Mercurial > hg > mpdl-group
comparison software/mpdl-services/mpiwg-mpdl-lt/src/de/mpg/mpiwg/berlin/mpdl/lt/text/norm/lang/MpdlNormalizerLexNL.java @ 19:4a3641ae14d2
Erstellung
author | Josef Willenborg <jwillenborg@mpiwg-berlin.mpg.de> |
---|---|
date | Wed, 09 Nov 2011 15:32:05 +0100 |
parents | |
children | e845310098ba |
comparison
equal
deleted
inserted
replaced
18:dc5e9fcb3fdc | 19:4a3641ae14d2 |
---|---|
1 /* The following code was generated by JFlex 1.4.3 on 21.07.11 11:22 */ | |
2 | |
3 /* | |
4 * Normalization rules for Dutch text | |
5 * [this is a JFlex specification] | |
6 * | |
7 * Wolfgang Schmidle | |
8 * version 2011-07-12 | |
9 * | |
10 */ | |
11 | |
12 package de.mpg.mpiwg.berlin.mpdl.lt.text.norm.lang; | |
13 | |
14 | |
15 /** | |
16 * This class is a scanner generated by | |
17 * <a href="http://www.jflex.de/">JFlex</a> 1.4.3 | |
18 * on 21.07.11 11:22 from the specification file | |
19 * <tt>MpdlNormalizerLexNL.lex</tt> | |
20 */ | |
21 public class MpdlNormalizerLexNL { | |
22 | |
23 /** This character denotes the end of file */ | |
24 public static final int YYEOF = -1; | |
25 | |
26 /** initial size of the lookahead buffer */ | |
27 private static final int ZZ_BUFFERSIZE = 16384; | |
28 | |
29 /** lexical states */ | |
30 public static final int SEARCH = 6; | |
31 public static final int DICT = 4; | |
32 public static final int YYINITIAL = 0; | |
33 public static final int DISP = 2; | |
34 | |
35 /** | |
36 * ZZ_LEXSTATE[l] is the state in the DFA for the lexical state l | |
37 * ZZ_LEXSTATE[l+1] is the state in the DFA for the lexical state l | |
38 * at the beginning of a line | |
39 * l is of the form l = 2*k, k a non negative integer | |
40 */ | |
41 private static final int ZZ_LEXSTATE[] = { | |
42 0, 0, 1, 1, 2, 2, 3, 3 | |
43 }; | |
44 | |
45 /** | |
46 * Translates characters to character classes | |
47 */ | |
48 private static final String ZZ_CMAP_PACKED = | |
49 "\12\0\1\3\25\0\1\2\14\0\1\1\2\0\1\1\17\0\1\5"+ | |
50 "\40\0\1\1\2\0\1\1\20\0\1\1\5\0\1\1\1\0\1\1"+ | |
51 "\u0101\0\1\4\ufe80\0"; | |
52 | |
53 /** | |
54 * Translates characters to character classes | |
55 */ | |
56 private static final char [] ZZ_CMAP = zzUnpackCMap(ZZ_CMAP_PACKED); | |
57 | |
58 /** | |
59 * Translates DFA states to action switch labels. | |
60 */ | |
61 private static final int [] ZZ_ACTION = zzUnpackAction(); | |
62 | |
63 private static final String ZZ_ACTION_PACKED_0 = | |
64 "\4\0\2\1\1\2\1\3\1\4\1\5\1\6"; | |
65 | |
66 private static int [] zzUnpackAction() { | |
67 int [] result = new int[11]; | |
68 int offset = 0; | |
69 offset = zzUnpackAction(ZZ_ACTION_PACKED_0, offset, result); | |
70 return result; | |
71 } | |
72 | |
73 private static int zzUnpackAction(String packed, int offset, int [] result) { | |
74 int i = 0; /* index in packed string */ | |
75 int j = offset; /* index in unpacked array */ | |
76 int l = packed.length(); | |
77 while (i < l) { | |
78 int count = packed.charAt(i++); | |
79 int value = packed.charAt(i++); | |
80 do result[j++] = value; while (--count > 0); | |
81 } | |
82 return j; | |
83 } | |
84 | |
85 | |
86 /** | |
87 * Translates a state to a row index in the transition table | |
88 */ | |
89 private static final int [] ZZ_ROWMAP = zzUnpackRowMap(); | |
90 | |
91 private static final String ZZ_ROWMAP_PACKED_0 = | |
92 "\0\0\0\6\0\14\0\22\0\30\0\36\0\30\0\30"+ | |
93 "\0\30\0\30\0\30"; | |
94 | |
95 private static int [] zzUnpackRowMap() { | |
96 int [] result = new int[11]; | |
97 int offset = 0; | |
98 offset = zzUnpackRowMap(ZZ_ROWMAP_PACKED_0, offset, result); | |
99 return result; | |
100 } | |
101 | |
102 private static int zzUnpackRowMap(String packed, int offset, int [] result) { | |
103 int i = 0; /* index in packed string */ | |
104 int j = offset; /* index in unpacked array */ | |
105 int l = packed.length(); | |
106 while (i < l) { | |
107 int high = packed.charAt(i++) << 16; | |
108 result[j++] = high | packed.charAt(i++); | |
109 } | |
110 return j; | |
111 } | |
112 | |
113 /** | |
114 * The transition table of the DFA | |
115 */ | |
116 private static final int [] ZZ_TRANS = zzUnpackTrans(); | |
117 | |
118 private static final String ZZ_TRANS_PACKED_0 = | |
119 "\1\5\1\6\1\5\1\0\1\5\1\7\1\5\1\6"+ | |
120 "\1\5\1\10\1\11\1\7\1\5\1\6\1\5\1\12"+ | |
121 "\1\11\1\7\1\5\1\6\1\5\1\13\1\11\1\7"+ | |
122 "\10\0\1\5\3\0"; | |
123 | |
124 private static int [] zzUnpackTrans() { | |
125 int [] result = new int[36]; | |
126 int offset = 0; | |
127 offset = zzUnpackTrans(ZZ_TRANS_PACKED_0, offset, result); | |
128 return result; | |
129 } | |
130 | |
131 private static int zzUnpackTrans(String packed, int offset, int [] result) { | |
132 int i = 0; /* index in packed string */ | |
133 int j = offset; /* index in unpacked array */ | |
134 int l = packed.length(); | |
135 while (i < l) { | |
136 int count = packed.charAt(i++); | |
137 int value = packed.charAt(i++); | |
138 value--; | |
139 do result[j++] = value; while (--count > 0); | |
140 } | |
141 return j; | |
142 } | |
143 | |
144 | |
145 /* error codes */ | |
146 private static final int ZZ_UNKNOWN_ERROR = 0; | |
147 private static final int ZZ_NO_MATCH = 1; | |
148 private static final int ZZ_PUSHBACK_2BIG = 2; | |
149 | |
150 /* error messages for the codes above */ | |
151 private static final String ZZ_ERROR_MSG[] = { | |
152 "Unkown internal scanner error", | |
153 "Error: could not match input", | |
154 "Error: pushback value was too large" | |
155 }; | |
156 | |
157 /** | |
158 * ZZ_ATTRIBUTE[aState] contains the attributes of state <code>aState</code> | |
159 */ | |
160 private static final int [] ZZ_ATTRIBUTE = zzUnpackAttribute(); | |
161 | |
162 private static final String ZZ_ATTRIBUTE_PACKED_0 = | |
163 "\4\0\1\11\1\1\5\11"; | |
164 | |
165 private static int [] zzUnpackAttribute() { | |
166 int [] result = new int[11]; | |
167 int offset = 0; | |
168 offset = zzUnpackAttribute(ZZ_ATTRIBUTE_PACKED_0, offset, result); | |
169 return result; | |
170 } | |
171 | |
172 private static int zzUnpackAttribute(String packed, int offset, int [] result) { | |
173 int i = 0; /* index in packed string */ | |
174 int j = offset; /* index in unpacked array */ | |
175 int l = packed.length(); | |
176 while (i < l) { | |
177 int count = packed.charAt(i++); | |
178 int value = packed.charAt(i++); | |
179 do result[j++] = value; while (--count > 0); | |
180 } | |
181 return j; | |
182 } | |
183 | |
184 /** the input device */ | |
185 private java.io.Reader zzReader; | |
186 | |
187 /** the current state of the DFA */ | |
188 private int zzState; | |
189 | |
190 /** the current lexical state */ | |
191 private int zzLexicalState = YYINITIAL; | |
192 | |
193 /** this buffer contains the current text to be matched and is | |
194 the source of the yytext() string */ | |
195 private char zzBuffer[] = new char[ZZ_BUFFERSIZE]; | |
196 | |
197 /** the textposition at the last accepting state */ | |
198 private int zzMarkedPos; | |
199 | |
200 /** the current text position in the buffer */ | |
201 private int zzCurrentPos; | |
202 | |
203 /** startRead marks the beginning of the yytext() string in the buffer */ | |
204 private int zzStartRead; | |
205 | |
206 /** endRead marks the last character in the buffer, that has been read | |
207 from input */ | |
208 private int zzEndRead; | |
209 | |
210 /** number of newlines encountered up to the start of the matched text */ | |
211 private int yyline; | |
212 | |
213 /** the number of characters up to the start of the matched text */ | |
214 private int yychar; | |
215 | |
216 /** | |
217 * the number of characters from the last newline up to the start of the | |
218 * matched text | |
219 */ | |
220 private int yycolumn; | |
221 | |
222 /** | |
223 * zzAtBOL == true <=> the scanner is currently at the beginning of a line | |
224 */ | |
225 private boolean zzAtBOL = true; | |
226 | |
227 /** zzAtEOF == true <=> the scanner is at the EOF */ | |
228 private boolean zzAtEOF; | |
229 | |
230 /** denotes if the user-EOF-code has already been executed */ | |
231 private boolean zzEOFDone; | |
232 | |
233 /* user code: */ | |
234 private String original = ""; | |
235 private String normalized = ""; | |
236 private int problem = 0; | |
237 | |
238 private void add (String norm) { | |
239 original += yytext(); | |
240 normalized += norm; | |
241 } | |
242 | |
243 private static final String LB = "[\u002d\u00ad] "; | |
244 | |
245 | |
246 /** | |
247 * Creates a new scanner | |
248 * There is also a java.io.InputStream version of this constructor. | |
249 * | |
250 * @param in the java.io.Reader to read input from. | |
251 */ | |
252 public MpdlNormalizerLexNL(java.io.Reader in) { | |
253 this.zzReader = in; | |
254 } | |
255 | |
256 /** | |
257 * Creates a new scanner. | |
258 * There is also java.io.Reader version of this constructor. | |
259 * | |
260 * @param in the java.io.Inputstream to read input from. | |
261 */ | |
262 public MpdlNormalizerLexNL(java.io.InputStream in) { | |
263 this(new java.io.InputStreamReader(in)); | |
264 } | |
265 | |
266 /** | |
267 * Unpacks the compressed character translation table. | |
268 * | |
269 * @param packed the packed character translation table | |
270 * @return the unpacked character translation table | |
271 */ | |
272 private static char [] zzUnpackCMap(String packed) { | |
273 char [] map = new char[0x10000]; | |
274 int i = 0; /* index in packed string */ | |
275 int j = 0; /* index in unpacked array */ | |
276 while (i < 46) { | |
277 int count = packed.charAt(i++); | |
278 char value = packed.charAt(i++); | |
279 do map[j++] = value; while (--count > 0); | |
280 } | |
281 return map; | |
282 } | |
283 | |
284 | |
285 /** | |
286 * Refills the input buffer. | |
287 * | |
288 * @return <code>false</code>, iff there was new input. | |
289 * | |
290 * @exception java.io.IOException if any I/O-Error occurs | |
291 */ | |
292 private boolean zzRefill() throws java.io.IOException { | |
293 | |
294 /* first: make room (if you can) */ | |
295 if (zzStartRead > 0) { | |
296 System.arraycopy(zzBuffer, zzStartRead, | |
297 zzBuffer, 0, | |
298 zzEndRead-zzStartRead); | |
299 | |
300 /* translate stored positions */ | |
301 zzEndRead-= zzStartRead; | |
302 zzCurrentPos-= zzStartRead; | |
303 zzMarkedPos-= zzStartRead; | |
304 zzStartRead = 0; | |
305 } | |
306 | |
307 /* is the buffer big enough? */ | |
308 if (zzCurrentPos >= zzBuffer.length) { | |
309 /* if not: blow it up */ | |
310 char newBuffer[] = new char[zzCurrentPos*2]; | |
311 System.arraycopy(zzBuffer, 0, newBuffer, 0, zzBuffer.length); | |
312 zzBuffer = newBuffer; | |
313 } | |
314 | |
315 /* finally: fill the buffer with new input */ | |
316 int numRead = zzReader.read(zzBuffer, zzEndRead, | |
317 zzBuffer.length-zzEndRead); | |
318 | |
319 if (numRead > 0) { | |
320 zzEndRead+= numRead; | |
321 return false; | |
322 } | |
323 // unlikely but not impossible: read 0 characters, but not at end of stream | |
324 if (numRead == 0) { | |
325 int c = zzReader.read(); | |
326 if (c == -1) { | |
327 return true; | |
328 } else { | |
329 zzBuffer[zzEndRead++] = (char) c; | |
330 return false; | |
331 } | |
332 } | |
333 | |
334 // numRead < 0 | |
335 return true; | |
336 } | |
337 | |
338 | |
339 /** | |
340 * Closes the input stream. | |
341 */ | |
342 public final void yyclose() throws java.io.IOException { | |
343 zzAtEOF = true; /* indicate end of file */ | |
344 zzEndRead = zzStartRead; /* invalidate buffer */ | |
345 | |
346 if (zzReader != null) | |
347 zzReader.close(); | |
348 } | |
349 | |
350 | |
351 /** | |
352 * Resets the scanner to read from a new input stream. | |
353 * Does not close the old reader. | |
354 * | |
355 * All internal variables are reset, the old input stream | |
356 * <b>cannot</b> be reused (internal buffer is discarded and lost). | |
357 * Lexical state is set to <tt>ZZ_INITIAL</tt>. | |
358 * | |
359 * @param reader the new input stream | |
360 */ | |
361 public final void yyreset(java.io.Reader reader) { | |
362 zzReader = reader; | |
363 zzAtBOL = true; | |
364 zzAtEOF = false; | |
365 zzEOFDone = false; | |
366 zzEndRead = zzStartRead = 0; | |
367 zzCurrentPos = zzMarkedPos = 0; | |
368 yyline = yychar = yycolumn = 0; | |
369 zzLexicalState = YYINITIAL; | |
370 } | |
371 | |
372 | |
373 /** | |
374 * Returns the current lexical state. | |
375 */ | |
376 public final int yystate() { | |
377 return zzLexicalState; | |
378 } | |
379 | |
380 | |
381 /** | |
382 * Enters a new lexical state | |
383 * | |
384 * @param newState the new lexical state | |
385 */ | |
386 public final void yybegin(int newState) { | |
387 zzLexicalState = newState; | |
388 } | |
389 | |
390 | |
391 /** | |
392 * Returns the text matched by the current regular expression. | |
393 */ | |
394 public final String yytext() { | |
395 return new String( zzBuffer, zzStartRead, zzMarkedPos-zzStartRead ); | |
396 } | |
397 | |
398 | |
399 /** | |
400 * Returns the character at position <tt>pos</tt> from the | |
401 * matched text. | |
402 * | |
403 * It is equivalent to yytext().charAt(pos), but faster | |
404 * | |
405 * @param pos the position of the character to fetch. | |
406 * A value from 0 to yylength()-1. | |
407 * | |
408 * @return the character at position pos | |
409 */ | |
410 public final char yycharat(int pos) { | |
411 return zzBuffer[zzStartRead+pos]; | |
412 } | |
413 | |
414 | |
415 /** | |
416 * Returns the length of the matched text region. | |
417 */ | |
418 public final int yylength() { | |
419 return zzMarkedPos-zzStartRead; | |
420 } | |
421 | |
422 | |
423 /** | |
424 * Reports an error that occured while scanning. | |
425 * | |
426 * In a wellformed scanner (no or only correct usage of | |
427 * yypushback(int) and a match-all fallback rule) this method | |
428 * will only be called with things that "Can't Possibly Happen". | |
429 * If this method is called, something is seriously wrong | |
430 * (e.g. a JFlex bug producing a faulty scanner etc.). | |
431 * | |
432 * Usual syntax/scanner level error handling should be done | |
433 * in error fallback rules. | |
434 * | |
435 * @param errorCode the code of the errormessage to display | |
436 */ | |
437 private void zzScanError(int errorCode) { | |
438 String message; | |
439 try { | |
440 message = ZZ_ERROR_MSG[errorCode]; | |
441 } | |
442 catch (ArrayIndexOutOfBoundsException e) { | |
443 message = ZZ_ERROR_MSG[ZZ_UNKNOWN_ERROR]; | |
444 } | |
445 | |
446 throw new Error(message); | |
447 } | |
448 | |
449 | |
450 /** | |
451 * Pushes the specified amount of characters back into the input stream. | |
452 * | |
453 * They will be read again by then next call of the scanning method | |
454 * | |
455 * @param number the number of characters to be read again. | |
456 * This number must not be greater than yylength()! | |
457 */ | |
458 public void yypushback(int number) { | |
459 if ( number > yylength() ) | |
460 zzScanError(ZZ_PUSHBACK_2BIG); | |
461 | |
462 zzMarkedPos -= number; | |
463 } | |
464 | |
465 | |
466 /** | |
467 * Resumes scanning until the next regular expression is matched, | |
468 * the end of input is encountered or an I/O-Error occurs. | |
469 * | |
470 * @return the next token | |
471 * @exception java.io.IOException if any I/O-Error occurs | |
472 */ | |
473 public java.lang.String yylex() throws java.io.IOException { | |
474 int zzInput; | |
475 int zzAction; | |
476 | |
477 // cached fields: | |
478 int zzCurrentPosL; | |
479 int zzMarkedPosL; | |
480 int zzEndReadL = zzEndRead; | |
481 char [] zzBufferL = zzBuffer; | |
482 char [] zzCMapL = ZZ_CMAP; | |
483 | |
484 int [] zzTransL = ZZ_TRANS; | |
485 int [] zzRowMapL = ZZ_ROWMAP; | |
486 int [] zzAttrL = ZZ_ATTRIBUTE; | |
487 | |
488 while (true) { | |
489 zzMarkedPosL = zzMarkedPos; | |
490 | |
491 zzAction = -1; | |
492 | |
493 zzCurrentPosL = zzCurrentPos = zzStartRead = zzMarkedPosL; | |
494 | |
495 zzState = ZZ_LEXSTATE[zzLexicalState]; | |
496 | |
497 | |
498 zzForAction: { | |
499 while (true) { | |
500 | |
501 if (zzCurrentPosL < zzEndReadL) | |
502 zzInput = zzBufferL[zzCurrentPosL++]; | |
503 else if (zzAtEOF) { | |
504 zzInput = YYEOF; | |
505 break zzForAction; | |
506 } | |
507 else { | |
508 // store back cached positions | |
509 zzCurrentPos = zzCurrentPosL; | |
510 zzMarkedPos = zzMarkedPosL; | |
511 boolean eof = zzRefill(); | |
512 // get translated positions and possibly new buffer | |
513 zzCurrentPosL = zzCurrentPos; | |
514 zzMarkedPosL = zzMarkedPos; | |
515 zzBufferL = zzBuffer; | |
516 zzEndReadL = zzEndRead; | |
517 if (eof) { | |
518 zzInput = YYEOF; | |
519 break zzForAction; | |
520 } | |
521 else { | |
522 zzInput = zzBufferL[zzCurrentPosL++]; | |
523 } | |
524 } | |
525 int zzNext = zzTransL[ zzRowMapL[zzState] + zzCMapL[zzInput] ]; | |
526 if (zzNext == -1) break zzForAction; | |
527 zzState = zzNext; | |
528 | |
529 int zzAttributes = zzAttrL[zzState]; | |
530 if ( (zzAttributes & 1) == 1 ) { | |
531 zzAction = zzState; | |
532 zzMarkedPosL = zzCurrentPosL; | |
533 if ( (zzAttributes & 8) == 8 ) break zzForAction; | |
534 } | |
535 | |
536 } | |
537 } | |
538 | |
539 // store back cached position | |
540 zzMarkedPos = zzMarkedPosL; | |
541 | |
542 switch (zzAction < 0 ? zzAction : ZZ_ACTION[zzAction]) { | |
543 case 5: | |
544 { switch (problem) { | |
545 case 1: return ""; | |
546 default: return normalized.replaceAll(LB, ""); | |
547 } | |
548 } | |
549 case 7: break; | |
550 case 2: | |
551 { problem = 1; add(yytext()); | |
552 } | |
553 case 8: break; | |
554 case 4: | |
555 { add("s"); | |
556 } | |
557 case 9: break; | |
558 case 3: | |
559 { switch (problem) { | |
560 case 1: return original; | |
561 default: return normalized; | |
562 } | |
563 } | |
564 case 10: break; | |
565 case 6: | |
566 { switch (problem) { | |
567 case 1: return original; | |
568 default: return normalized.replaceAll(LB, "").toLowerCase(); | |
569 } | |
570 } | |
571 case 11: break; | |
572 case 1: | |
573 { add(yytext()); | |
574 } | |
575 case 12: break; | |
576 default: | |
577 if (zzInput == YYEOF && zzStartRead == zzCurrentPos) { | |
578 zzAtEOF = true; | |
579 return null; | |
580 } | |
581 else { | |
582 zzScanError(ZZ_NO_MATCH); | |
583 } | |
584 } | |
585 } | |
586 } | |
587 | |
588 | |
589 } |