("ent")) break; return;
332 case 'o': if (ends("ion") && j >= 0 && (b[j] == 's' || b[j] == 't')) break;
333 /* j >= 0 fixes Bug 2 */
334 if (ends("ou")) break; return;
335 /* takes care of -ous */
336 case 's': if (ends("ism")) break; return;
337 case 't': if (ends("ate")) break;
338 if (ends("iti")) break; return;
339 case 'u': if (ends("ous")) break; return;
340 case 'v': if (ends("ive")) break; return;
341 case 'z': if (ends("ize")) break; return;
342 default: return;
343 }
344 if (m() > 1) k = j;
345 }
346
347 /* step6() removes a final -e if m() > 1. */
348
349 private final void step6()
350 { j = k;
351 if (b[k] == 'e')
352 { int a = m();
353 if (a > 1 || a == 1 && !cvc(k-1)) k--;
354 }
355 if (b[k] == 'l' && doublec(k) && m() > 1) k--;
356 }
357
358 /** Stem the word placed into the Stemmer buffer through calls to add().
359 * Returns true if the stemming process resulted in a word different
360 * from the input. You can retrieve the result with
361 * getResultLength()/getResultBuffer() or toString().
362 */
363 public void stem()
364 { k = i - 1;
365 if (k > 1) { step1(); step2(); step3(); step4(); step5(); step6(); }
366 i_end = k+1; i = 0;
367 }
368
369 /** Test program for demonstrating the Stemmer. It reads text from a
370 * a list of files, stems each word, and writes the result to standard
371 * output. Note that the word stemmed is expected to be in lower case:
372 * forcing lower case must be done outside the Stemmer class.
373 * Usage: Stemmer file-name file-name ...
374 */
375 public static void main(String[] args)
376 {
377 char[] w = new char[501];
378 Stemmer s = new Stemmer();
379 for (int i = 0; i < args.length; i++)
380 try
381 {
382 FileInputStream in = new FileInputStream(args[i]);
383
384 try
385 { while(true)
386
387 { int ch = in.read();
388 if (Character.isLetter((char) ch))
389 {
390 int j = 0;
391 while(true)
392 { ch = Character.toLowerCase((char) ch);
393 w[j] = (char) ch;
394 if (j < 500) j++;
395 ch = in.read();
396 if (!Character.isLetter((char) ch))
397 {
398 /* to test add(char ch) */
399 for (int c = 0; c < j; c++) s.add(w[c]);
400
401 /* or, to test add(char[] w, int j) */
402 /* s.add(w, j); */
403
404 s.stem();
405 { String u;
406
407 /* and now, to test toString() : */
408 u = s.toString();
409
410 /* to test getResultBuffer(), getResultLength() : */
411 /* u = new String(s.getResultBuffer(), 0, s.getResultLength()); */
412
413 System.out.print(u);
414 }
415 break;
416 }
417 }
418 }
419 if (ch < 0) break;
420 System.out.print((char)ch);
421 }
422 }
423 catch (IOException e)
424 { System.out.println("error reading " + args[i]);
425 break;
426 }
427 }
428 catch (FileNotFoundException e)
429 { System.out.println("file " + args[i] + " not found");
430 break;
431 }
432 }
433 }