143 |
short secondary; |
short secondary; |
144 |
short tertiary; |
short tertiary; |
145 |
|
|
146 |
CollationElement(String char_seq, int primary, short secondary, short tertiary) |
CollationElement (String char_seq, int primary, short secondary, short tertiary) |
147 |
{ |
{ |
148 |
this.char_seq = char_seq; |
this.char_seq = char_seq; |
149 |
this.primary = primary; |
this.primary = primary; |
175 |
* @exception ParseException If the rule string contains syntax errors. |
* @exception ParseException If the rule string contains syntax errors. |
176 |
*/ |
*/ |
177 |
public RuleBasedCollator (String rules) throws ParseException |
public RuleBasedCollator (String rules) throws ParseException |
178 |
{ |
{ |
179 |
|
if (rules.equals ("")) |
180 |
|
throw new ParseException ("empty rule set", 0); |
181 |
|
|
182 |
this.rules = rules; |
this.rules = rules; |
183 |
|
Vector vec = new Vector(); |
|
if (rules.equals("")) |
|
|
throw new IllegalArgumentException("Empty rule set"); |
|
|
|
|
|
Vector v = new Vector(); |
|
184 |
boolean ignore_chars = true; |
boolean ignore_chars = true; |
185 |
int primary_seq = 0; |
int primary_seq = 0; |
186 |
short secondary_seq = 0; |
short secondary_seq = 0; |
187 |
short tertiary_seq = 0; |
short tertiary_seq = 0; |
188 |
StringBuffer sb = new StringBuffer(""); |
StringBuffer argument = new StringBuffer(); |
189 |
int len = rules.length(); |
int len = rules.length(); |
190 |
|
|
191 |
for (int index = 0; index < len; ++index) |
for (int index = 0; index < len; ++index) |
194 |
|
|
195 |
// Just skip whitespace. |
// Just skip whitespace. |
196 |
if (Character.isWhitespace (c)) |
if (Character.isWhitespace (c)) |
197 |
continue; |
continue; |
198 |
|
|
199 |
// Primary difference |
// Primary difference. |
200 |
if (c == '<') |
if (c == '<') |
201 |
{ |
{ |
202 |
ignore_chars = false; |
ignore_chars = false; |
203 |
CollationElement e = new CollationElement(sb.toString(), primary_seq, |
CollationElement e = new CollationElement (argument.toString(), primary_seq, |
204 |
secondary_seq, |
secondary_seq, |
205 |
tertiary_seq); |
tertiary_seq); |
206 |
secondary_seq = 0; |
secondary_seq = 0; |
207 |
tertiary_seq = 0; |
tertiary_seq = 0; |
208 |
++primary_seq; |
++primary_seq; |
209 |
|
|
210 |
v.add(e); |
vec.add (e); |
211 |
sb.setLength(0); |
argument.setLength (0); |
212 |
continue; |
continue; |
213 |
} |
} |
214 |
|
|
215 |
// Secondary difference |
// Secondary difference. |
216 |
if (c == ';') |
if (c == ';') |
217 |
{ |
{ |
218 |
if (primary_seq == 0) |
if (primary_seq == 0) |
219 |
throw new ParseException (rules, index); |
throw new ParseException (rules, index); |
220 |
|
|
221 |
CollationElement e = new CollationElement(sb.toString(), primary_seq, |
CollationElement e = new CollationElement (argument.toString(), primary_seq, |
222 |
secondary_seq, |
secondary_seq, |
223 |
tertiary_seq); |
tertiary_seq); |
224 |
++secondary_seq; |
++secondary_seq; |
225 |
tertiary_seq = 0; |
tertiary_seq = 0; |
226 |
|
|
227 |
v.add(e); |
vec.add (e); |
228 |
sb.setLength(0); |
argument.setLength (0); |
229 |
continue; |
continue; |
230 |
} |
} |
231 |
|
|
232 |
// Tertiary difference |
// Tertiary difference. |
233 |
if (c == ',') |
if (c == ',') |
234 |
{ |
{ |
235 |
if (primary_seq == 0) |
if (primary_seq == 0) |
236 |
throw new ParseException (rules, index); |
throw new ParseException (rules, index); |
237 |
|
|
238 |
CollationElement e = new CollationElement(sb.toString(), primary_seq, |
CollationElement e = new CollationElement (argument.toString(), |
239 |
secondary_seq, |
primary_seq, |
240 |
tertiary_seq); |
secondary_seq, |
241 |
|
tertiary_seq); |
242 |
++tertiary_seq; |
++tertiary_seq; |
243 |
|
|
244 |
v.add(e); |
vec.add (e); |
245 |
sb.setLength(0); |
argument.setLength (0); |
246 |
continue; |
continue; |
247 |
} |
} |
248 |
|
|
249 |
// Is equal to |
// Is equal to. |
250 |
if (c == '=') |
if (c == '=') |
251 |
{ |
{ |
252 |
if (primary_seq == 0) |
if (primary_seq == 0) |
253 |
throw new ParseException (rules, index); |
throw new ParseException (rules, index); |
254 |
|
|
255 |
CollationElement e = new CollationElement(sb.toString(), primary_seq, |
CollationElement e = new CollationElement (argument.toString(), |
256 |
secondary_seq, |
primary_seq, |
257 |
tertiary_seq); |
secondary_seq, |
258 |
v.add(e); |
tertiary_seq); |
259 |
sb.setLength(0); |
vec.add (e); |
260 |
|
argument.setLength (0); |
261 |
continue; |
continue; |
262 |
} |
} |
263 |
|
|
264 |
// Sort accents backwards |
// Sort accents backwards. |
265 |
if (c == '@') |
if (c == '@') |
266 |
{ |
{ |
267 |
throw new ParseException("French style accents not implemented yet", 0); |
throw new ParseException ("French style accents not implemented yet", 0); |
268 |
} |
} |
269 |
|
|
270 |
// Reset command |
// Reset command. |
271 |
if (c == '&') |
if (c == '&') |
272 |
{ |
{ |
273 |
throw new ParseException("Reset not implemented yet", 0); |
throw new ParseException ("Reset not implemented yet", 0); |
274 |
} |
} |
275 |
|
|
276 |
// See if we are still reading characters to skip |
// See if we are still reading characters to skip. |
277 |
if (ignore_chars == true) |
if (ignore_chars == true) |
278 |
{ |
{ |
279 |
CollationElement e = new CollationElement(c + "", 0, (short)0, |
CollationElement e = new CollationElement (c + "", 0, (short) 0, |
280 |
(short)0); |
(short) 0); |
281 |
v.add(e); |
vec.add (e); |
282 |
continue; |
continue; |
283 |
} |
} |
284 |
|
|
285 |
sb.append(c); |
argument.append (c); |
286 |
} |
} |
287 |
|
|
288 |
if (sb.length() > 0) |
if (argument.length() > 0) |
289 |
{ |
{ |
290 |
CollationElement e = new CollationElement (sb.toString(), primary_seq, |
CollationElement e = new CollationElement (argument.toString(), primary_seq, |
291 |
secondary_seq, tertiary_seq); |
secondary_seq, tertiary_seq); |
292 |
v.add (e); |
vec.add (e); |
293 |
} |
} |
294 |
|
|
295 |
ce_table = v.toArray(); |
ce_table = vec.toArray(); |
296 |
} |
} |
297 |
|
|
298 |
/** |
/** |
302 |
*/ |
*/ |
303 |
public Object clone() |
public Object clone() |
304 |
{ |
{ |
305 |
return super.clone(); |
RuleBasedCollator c = (RuleBasedCollator) super.clone (); |
306 |
|
return c; |
307 |
} |
} |
308 |
|
|
309 |
/** |
/** |
320 |
*/ |
*/ |
321 |
public int compare (String source, String target) |
public int compare (String source, String target) |
322 |
{ |
{ |
323 |
CollationElementIterator cei1 = getCollationElementIterator (source); |
CollationElementIterator cs = getCollationElementIterator (source); |
324 |
CollationElementIterator cei2 = getCollationElementIterator (target); |
CollationElementIterator ct = getCollationElementIterator (target); |
325 |
|
|
326 |
for(;;) |
while (true) |
327 |
{ |
{ |
328 |
int ord1 = cei1.next(); |
int os = cs.next(); |
329 |
int ord2 = cei2.next(); |
int ot = ct.next(); |
330 |
|
|
331 |
// Check for end of string |
// Check for end of string |
332 |
if (ord1 == CollationElementIterator.NULLORDER) |
if (os == CollationElementIterator.NULLORDER |
333 |
if (ord2 == CollationElementIterator.NULLORDER) |
&& ot == CollationElementIterator.NULLORDER) |
334 |
return(0); |
{ |
335 |
else |
// Source and target string are equal. |
336 |
return(-1); |
return 0; |
337 |
else if (ord2 == CollationElementIterator.NULLORDER) |
} |
338 |
return(1); |
else if (os == CollationElementIterator.NULLORDER) |
339 |
|
{ |
340 |
|
// Source string is shorter, so return "less than". |
341 |
|
return -1; |
342 |
|
} |
343 |
|
else if (ot == CollationElementIterator.NULLORDER) |
344 |
|
{ |
345 |
|
// Target string is shorter, so return "greater than". |
346 |
|
return 1; |
347 |
|
} |
348 |
|
|
349 |
// We know chars are totally equal, so skip |
// We know chars are totally equal, so skip |
350 |
if (ord1 == ord2) |
if (os == ot) |
351 |
continue; |
continue; |
352 |
|
|
353 |
// Check for primary strength differences |
// Check for primary strength differences |
354 |
int prim1 = cei1.primaryOrder(ord1); |
int prim1 = cs.primaryOrder (os); |
355 |
int prim2 = cei2.primaryOrder(ord2); |
int prim2 = ct.primaryOrder (ot); |
356 |
|
|
357 |
if (prim1 < prim2) |
if (prim1 < prim2) |
358 |
return(-1); |
return -1; |
359 |
else if (prim1 > prim2) |
else if (prim1 > prim2) |
360 |
return(1); |
return 1; |
361 |
else if (getStrength() == PRIMARY) |
else if (getStrength() == PRIMARY) |
362 |
continue; |
continue; |
363 |
|
|
364 |
// Check for secondary strength differences |
// Check for secondary strength differences |
365 |
int sec1 = cei1.secondaryOrder(ord1); |
int sec1 = cs.secondaryOrder (os); |
366 |
int sec2 = cei2.secondaryOrder(ord2); |
int sec2 = ct.secondaryOrder (ot); |
367 |
|
|
368 |
if (sec1 < sec2) |
if (sec1 < sec2) |
369 |
return(-1); |
return -1; |
370 |
else if (sec1 > sec2) |
else if (sec1 > sec2) |
371 |
return(1); |
return 1; |
372 |
else if (getStrength() == SECONDARY) |
else if (getStrength() == SECONDARY) |
373 |
continue; |
continue; |
374 |
|
|
375 |
// Check for tertiary differences |
// Check for tertiary differences |
376 |
int tert1 = cei1.tertiaryOrder(ord1); |
int tert1 = cs.tertiaryOrder (os); |
377 |
int tert2 = cei2.tertiaryOrder(ord1); |
int tert2 = ct.tertiaryOrder (ot); |
378 |
|
|
379 |
if (tert1 < tert2) |
if (tert1 < tert2) |
380 |
return(-1); |
return -1; |
381 |
else if (tert1 > tert2) |
else if (tert1 > tert2) |
382 |
return(1); |
return 1; |
383 |
} |
} |
384 |
} |
} |
385 |
|
|
395 |
public boolean equals (Object obj) |
public boolean equals (Object obj) |
396 |
{ |
{ |
397 |
if (obj == this) |
if (obj == this) |
398 |
return(true); |
return true; |
399 |
else |
|
400 |
return(false); |
return false; |
401 |
} |
} |
402 |
|
|
403 |
/** |
/** |
412 |
public CollationElementIterator getCollationElementIterator (String source) |
public CollationElementIterator getCollationElementIterator (String source) |
413 |
{ |
{ |
414 |
return new CollationElementIterator (source, this); |
return new CollationElementIterator (source, this); |
415 |
} |
} |
416 |
|
|
417 |
/** |
/** |
418 |
* This method returns an instance of <code>CollationElementIterator</code> |
* This method returns an instance of <code>CollationElementIterator</code> |
425 |
*/ |
*/ |
426 |
public CollationElementIterator getCollationElementIterator (CharacterIterator source) |
public CollationElementIterator getCollationElementIterator (CharacterIterator source) |
427 |
{ |
{ |
428 |
StringBuffer sb = new StringBuffer(""); |
StringBuffer sb = new StringBuffer(); |
429 |
|
|
430 |
// Right now we assume that we will read from the beginning of the string. |
// Right now we assume that we will read from the beginning of the string. |
431 |
for (char c = source.first(); c != CharacterIterator.DONE; c = source.next()) |
for (char c = source.first(); c != CharacterIterator.DONE; c = source.next()) |
450 |
public CollationKey getCollationKey (String source) |
public CollationKey getCollationKey (String source) |
451 |
{ |
{ |
452 |
CollationElementIterator cei = getCollationElementIterator (source); |
CollationElementIterator cei = getCollationElementIterator (source); |
453 |
Vector vect = new Vector(25); |
Vector vec = new Vector (25); |
454 |
|
|
455 |
int ord = cei.next(); |
int ord = cei.next(); |
456 |
cei.reset(); //set to start of string |
cei.reset(); //set to start of string |
460 |
switch (getStrength()) |
switch (getStrength()) |
461 |
{ |
{ |
462 |
case PRIMARY: |
case PRIMARY: |
463 |
ord = cei.primaryOrder(ord); |
ord = cei.primaryOrder (ord); |
464 |
break; |
break; |
465 |
|
|
466 |
case SECONDARY: |
case SECONDARY: |
467 |
ord = cei.secondaryOrder(ord); |
ord = cei.secondaryOrder (ord); |
468 |
|
|
469 |
default: |
default: |
470 |
break; |
break; |
471 |
} |
} |
472 |
|
|
473 |
vect.add(new Integer(ord)); |
vec.add (new Integer (ord)); |
474 |
ord = cei.next(); //increment to next key |
ord = cei.next(); //increment to next key |
475 |
} |
} |
476 |
|
|
477 |
Object[] objarr = vect.toArray(); |
Object[] objarr = vec.toArray(); |
478 |
byte[] key = new byte[objarr.length * 4]; |
byte[] key = new byte [objarr.length * 4]; |
479 |
|
|
480 |
for (int i = 0; i < objarr.length; i++) |
for (int i = 0; i < objarr.length; i++) |
481 |
{ |
{ |
482 |
int j = ((Integer)objarr[i]).intValue(); |
int j = ((Integer) objarr [i]).intValue(); |
483 |
key [i * 4] = (byte)((j & 0xFF000000) >> 24); |
key [i * 4] = (byte) ((j & 0xFF000000) >> 24); |
484 |
key [i * 4 + 1] = (byte)((j & 0x00FF0000) >> 16); |
key [i * 4 + 1] = (byte) ((j & 0x00FF0000) >> 16); |
485 |
key [i * 4 + 2] = (byte)((j & 0x0000FF00) >> 8); |
key [i * 4 + 2] = (byte) ((j & 0x0000FF00) >> 8); |
486 |
key [i * 4 + 3] = (byte)(j & 0x000000FF); |
key [i * 4 + 3] = (byte) (j & 0x000000FF); |
487 |
} |
} |
488 |
|
|
489 |
return new CollationKey (this, source, key); |
return new CollationKey (this, source, key); |