Mercurial > hg > mpdl-group
comparison software/mpdl-services/mpiwg-mpdl-lt/src/de/mpg/mpiwg/berlin/mpdl/lt/text/norm/lang/MpdlNormalizerLexAR.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 Arabic text | |
5 * [this is a JFlex specification] | |
6 * | |
7 * Wolfgang Schmidle | |
8 * version 2011-02-28 | |
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>MpdlNormalizerLexAR.lex</tt> | |
20 */ | |
21 public class MpdlNormalizerLexAR { | |
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\4"+ | |
50 "\40\0\1\1\2\0\1\1\20\0\1\1\5\0\1\1\1\0\1\1"+ | |
51 "\uff82\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"; | |
65 | |
66 private static int [] zzUnpackAction() { | |
67 int [] result = new int[10]; | |
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\5\0\12\0\17\0\24\0\31\0\24\0\24"+ | |
93 "\0\24\0\24"; | |
94 | |
95 private static int [] zzUnpackRowMap() { | |
96 int [] result = new int[10]; | |
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\7\1\5\1\6\1\5"+ | |
120 "\1\10\1\7\1\5\1\6\1\5\1\11\1\7\1\5"+ | |
121 "\1\6\1\5\1\12\1\7\7\0\1\5\2\0"; | |
122 | |
123 private static int [] zzUnpackTrans() { | |
124 int [] result = new int[30]; | |
125 int offset = 0; | |
126 offset = zzUnpackTrans(ZZ_TRANS_PACKED_0, offset, result); | |
127 return result; | |
128 } | |
129 | |
130 private static int zzUnpackTrans(String packed, int offset, int [] result) { | |
131 int i = 0; /* index in packed string */ | |
132 int j = offset; /* index in unpacked array */ | |
133 int l = packed.length(); | |
134 while (i < l) { | |
135 int count = packed.charAt(i++); | |
136 int value = packed.charAt(i++); | |
137 value--; | |
138 do result[j++] = value; while (--count > 0); | |
139 } | |
140 return j; | |
141 } | |
142 | |
143 | |
144 /* error codes */ | |
145 private static final int ZZ_UNKNOWN_ERROR = 0; | |
146 private static final int ZZ_NO_MATCH = 1; | |
147 private static final int ZZ_PUSHBACK_2BIG = 2; | |
148 | |
149 /* error messages for the codes above */ | |
150 private static final String ZZ_ERROR_MSG[] = { | |
151 "Unkown internal scanner error", | |
152 "Error: could not match input", | |
153 "Error: pushback value was too large" | |
154 }; | |
155 | |
156 /** | |
157 * ZZ_ATTRIBUTE[aState] contains the attributes of state <code>aState</code> | |
158 */ | |
159 private static final int [] ZZ_ATTRIBUTE = zzUnpackAttribute(); | |
160 | |
161 private static final String ZZ_ATTRIBUTE_PACKED_0 = | |
162 "\4\0\1\11\1\1\4\11"; | |
163 | |
164 private static int [] zzUnpackAttribute() { | |
165 int [] result = new int[10]; | |
166 int offset = 0; | |
167 offset = zzUnpackAttribute(ZZ_ATTRIBUTE_PACKED_0, offset, result); | |
168 return result; | |
169 } | |
170 | |
171 private static int zzUnpackAttribute(String packed, int offset, int [] result) { | |
172 int i = 0; /* index in packed string */ | |
173 int j = offset; /* index in unpacked array */ | |
174 int l = packed.length(); | |
175 while (i < l) { | |
176 int count = packed.charAt(i++); | |
177 int value = packed.charAt(i++); | |
178 do result[j++] = value; while (--count > 0); | |
179 } | |
180 return j; | |
181 } | |
182 | |
183 /** the input device */ | |
184 private java.io.Reader zzReader; | |
185 | |
186 /** the current state of the DFA */ | |
187 private int zzState; | |
188 | |
189 /** the current lexical state */ | |
190 private int zzLexicalState = YYINITIAL; | |
191 | |
192 /** this buffer contains the current text to be matched and is | |
193 the source of the yytext() string */ | |
194 private char zzBuffer[] = new char[ZZ_BUFFERSIZE]; | |
195 | |
196 /** the textposition at the last accepting state */ | |
197 private int zzMarkedPos; | |
198 | |
199 /** the current text position in the buffer */ | |
200 private int zzCurrentPos; | |
201 | |
202 /** startRead marks the beginning of the yytext() string in the buffer */ | |
203 private int zzStartRead; | |
204 | |
205 /** endRead marks the last character in the buffer, that has been read | |
206 from input */ | |
207 private int zzEndRead; | |
208 | |
209 /** number of newlines encountered up to the start of the matched text */ | |
210 private int yyline; | |
211 | |
212 /** the number of characters up to the start of the matched text */ | |
213 private int yychar; | |
214 | |
215 /** | |
216 * the number of characters from the last newline up to the start of the | |
217 * matched text | |
218 */ | |
219 private int yycolumn; | |
220 | |
221 /** | |
222 * zzAtBOL == true <=> the scanner is currently at the beginning of a line | |
223 */ | |
224 private boolean zzAtBOL = true; | |
225 | |
226 /** zzAtEOF == true <=> the scanner is at the EOF */ | |
227 private boolean zzAtEOF; | |
228 | |
229 /** denotes if the user-EOF-code has already been executed */ | |
230 private boolean zzEOFDone; | |
231 | |
232 /* user code: */ | |
233 private String original = ""; | |
234 private String normalized = ""; | |
235 private int problem = 0; | |
236 | |
237 private void add (String norm) { | |
238 original += yytext(); | |
239 normalized += norm; | |
240 } | |
241 | |
242 private static final String LB = "[\u002d\u00ad] "; | |
243 | |
244 | |
245 /** | |
246 * Creates a new scanner | |
247 * There is also a java.io.InputStream version of this constructor. | |
248 * | |
249 * @param in the java.io.Reader to read input from. | |
250 */ | |
251 public MpdlNormalizerLexAR(java.io.Reader in) { | |
252 this.zzReader = in; | |
253 } | |
254 | |
255 /** | |
256 * Creates a new scanner. | |
257 * There is also java.io.Reader version of this constructor. | |
258 * | |
259 * @param in the java.io.Inputstream to read input from. | |
260 */ | |
261 public MpdlNormalizerLexAR(java.io.InputStream in) { | |
262 this(new java.io.InputStreamReader(in)); | |
263 } | |
264 | |
265 /** | |
266 * Unpacks the compressed character translation table. | |
267 * | |
268 * @param packed the packed character translation table | |
269 * @return the unpacked character translation table | |
270 */ | |
271 private static char [] zzUnpackCMap(String packed) { | |
272 char [] map = new char[0x10000]; | |
273 int i = 0; /* index in packed string */ | |
274 int j = 0; /* index in unpacked array */ | |
275 while (i < 42) { | |
276 int count = packed.charAt(i++); | |
277 char value = packed.charAt(i++); | |
278 do map[j++] = value; while (--count > 0); | |
279 } | |
280 return map; | |
281 } | |
282 | |
283 | |
284 /** | |
285 * Refills the input buffer. | |
286 * | |
287 * @return <code>false</code>, iff there was new input. | |
288 * | |
289 * @exception java.io.IOException if any I/O-Error occurs | |
290 */ | |
291 private boolean zzRefill() throws java.io.IOException { | |
292 | |
293 /* first: make room (if you can) */ | |
294 if (zzStartRead > 0) { | |
295 System.arraycopy(zzBuffer, zzStartRead, | |
296 zzBuffer, 0, | |
297 zzEndRead-zzStartRead); | |
298 | |
299 /* translate stored positions */ | |
300 zzEndRead-= zzStartRead; | |
301 zzCurrentPos-= zzStartRead; | |
302 zzMarkedPos-= zzStartRead; | |
303 zzStartRead = 0; | |
304 } | |
305 | |
306 /* is the buffer big enough? */ | |
307 if (zzCurrentPos >= zzBuffer.length) { | |
308 /* if not: blow it up */ | |
309 char newBuffer[] = new char[zzCurrentPos*2]; | |
310 System.arraycopy(zzBuffer, 0, newBuffer, 0, zzBuffer.length); | |
311 zzBuffer = newBuffer; | |
312 } | |
313 | |
314 /* finally: fill the buffer with new input */ | |
315 int numRead = zzReader.read(zzBuffer, zzEndRead, | |
316 zzBuffer.length-zzEndRead); | |
317 | |
318 if (numRead > 0) { | |
319 zzEndRead+= numRead; | |
320 return false; | |
321 } | |
322 // unlikely but not impossible: read 0 characters, but not at end of stream | |
323 if (numRead == 0) { | |
324 int c = zzReader.read(); | |
325 if (c == -1) { | |
326 return true; | |
327 } else { | |
328 zzBuffer[zzEndRead++] = (char) c; | |
329 return false; | |
330 } | |
331 } | |
332 | |
333 // numRead < 0 | |
334 return true; | |
335 } | |
336 | |
337 | |
338 /** | |
339 * Closes the input stream. | |
340 */ | |
341 public final void yyclose() throws java.io.IOException { | |
342 zzAtEOF = true; /* indicate end of file */ | |
343 zzEndRead = zzStartRead; /* invalidate buffer */ | |
344 | |
345 if (zzReader != null) | |
346 zzReader.close(); | |
347 } | |
348 | |
349 | |
350 /** | |
351 * Resets the scanner to read from a new input stream. | |
352 * Does not close the old reader. | |
353 * | |
354 * All internal variables are reset, the old input stream | |
355 * <b>cannot</b> be reused (internal buffer is discarded and lost). | |
356 * Lexical state is set to <tt>ZZ_INITIAL</tt>. | |
357 * | |
358 * @param reader the new input stream | |
359 */ | |
360 public final void yyreset(java.io.Reader reader) { | |
361 zzReader = reader; | |
362 zzAtBOL = true; | |
363 zzAtEOF = false; | |
364 zzEOFDone = false; | |
365 zzEndRead = zzStartRead = 0; | |
366 zzCurrentPos = zzMarkedPos = 0; | |
367 yyline = yychar = yycolumn = 0; | |
368 zzLexicalState = YYINITIAL; | |
369 } | |
370 | |
371 | |
372 /** | |
373 * Returns the current lexical state. | |
374 */ | |
375 public final int yystate() { | |
376 return zzLexicalState; | |
377 } | |
378 | |
379 | |
380 /** | |
381 * Enters a new lexical state | |
382 * | |
383 * @param newState the new lexical state | |
384 */ | |
385 public final void yybegin(int newState) { | |
386 zzLexicalState = newState; | |
387 } | |
388 | |
389 | |
390 /** | |
391 * Returns the text matched by the current regular expression. | |
392 */ | |
393 public final String yytext() { | |
394 return new String( zzBuffer, zzStartRead, zzMarkedPos-zzStartRead ); | |
395 } | |
396 | |
397 | |
398 /** | |
399 * Returns the character at position <tt>pos</tt> from the | |
400 * matched text. | |
401 * | |
402 * It is equivalent to yytext().charAt(pos), but faster | |
403 * | |
404 * @param pos the position of the character to fetch. | |
405 * A value from 0 to yylength()-1. | |
406 * | |
407 * @return the character at position pos | |
408 */ | |
409 public final char yycharat(int pos) { | |
410 return zzBuffer[zzStartRead+pos]; | |
411 } | |
412 | |
413 | |
414 /** | |
415 * Returns the length of the matched text region. | |
416 */ | |
417 public final int yylength() { | |
418 return zzMarkedPos-zzStartRead; | |
419 } | |
420 | |
421 | |
422 /** | |
423 * Reports an error that occured while scanning. | |
424 * | |
425 * In a wellformed scanner (no or only correct usage of | |
426 * yypushback(int) and a match-all fallback rule) this method | |
427 * will only be called with things that "Can't Possibly Happen". | |
428 * If this method is called, something is seriously wrong | |
429 * (e.g. a JFlex bug producing a faulty scanner etc.). | |
430 * | |
431 * Usual syntax/scanner level error handling should be done | |
432 * in error fallback rules. | |
433 * | |
434 * @param errorCode the code of the errormessage to display | |
435 */ | |
436 private void zzScanError(int errorCode) { | |
437 String message; | |
438 try { | |
439 message = ZZ_ERROR_MSG[errorCode]; | |
440 } | |
441 catch (ArrayIndexOutOfBoundsException e) { | |
442 message = ZZ_ERROR_MSG[ZZ_UNKNOWN_ERROR]; | |
443 } | |
444 | |
445 throw new Error(message); | |
446 } | |
447 | |
448 | |
449 /** | |
450 * Pushes the specified amount of characters back into the input stream. | |
451 * | |
452 * They will be read again by then next call of the scanning method | |
453 * | |
454 * @param number the number of characters to be read again. | |
455 * This number must not be greater than yylength()! | |
456 */ | |
457 public void yypushback(int number) { | |
458 if ( number > yylength() ) | |
459 zzScanError(ZZ_PUSHBACK_2BIG); | |
460 | |
461 zzMarkedPos -= number; | |
462 } | |
463 | |
464 | |
465 /** | |
466 * Resumes scanning until the next regular expression is matched, | |
467 * the end of input is encountered or an I/O-Error occurs. | |
468 * | |
469 * @return the next token | |
470 * @exception java.io.IOException if any I/O-Error occurs | |
471 */ | |
472 public java.lang.String yylex() throws java.io.IOException { | |
473 int zzInput; | |
474 int zzAction; | |
475 | |
476 // cached fields: | |
477 int zzCurrentPosL; | |
478 int zzMarkedPosL; | |
479 int zzEndReadL = zzEndRead; | |
480 char [] zzBufferL = zzBuffer; | |
481 char [] zzCMapL = ZZ_CMAP; | |
482 | |
483 int [] zzTransL = ZZ_TRANS; | |
484 int [] zzRowMapL = ZZ_ROWMAP; | |
485 int [] zzAttrL = ZZ_ATTRIBUTE; | |
486 | |
487 while (true) { | |
488 zzMarkedPosL = zzMarkedPos; | |
489 | |
490 zzAction = -1; | |
491 | |
492 zzCurrentPosL = zzCurrentPos = zzStartRead = zzMarkedPosL; | |
493 | |
494 zzState = ZZ_LEXSTATE[zzLexicalState]; | |
495 | |
496 | |
497 zzForAction: { | |
498 while (true) { | |
499 | |
500 if (zzCurrentPosL < zzEndReadL) | |
501 zzInput = zzBufferL[zzCurrentPosL++]; | |
502 else if (zzAtEOF) { | |
503 zzInput = YYEOF; | |
504 break zzForAction; | |
505 } | |
506 else { | |
507 // store back cached positions | |
508 zzCurrentPos = zzCurrentPosL; | |
509 zzMarkedPos = zzMarkedPosL; | |
510 boolean eof = zzRefill(); | |
511 // get translated positions and possibly new buffer | |
512 zzCurrentPosL = zzCurrentPos; | |
513 zzMarkedPosL = zzMarkedPos; | |
514 zzBufferL = zzBuffer; | |
515 zzEndReadL = zzEndRead; | |
516 if (eof) { | |
517 zzInput = YYEOF; | |
518 break zzForAction; | |
519 } | |
520 else { | |
521 zzInput = zzBufferL[zzCurrentPosL++]; | |
522 } | |
523 } | |
524 int zzNext = zzTransL[ zzRowMapL[zzState] + zzCMapL[zzInput] ]; | |
525 if (zzNext == -1) break zzForAction; | |
526 zzState = zzNext; | |
527 | |
528 int zzAttributes = zzAttrL[zzState]; | |
529 if ( (zzAttributes & 1) == 1 ) { | |
530 zzAction = zzState; | |
531 zzMarkedPosL = zzCurrentPosL; | |
532 if ( (zzAttributes & 8) == 8 ) break zzForAction; | |
533 } | |
534 | |
535 } | |
536 } | |
537 | |
538 // store back cached position | |
539 zzMarkedPos = zzMarkedPosL; | |
540 | |
541 switch (zzAction < 0 ? zzAction : ZZ_ACTION[zzAction]) { | |
542 case 5: | |
543 { switch (problem) { | |
544 case 1: return original; | |
545 default: return normalized.replaceAll(LB, ""); | |
546 } | |
547 } | |
548 case 6: break; | |
549 case 4: | |
550 { switch (problem) { | |
551 case 1: return ""; | |
552 default: return normalized.replaceAll(LB, ""); | |
553 } | |
554 } | |
555 case 7: break; | |
556 case 2: | |
557 { problem = 1; add(yytext()); | |
558 } | |
559 case 8: break; | |
560 case 3: | |
561 { switch (problem) { | |
562 case 1: return original; | |
563 default: return normalized; | |
564 } | |
565 } | |
566 case 9: break; | |
567 case 1: | |
568 { add(yytext()); | |
569 } | |
570 case 10: break; | |
571 default: | |
572 if (zzInput == YYEOF && zzStartRead == zzCurrentPos) { | |
573 zzAtEOF = true; | |
574 return null; | |
575 } | |
576 else { | |
577 zzScanError(ZZ_NO_MATCH); | |
578 } | |
579 } | |
580 } | |
581 } | |
582 | |
583 | |
584 } |