author | thartmann |
Fri, 27 Nov 2015 09:37:33 +0100 | |
changeset 34517 | c6e795a80c80 |
parent 33663 | 2cd62a4bd471 |
child 34711 | 65544417508e |
permissions | -rw-r--r-- |
33663 | 1 |
/* |
2 |
* Copyright (c) 2015, Oracle and/or its affiliates. All rights reserved. |
|
3 |
* DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. |
|
4 |
* |
|
5 |
* This code is free software; you can redistribute it and/or modify it |
|
6 |
* under the terms of the GNU General Public License version 2 only, as |
|
7 |
* published by the Free Software Foundation. Oracle designates this |
|
8 |
* particular file as subject to the "Classpath" exception as provided |
|
9 |
* by Oracle in the LICENSE file that accompanied this code. |
|
10 |
* |
|
11 |
* This code is distributed in the hope that it will be useful, but WITHOUT |
|
12 |
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or |
|
13 |
* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License |
|
14 |
* version 2 for more details (a copy is included in the LICENSE file that |
|
15 |
* accompanied this code). |
|
16 |
* |
|
17 |
* You should have received a copy of the GNU General Public License version |
|
18 |
* 2 along with this work; if not, write to the Free Software Foundation, |
|
19 |
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. |
|
20 |
* |
|
21 |
* Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA |
|
22 |
* or visit www.oracle.com if you need additional information or have any |
|
23 |
* questions. |
|
24 |
*/ |
|
25 |
||
26 |
package java.lang; |
|
27 |
||
28 |
import java.util.Arrays; |
|
29 |
import java.util.Locale; |
|
30 |
import java.util.Spliterator; |
|
31 |
import java.util.function.IntConsumer; |
|
32 |
import jdk.internal.HotSpotIntrinsicCandidate; |
|
33 |
||
34 |
import static java.lang.String.UTF16; |
|
35 |
import static java.lang.String.LATIN1; |
|
36 |
import static java.lang.String.checkIndex; |
|
37 |
import static java.lang.String.checkOffset; |
|
34517
c6e795a80c80
8142303: C2 compilation fails with "bad AD file"
thartmann
parents:
33663
diff
changeset
|
38 |
import static java.lang.String.checkBoundsOffCount; |
33663 | 39 |
|
40 |
final class StringUTF16 { |
|
41 |
||
42 |
public static byte[] newBytesFor(int len) { |
|
43 |
if (len < 0) { |
|
44 |
throw new NegativeArraySizeException(); |
|
45 |
} |
|
46 |
if (len > MAX_LENGTH) { |
|
47 |
throw new OutOfMemoryError("UTF16 String size is " + len + |
|
48 |
", should be less than " + MAX_LENGTH); |
|
49 |
} |
|
50 |
return new byte[len << 1]; |
|
51 |
} |
|
52 |
||
53 |
@HotSpotIntrinsicCandidate |
|
54 |
public static void putChar(byte[] val, int index, int c) { |
|
55 |
index <<= 1; |
|
56 |
val[index++] = (byte)(c >> HI_BYTE_SHIFT); |
|
57 |
val[index] = (byte)(c >> LO_BYTE_SHIFT); |
|
58 |
} |
|
59 |
||
60 |
@HotSpotIntrinsicCandidate |
|
61 |
public static char getChar(byte[] val, int index) { |
|
62 |
index <<= 1; |
|
63 |
return (char)(((val[index++] & 0xff) << HI_BYTE_SHIFT) | |
|
64 |
((val[index] & 0xff) << LO_BYTE_SHIFT)); |
|
65 |
} |
|
66 |
||
67 |
public static char charAt(byte[] value, int index) { |
|
68 |
if (index < 0 || index >= value.length >> 1) { |
|
69 |
throw new StringIndexOutOfBoundsException(index); |
|
70 |
} |
|
71 |
return getChar(value, index); |
|
72 |
} |
|
73 |
||
74 |
public static int length(byte[] value) { |
|
75 |
return value.length >> 1; |
|
76 |
} |
|
77 |
||
78 |
public static int codePointAt(byte[] value, int index, int end) { |
|
79 |
char c1 = getChar(value, index); |
|
80 |
if (Character.isHighSurrogate(c1) && ++index < end) { |
|
81 |
char c2 = getChar(value, index); |
|
82 |
if (Character.isLowSurrogate(c2)) { |
|
83 |
return Character.toCodePoint(c1, c2); |
|
84 |
} |
|
85 |
} |
|
86 |
return c1; |
|
87 |
} |
|
88 |
||
89 |
public static int codePointBefore(byte[] value, int index) { |
|
90 |
char c2 = getChar(value, --index); |
|
91 |
if (Character.isLowSurrogate(c2) && index > 0) { |
|
92 |
char c1 = getChar(value, --index); |
|
93 |
if (Character.isHighSurrogate(c1)) { |
|
94 |
return Character.toCodePoint(c1, c2); |
|
95 |
} |
|
96 |
} |
|
97 |
return c2; |
|
98 |
} |
|
99 |
||
100 |
public static int codePointCount(byte[] value, int beginIndex, int endIndex) { |
|
101 |
int count = endIndex - beginIndex; |
|
102 |
for (int i = beginIndex; i < endIndex; ) { |
|
103 |
if (Character.isHighSurrogate(getChar(value, i++)) && |
|
104 |
i < endIndex && |
|
105 |
Character.isLowSurrogate(getChar(value, i))) { |
|
106 |
count--; |
|
107 |
i++; |
|
108 |
} |
|
109 |
} |
|
110 |
return count; |
|
111 |
} |
|
112 |
||
113 |
public static char[] toChars(byte[] value) { |
|
114 |
char[] dst = new char[value.length >> 1]; |
|
115 |
getChars(value, 0, dst.length, dst, 0); |
|
116 |
return dst; |
|
117 |
} |
|
118 |
||
119 |
@HotSpotIntrinsicCandidate |
|
120 |
public static byte[] toBytes(char[] value, int off, int len) { |
|
121 |
byte[] val = newBytesFor(len); |
|
122 |
for (int i = 0; i < len; i++) { |
|
123 |
putChar(val, i, value[off++]); |
|
124 |
} |
|
125 |
return val; |
|
126 |
} |
|
127 |
||
128 |
public static byte[] compress(char[] val, int off, int len) { |
|
129 |
byte[] ret = new byte[len]; |
|
130 |
if (compress(val, off, ret, 0, len) == len) { |
|
131 |
return ret; |
|
132 |
} |
|
133 |
return null; |
|
134 |
} |
|
135 |
||
136 |
public static byte[] compress(byte[] val, int off, int len) { |
|
137 |
byte[] ret = new byte[len]; |
|
138 |
if (compress(val, off, ret, 0, len) == len) { |
|
139 |
return ret; |
|
140 |
} |
|
141 |
return null; |
|
142 |
} |
|
143 |
||
144 |
// compressedCopy char[] -> byte[] |
|
145 |
@HotSpotIntrinsicCandidate |
|
146 |
private static int compress(char[] src, int srcOff, byte[] dst, int dstOff, int len) { |
|
147 |
for (int i = 0; i < len; i++) { |
|
148 |
int c = src[srcOff++]; |
|
149 |
if (c >>> 8 != 0) { |
|
150 |
return 0; |
|
151 |
} |
|
152 |
dst[dstOff++] = (byte)c; |
|
153 |
} |
|
154 |
return len; |
|
155 |
} |
|
156 |
||
157 |
// compressedCopy byte[] -> byte[] |
|
158 |
@HotSpotIntrinsicCandidate |
|
159 |
public static int compress(byte[] src, int srcOff, byte[] dst, int dstOff, int len) { |
|
34517
c6e795a80c80
8142303: C2 compilation fails with "bad AD file"
thartmann
parents:
33663
diff
changeset
|
160 |
// We need a range check here because 'getChar' has no checks |
c6e795a80c80
8142303: C2 compilation fails with "bad AD file"
thartmann
parents:
33663
diff
changeset
|
161 |
checkBoundsOffCount(srcOff, len, src.length); |
33663 | 162 |
for (int i = 0; i < len; i++) { |
163 |
int c = getChar(src, srcOff++); |
|
164 |
if (c >>> 8 != 0) { |
|
165 |
return 0; |
|
166 |
} |
|
167 |
dst[dstOff++] = (byte)c; |
|
168 |
} |
|
169 |
return len; |
|
170 |
} |
|
171 |
||
172 |
public static byte[] toBytes(int[] val, int index, int len) { |
|
173 |
final int end = index + len; |
|
174 |
// Pass 1: Compute precise size of char[] |
|
175 |
int n = len; |
|
176 |
for (int i = index; i < end; i++) { |
|
177 |
int cp = val[i]; |
|
178 |
if (Character.isBmpCodePoint(cp)) |
|
179 |
continue; |
|
180 |
else if (Character.isValidCodePoint(cp)) |
|
181 |
n++; |
|
182 |
else throw new IllegalArgumentException(Integer.toString(cp)); |
|
183 |
} |
|
184 |
// Pass 2: Allocate and fill in <high, low> pair |
|
185 |
byte[] buf = newBytesFor(n); |
|
186 |
for (int i = index, j = 0; i < end; i++, j++) { |
|
187 |
int cp = val[i]; |
|
188 |
if (Character.isBmpCodePoint(cp)) { |
|
189 |
putChar(buf, j, cp); |
|
190 |
} else { |
|
191 |
putChar(buf, j++, Character.highSurrogate(cp)); |
|
192 |
putChar(buf, j, Character.lowSurrogate(cp)); |
|
193 |
} |
|
194 |
} |
|
195 |
return buf; |
|
196 |
} |
|
197 |
||
198 |
public static byte[] toBytes(char c) { |
|
199 |
byte[] result = new byte[2]; |
|
200 |
putChar(result, 0, c); |
|
201 |
return result; |
|
202 |
} |
|
203 |
||
204 |
@HotSpotIntrinsicCandidate |
|
205 |
public static void getChars(byte[] value, int srcBegin, int srcEnd, char dst[], int dstBegin) { |
|
34517
c6e795a80c80
8142303: C2 compilation fails with "bad AD file"
thartmann
parents:
33663
diff
changeset
|
206 |
// We need a range check here because 'getChar' has no checks |
c6e795a80c80
8142303: C2 compilation fails with "bad AD file"
thartmann
parents:
33663
diff
changeset
|
207 |
checkBoundsOffCount(srcBegin, srcEnd - srcBegin, value.length); |
33663 | 208 |
for (int i = srcBegin; i < srcEnd; i++) { |
209 |
dst[dstBegin++] = getChar(value, i); |
|
210 |
} |
|
211 |
} |
|
212 |
||
213 |
/* @see java.lang.String.getBytes(int, int, byte[], int) */ |
|
214 |
public static void getBytes(byte[] value, int srcBegin, int srcEnd, byte dst[], int dstBegin) { |
|
215 |
srcBegin <<= 1; |
|
216 |
srcEnd <<= 1; |
|
217 |
for (int i = srcBegin + (1 >> LO_BYTE_SHIFT); i < srcEnd; i += 2) { |
|
218 |
dst[dstBegin++] = value[i]; |
|
219 |
} |
|
220 |
} |
|
221 |
||
222 |
@HotSpotIntrinsicCandidate |
|
223 |
public static boolean equals(byte[] value, byte[] other) { |
|
224 |
if (value.length == other.length) { |
|
225 |
int len = value.length >> 1; |
|
226 |
for (int i = 0; i < len; i++) { |
|
227 |
if (getChar(value, i) != getChar(other, i)) { |
|
228 |
return false; |
|
229 |
} |
|
230 |
} |
|
231 |
return true; |
|
232 |
} |
|
233 |
return false; |
|
234 |
} |
|
235 |
||
236 |
@HotSpotIntrinsicCandidate |
|
237 |
public static int compareTo(byte[] value, byte[] other) { |
|
238 |
int len1 = length(value); |
|
239 |
int len2 = length(other); |
|
240 |
int lim = Math.min(len1, len2); |
|
241 |
for (int k = 0; k < lim; k++) { |
|
242 |
char c1 = getChar(value, k); |
|
243 |
char c2 = getChar(other, k); |
|
244 |
if (c1 != c2) { |
|
245 |
return c1 - c2; |
|
246 |
} |
|
247 |
} |
|
248 |
return len1 - len2; |
|
249 |
} |
|
250 |
||
251 |
@HotSpotIntrinsicCandidate |
|
252 |
public static int compareToLatin1(byte[] value, byte[] other) { |
|
253 |
int len1 = length(value); |
|
254 |
int len2 = StringLatin1.length(other); |
|
255 |
int lim = Math.min(len1, len2); |
|
256 |
for (int k = 0; k < lim; k++) { |
|
257 |
char c1 = getChar(value, k); |
|
258 |
char c2 = StringLatin1.getChar(other, k); |
|
259 |
if (c1 != c2) { |
|
260 |
return c1 - c2; |
|
261 |
} |
|
262 |
} |
|
263 |
return len1 - len2; |
|
264 |
} |
|
265 |
||
266 |
public static int hashCode(byte[] value) { |
|
267 |
int h = 0; |
|
268 |
int length = value.length >> 1; |
|
269 |
for (int i = 0; i < length; i++) { |
|
270 |
h = 31 * h + getChar(value, i); |
|
271 |
} |
|
272 |
return h; |
|
273 |
} |
|
274 |
||
275 |
public static int indexOf(byte[] value, int ch, int fromIndex) { |
|
276 |
int max = value.length >> 1; |
|
277 |
if (fromIndex < 0) { |
|
278 |
fromIndex = 0; |
|
279 |
} else if (fromIndex >= max) { |
|
280 |
// Note: fromIndex might be near -1>>>1. |
|
281 |
return -1; |
|
282 |
} |
|
283 |
if (ch < Character.MIN_SUPPLEMENTARY_CODE_POINT) { |
|
284 |
// handle most cases here (ch is a BMP code point or a |
|
285 |
// negative value (invalid code point)) |
|
286 |
return indexOfChar(value, ch, fromIndex, max); |
|
287 |
} else { |
|
288 |
return indexOfSupplementary(value, ch, fromIndex, max); |
|
289 |
} |
|
290 |
} |
|
291 |
||
292 |
@HotSpotIntrinsicCandidate |
|
293 |
public static int indexOf(byte[] value, byte[] str) { |
|
294 |
if (str.length == 0) { |
|
295 |
return 0; |
|
296 |
} |
|
297 |
if (value.length == 0) { |
|
298 |
return -1; |
|
299 |
} |
|
300 |
return indexOf(value, length(value), str, length(str), 0); |
|
301 |
} |
|
302 |
||
303 |
@HotSpotIntrinsicCandidate |
|
304 |
public static int indexOf(byte[] value, int valueCount, byte[] str, int strCount, int fromIndex) { |
|
305 |
char first = getChar(str, 0); |
|
306 |
int max = (valueCount - strCount); |
|
307 |
for (int i = fromIndex; i <= max; i++) { |
|
308 |
// Look for first character. |
|
309 |
if (getChar(value, i) != first) { |
|
310 |
while (++i <= max && getChar(value, i) != first); |
|
311 |
} |
|
312 |
// Found first character, now look at the rest of value |
|
313 |
if (i <= max) { |
|
314 |
int j = i + 1; |
|
315 |
int end = j + strCount - 1; |
|
316 |
for (int k = 1; j < end && getChar(value, j) == getChar(str, k); j++, k++); |
|
317 |
if (j == end) { |
|
318 |
// Found whole string. |
|
319 |
return i; |
|
320 |
} |
|
321 |
} |
|
322 |
} |
|
323 |
return -1; |
|
324 |
} |
|
325 |
||
326 |
/** |
|
327 |
* Handles indexOf Latin1 substring in UTF16 string. |
|
328 |
*/ |
|
329 |
@HotSpotIntrinsicCandidate |
|
330 |
public static int indexOfLatin1(byte[] value, byte[] str) { |
|
331 |
if (str.length == 0) { |
|
332 |
return 0; |
|
333 |
} |
|
334 |
if (value.length == 0) { |
|
335 |
return -1; |
|
336 |
} |
|
337 |
return indexOfLatin1(value, length(value), str, str.length, 0); |
|
338 |
} |
|
339 |
||
340 |
@HotSpotIntrinsicCandidate |
|
341 |
public static int indexOfLatin1(byte[] src, int srcCount, byte[] tgt, int tgtCount, int fromIndex) { |
|
342 |
char first = (char)(tgt[0] & 0xff); |
|
343 |
int max = (srcCount - tgtCount); |
|
344 |
for (int i = fromIndex; i <= max; i++) { |
|
345 |
// Look for first character. |
|
346 |
if (getChar(src, i) != first) { |
|
347 |
while (++i <= max && getChar(src, i) != first); |
|
348 |
} |
|
349 |
// Found first character, now look at the rest of v2 |
|
350 |
if (i <= max) { |
|
351 |
int j = i + 1; |
|
352 |
int end = j + tgtCount - 1; |
|
353 |
for (int k = 1; |
|
354 |
j < end && getChar(src, j) == (tgt[k] & 0xff); |
|
355 |
j++, k++); |
|
356 |
if (j == end) { |
|
357 |
// Found whole string. |
|
358 |
return i; |
|
359 |
} |
|
360 |
} |
|
361 |
} |
|
362 |
return -1; |
|
363 |
} |
|
364 |
||
365 |
@HotSpotIntrinsicCandidate |
|
366 |
private static int indexOfChar(byte[] value, int ch, int fromIndex, int max) { |
|
367 |
for (int i = fromIndex; i < max; i++) { |
|
368 |
if (getChar(value, i) == ch) { |
|
369 |
return i; |
|
370 |
} |
|
371 |
} |
|
372 |
return -1; |
|
373 |
} |
|
374 |
||
375 |
/** |
|
376 |
* Handles (rare) calls of indexOf with a supplementary character. |
|
377 |
*/ |
|
378 |
private static int indexOfSupplementary(byte[] value, int ch, int fromIndex, int max) { |
|
379 |
if (Character.isValidCodePoint(ch)) { |
|
380 |
final char hi = Character.highSurrogate(ch); |
|
381 |
final char lo = Character.lowSurrogate(ch); |
|
382 |
for (int i = fromIndex; i < max - 1; i++) { |
|
383 |
if (getChar(value, i) == hi && getChar(value, i + 1 ) == lo) { |
|
384 |
return i; |
|
385 |
} |
|
386 |
} |
|
387 |
} |
|
388 |
return -1; |
|
389 |
} |
|
390 |
||
391 |
public static int lastIndexOf(byte[] src, int srcCount, |
|
392 |
byte[] tgt, int tgtCount, int fromIndex) { |
|
393 |
int min = tgtCount - 1; |
|
394 |
int i = min + fromIndex; |
|
395 |
int strLastIndex = tgtCount - 1; |
|
396 |
char strLastChar = getChar(tgt, strLastIndex); |
|
397 |
||
398 |
startSearchForLastChar: |
|
399 |
while (true) { |
|
400 |
while (i >= min && getChar(src, i) != strLastChar) { |
|
401 |
i--; |
|
402 |
} |
|
403 |
if (i < min) { |
|
404 |
return -1; |
|
405 |
} |
|
406 |
int j = i - 1; |
|
407 |
int start = j - strLastIndex; |
|
408 |
int k = strLastIndex - 1; |
|
409 |
while (j > start) { |
|
410 |
if (getChar(src, j--) != getChar(tgt, k--)) { |
|
411 |
i--; |
|
412 |
continue startSearchForLastChar; |
|
413 |
} |
|
414 |
} |
|
415 |
return start + 1; |
|
416 |
} |
|
417 |
} |
|
418 |
||
419 |
public static int lastIndexOf(byte[] value, int ch, int fromIndex) { |
|
420 |
if (ch < Character.MIN_SUPPLEMENTARY_CODE_POINT) { |
|
421 |
// handle most cases here (ch is a BMP code point or a |
|
422 |
// negative value (invalid code point)) |
|
423 |
int i = Math.min(fromIndex, (value.length >> 1) - 1); |
|
424 |
for (; i >= 0; i--) { |
|
425 |
if (getChar(value, i) == ch) { |
|
426 |
return i; |
|
427 |
} |
|
428 |
} |
|
429 |
return -1; |
|
430 |
} else { |
|
431 |
return lastIndexOfSupplementary(value, ch, fromIndex); |
|
432 |
} |
|
433 |
} |
|
434 |
||
435 |
/** |
|
436 |
* Handles (rare) calls of lastIndexOf with a supplementary character. |
|
437 |
*/ |
|
438 |
private static int lastIndexOfSupplementary(final byte[] value, int ch, int fromIndex) { |
|
439 |
if (Character.isValidCodePoint(ch)) { |
|
440 |
char hi = Character.highSurrogate(ch); |
|
441 |
char lo = Character.lowSurrogate(ch); |
|
442 |
int i = Math.min(fromIndex, (value.length >> 1) - 2); |
|
443 |
for (; i >= 0; i--) { |
|
444 |
if (getChar(value, i) == hi && getChar(value, i + 1) == lo) { |
|
445 |
return i; |
|
446 |
} |
|
447 |
} |
|
448 |
} |
|
449 |
return -1; |
|
450 |
} |
|
451 |
||
452 |
public static String replace(byte[] value, char oldChar, char newChar) { |
|
453 |
int len = value.length >> 1; |
|
454 |
int i = -1; |
|
455 |
while (++i < len) { |
|
456 |
if (getChar(value, i) == oldChar) { |
|
457 |
break; |
|
458 |
} |
|
459 |
} |
|
460 |
if (i < len) { |
|
461 |
byte buf[] = new byte[value.length]; |
|
462 |
for (int j = 0; j < i; j++) { |
|
463 |
putChar(buf, j, getChar(value, j)); // TBD:arraycopy? |
|
464 |
} |
|
465 |
while (i < len) { |
|
466 |
char c = getChar(value, i); |
|
467 |
putChar(buf, i, c == oldChar ? newChar : c); |
|
468 |
i++; |
|
469 |
} |
|
470 |
// Check if we should try to compress to latin1 |
|
471 |
if (String.COMPACT_STRINGS && |
|
472 |
!StringLatin1.canEncode(oldChar) && |
|
473 |
StringLatin1.canEncode(newChar)) { |
|
474 |
byte[] val = compress(buf, 0, len); |
|
475 |
if (val != null) { |
|
476 |
return new String(val, LATIN1); |
|
477 |
} |
|
478 |
} |
|
479 |
return new String(buf, UTF16); |
|
480 |
} |
|
481 |
return null; |
|
482 |
} |
|
483 |
||
484 |
public static boolean regionMatchesCI(byte[] value, int toffset, |
|
485 |
byte[] other, int ooffset, int len) { |
|
486 |
int last = toffset + len; |
|
487 |
while (toffset < last) { |
|
488 |
char c1 = getChar(value, toffset++); |
|
489 |
char c2 = getChar(other, ooffset++); |
|
490 |
if (c1 == c2) { |
|
491 |
continue; |
|
492 |
} |
|
493 |
// try converting both characters to uppercase. |
|
494 |
// If the results match, then the comparison scan should |
|
495 |
// continue. |
|
496 |
char u1 = Character.toUpperCase(c1); |
|
497 |
char u2 = Character.toUpperCase(c2); |
|
498 |
if (u1 == u2) { |
|
499 |
continue; |
|
500 |
} |
|
501 |
// Unfortunately, conversion to uppercase does not work properly |
|
502 |
// for the Georgian alphabet, which has strange rules about case |
|
503 |
// conversion. So we need to make one last check before |
|
504 |
// exiting. |
|
505 |
if (Character.toLowerCase(u1) == Character.toLowerCase(u2)) { |
|
506 |
continue; |
|
507 |
} |
|
508 |
return false; |
|
509 |
} |
|
510 |
return true; |
|
511 |
} |
|
512 |
||
513 |
public static boolean regionMatchesCI_Latin1(byte[] value, int toffset, |
|
514 |
byte[] other, int ooffset, |
|
515 |
int len) { |
|
516 |
int last = toffset + len; |
|
517 |
while (toffset < last) { |
|
518 |
char c1 = getChar(value, toffset++); |
|
519 |
char c2 = (char)(other[ooffset++] & 0xff); |
|
520 |
if (c1 == c2) { |
|
521 |
continue; |
|
522 |
} |
|
523 |
char u1 = Character.toUpperCase(c1); |
|
524 |
char u2 = Character.toUpperCase(c2); |
|
525 |
if (u1 == u2) { |
|
526 |
continue; |
|
527 |
} |
|
528 |
if (Character.toLowerCase(u1) == Character.toLowerCase(u2)) { |
|
529 |
continue; |
|
530 |
} |
|
531 |
return false; |
|
532 |
} |
|
533 |
return true; |
|
534 |
} |
|
535 |
||
536 |
public static String toLowerCase(String str, byte[] value, Locale locale) { |
|
537 |
if (locale == null) { |
|
538 |
throw new NullPointerException(); |
|
539 |
} |
|
540 |
int first; |
|
541 |
boolean hasSurr = false; |
|
542 |
final int len = value.length >> 1; |
|
543 |
||
544 |
// Now check if there are any characters that need to be changed, or are surrogate |
|
545 |
for (first = 0 ; first < len; first++) { |
|
546 |
int cp = (int)getChar(value, first); |
|
547 |
if (Character.isSurrogate((char)cp)) { |
|
548 |
hasSurr = true; |
|
549 |
break; |
|
550 |
} |
|
551 |
if (cp != Character.toLowerCase(cp)) { // no need to check Character.ERROR |
|
552 |
break; |
|
553 |
} |
|
554 |
} |
|
555 |
if (first == len) |
|
556 |
return str; |
|
557 |
byte[] result = new byte[value.length]; |
|
558 |
System.arraycopy(value, 0, result, 0, first << 1); // Just copy the first few |
|
559 |
// lowerCase characters. |
|
560 |
String lang = locale.getLanguage(); |
|
561 |
if (lang == "tr" || lang == "az" || lang == "lt") { |
|
562 |
return toLowerCaseEx(str, value, result, first, locale, true); |
|
563 |
} |
|
564 |
if (hasSurr) { |
|
565 |
return toLowerCaseEx(str, value, result, first, locale, false); |
|
566 |
} |
|
567 |
int bits = 0; |
|
568 |
for (int i = first; i < len; i++) { |
|
569 |
int cp = (int)getChar(value, i); |
|
570 |
if (cp == '\u03A3' || // GREEK CAPITAL LETTER SIGMA |
|
571 |
Character.isSurrogate((char)cp)) { |
|
572 |
return toLowerCaseEx(str, value, result, i, locale, false); |
|
573 |
} |
|
574 |
if (cp == '\u0130') { // LATIN CAPITAL LETTER I WITH DOT ABOVE |
|
575 |
return toLowerCaseEx(str, value, result, i, locale, true); |
|
576 |
} |
|
577 |
cp = Character.toLowerCase(cp); |
|
578 |
if (!Character.isBmpCodePoint(cp)) { |
|
579 |
return toLowerCaseEx(str, value, result, i, locale, false); |
|
580 |
} |
|
581 |
bits |= cp; |
|
582 |
putChar(result, i, cp); |
|
583 |
} |
|
584 |
if (bits >>> 8 != 0) { |
|
585 |
return new String(result, UTF16); |
|
586 |
} else { |
|
587 |
return newString(result, 0, len); |
|
588 |
} |
|
589 |
} |
|
590 |
||
591 |
private static String toLowerCaseEx(String str, byte[] value, |
|
592 |
byte[] result, int first, Locale locale, |
|
593 |
boolean localeDependent) { |
|
594 |
int resultOffset = first; |
|
595 |
int length = value.length >> 1; |
|
596 |
int srcCount; |
|
597 |
for (int i = first; i < length; i += srcCount) { |
|
598 |
int srcChar = getChar(value, i); |
|
599 |
int lowerChar; |
|
600 |
char[] lowerCharArray; |
|
601 |
srcCount = 1; |
|
602 |
if (Character.isSurrogate((char)srcChar)) { |
|
603 |
srcChar = codePointAt(value, i, length); |
|
604 |
srcCount = Character.charCount(srcChar); |
|
605 |
} |
|
606 |
if (localeDependent || |
|
607 |
srcChar == '\u03A3' || // GREEK CAPITAL LETTER SIGMA |
|
608 |
srcChar == '\u0130') { // LATIN CAPITAL LETTER I WITH DOT ABOVE |
|
609 |
lowerChar = ConditionalSpecialCasing.toLowerCaseEx(str, i, locale); |
|
610 |
} else { |
|
611 |
lowerChar = Character.toLowerCase(srcChar); |
|
612 |
} |
|
613 |
if (Character.isBmpCodePoint(lowerChar)) { // Character.ERROR is not a bmp |
|
614 |
putChar(result, resultOffset++, lowerChar); |
|
615 |
} else { |
|
616 |
if (lowerChar == Character.ERROR) { |
|
617 |
lowerCharArray = ConditionalSpecialCasing.toLowerCaseCharArray(str, i, locale); |
|
618 |
} else { |
|
619 |
lowerCharArray = Character.toChars(lowerChar); |
|
620 |
} |
|
621 |
/* Grow result if needed */ |
|
622 |
int mapLen = lowerCharArray.length; |
|
623 |
if (mapLen > srcCount) { |
|
624 |
byte[] result2 = newBytesFor((result.length >> 1) + mapLen - srcCount); |
|
625 |
System.arraycopy(result, 0, result2, 0, resultOffset << 1); |
|
626 |
result = result2; |
|
627 |
} |
|
628 |
for (int x = 0; x < mapLen; ++x) { |
|
629 |
putChar(result, resultOffset++, lowerCharArray[x]); |
|
630 |
} |
|
631 |
} |
|
632 |
} |
|
633 |
return newString(result, 0, resultOffset); |
|
634 |
} |
|
635 |
||
636 |
public static String toUpperCase(String str, byte[] value, Locale locale) { |
|
637 |
if (locale == null) { |
|
638 |
throw new NullPointerException(); |
|
639 |
} |
|
640 |
int first; |
|
641 |
boolean hasSurr = false; |
|
642 |
final int len = value.length >> 1; |
|
643 |
||
644 |
// Now check if there are any characters that need to be changed, or are surrogate |
|
645 |
for (first = 0 ; first < len; first++) { |
|
646 |
int cp = (int)getChar(value, first); |
|
647 |
if (Character.isSurrogate((char)cp)) { |
|
648 |
hasSurr = true; |
|
649 |
break; |
|
650 |
} |
|
651 |
if (cp != Character.toUpperCaseEx(cp)) { // no need to check Character.ERROR |
|
652 |
break; |
|
653 |
} |
|
654 |
} |
|
655 |
if (first == len) { |
|
656 |
return str; |
|
657 |
} |
|
658 |
byte[] result = new byte[value.length]; |
|
659 |
System.arraycopy(value, 0, result, 0, first << 1); // Just copy the first few |
|
660 |
// upperCase characters. |
|
661 |
String lang = locale.getLanguage(); |
|
662 |
if (lang == "tr" || lang == "az" || lang == "lt") { |
|
663 |
return toUpperCaseEx(str, value, result, first, locale, true); |
|
664 |
} |
|
665 |
if (hasSurr) { |
|
666 |
return toUpperCaseEx(str, value, result, first, locale, false); |
|
667 |
} |
|
668 |
int bits = 0; |
|
669 |
for (int i = first; i < len; i++) { |
|
670 |
int cp = (int)getChar(value, i); |
|
671 |
if (Character.isSurrogate((char)cp)) { |
|
672 |
return toUpperCaseEx(str, value, result, i, locale, false); |
|
673 |
} |
|
674 |
cp = Character.toUpperCaseEx(cp); |
|
675 |
if (!Character.isBmpCodePoint(cp)) { // Character.ERROR is not bmp |
|
676 |
return toUpperCaseEx(str, value, result, i, locale, false); |
|
677 |
} |
|
678 |
bits |= cp; |
|
679 |
putChar(result, i, cp); |
|
680 |
} |
|
681 |
if (bits >>> 8 != 0) { |
|
682 |
return new String(result, UTF16); |
|
683 |
} else { |
|
684 |
return newString(result, 0, len); |
|
685 |
} |
|
686 |
} |
|
687 |
||
688 |
private static String toUpperCaseEx(String str, byte[] value, |
|
689 |
byte[] result, int first, |
|
690 |
Locale locale, boolean localeDependent) |
|
691 |
{ |
|
692 |
int resultOffset = first; |
|
693 |
int length = value.length >> 1; |
|
694 |
int srcCount; |
|
695 |
for (int i = first; i < length; i += srcCount) { |
|
696 |
int srcChar = getChar(value, i); |
|
697 |
int upperChar; |
|
698 |
char[] upperCharArray; |
|
699 |
srcCount = 1; |
|
700 |
if (Character.isSurrogate((char)srcChar)) { |
|
701 |
srcChar = codePointAt(value, i, length); |
|
702 |
srcCount = Character.charCount(srcChar); |
|
703 |
} |
|
704 |
if (localeDependent) { |
|
705 |
upperChar = ConditionalSpecialCasing.toUpperCaseEx(str, i, locale); |
|
706 |
} else { |
|
707 |
upperChar = Character.toUpperCaseEx(srcChar); |
|
708 |
} |
|
709 |
if (Character.isBmpCodePoint(upperChar)) { |
|
710 |
putChar(result, resultOffset++, upperChar); |
|
711 |
} else { |
|
712 |
if (upperChar == Character.ERROR) { |
|
713 |
if (localeDependent) { |
|
714 |
upperCharArray = |
|
715 |
ConditionalSpecialCasing.toUpperCaseCharArray(str, i, locale); |
|
716 |
} else { |
|
717 |
upperCharArray = Character.toUpperCaseCharArray(srcChar); |
|
718 |
} |
|
719 |
} else { |
|
720 |
upperCharArray = Character.toChars(upperChar); |
|
721 |
} |
|
722 |
/* Grow result if needed */ |
|
723 |
int mapLen = upperCharArray.length; |
|
724 |
if (mapLen > srcCount) { |
|
725 |
byte[] result2 = newBytesFor((result.length >> 1) + mapLen - srcCount); |
|
726 |
System.arraycopy(result, 0, result2, 0, resultOffset << 1); |
|
727 |
result = result2; |
|
728 |
} |
|
729 |
for (int x = 0; x < mapLen; ++x) { |
|
730 |
putChar(result, resultOffset++, upperCharArray[x]); |
|
731 |
} |
|
732 |
} |
|
733 |
} |
|
734 |
return newString(result, 0, resultOffset); |
|
735 |
} |
|
736 |
||
737 |
public static String trim(byte[] value) { |
|
738 |
int length = value.length >> 1; |
|
739 |
int len = length; |
|
740 |
int st = 0; |
|
741 |
while (st < len && getChar(value, st) <= ' ') { |
|
742 |
st++; |
|
743 |
} |
|
744 |
while (st < len && getChar(value, len - 1) <= ' ') { |
|
745 |
len--; |
|
746 |
} |
|
747 |
return ((st > 0) || (len < length )) ? |
|
748 |
new String(Arrays.copyOfRange(value, st << 1, len << 1), UTF16) : |
|
749 |
null; |
|
750 |
} |
|
751 |
||
752 |
public static void putChars(byte[] val, int index, char[] str, int off, int end) { |
|
753 |
while (off < end) { |
|
754 |
putChar(val, index++, str[off++]); |
|
755 |
} |
|
756 |
} |
|
757 |
||
758 |
public static String newString(byte[] val, int index, int len) { |
|
759 |
if (String.COMPACT_STRINGS) { |
|
760 |
byte[] buf = compress(val, index, len); |
|
761 |
if (buf != null) { |
|
762 |
return new String(buf, LATIN1); |
|
763 |
} |
|
764 |
} |
|
765 |
int last = index + len; |
|
766 |
return new String(Arrays.copyOfRange(val, index << 1, last << 1), UTF16); |
|
767 |
} |
|
768 |
||
769 |
public static void fillNull(byte[] val, int index, int end) { |
|
770 |
Arrays.fill(val, index << 1, end << 1, (byte)0); |
|
771 |
} |
|
772 |
||
773 |
static class CharsSpliterator implements Spliterator.OfInt { |
|
774 |
private final byte[] array; |
|
775 |
private int index; // current index, modified on advance/split |
|
776 |
private final int fence; // one past last index |
|
777 |
private final int cs; |
|
778 |
||
779 |
CharsSpliterator(byte[] array, int acs) { |
|
780 |
this(array, 0, array.length >> 1, acs); |
|
781 |
} |
|
782 |
||
783 |
CharsSpliterator(byte[] array, int origin, int fence, int acs) { |
|
784 |
this.array = array; |
|
785 |
this.index = origin; |
|
786 |
this.fence = fence; |
|
787 |
this.cs = acs | Spliterator.ORDERED | Spliterator.SIZED |
|
788 |
| Spliterator.SUBSIZED; |
|
789 |
} |
|
790 |
||
791 |
@Override |
|
792 |
public OfInt trySplit() { |
|
793 |
int lo = index, mid = (lo + fence) >>> 1; |
|
794 |
return (lo >= mid) |
|
795 |
? null |
|
796 |
: new CharsSpliterator(array, lo, index = mid, cs); |
|
797 |
} |
|
798 |
||
799 |
@Override |
|
800 |
public void forEachRemaining(IntConsumer action) { |
|
801 |
byte[] a; int i, hi; // hoist accesses and checks from loop |
|
802 |
if (action == null) |
|
803 |
throw new NullPointerException(); |
|
804 |
if (((a = array).length >> 1) >= (hi = fence) && |
|
805 |
(i = index) >= 0 && i < (index = hi)) { |
|
806 |
do { action.accept(getChar(a, i)); } while (++i < hi); |
|
807 |
} |
|
808 |
} |
|
809 |
||
810 |
@Override |
|
811 |
public boolean tryAdvance(IntConsumer action) { |
|
812 |
if (action == null) |
|
813 |
throw new NullPointerException(); |
|
814 |
if (index >= 0 && index < fence) { |
|
815 |
action.accept(getChar(array, index++)); |
|
816 |
return true; |
|
817 |
} |
|
818 |
return false; |
|
819 |
} |
|
820 |
||
821 |
@Override |
|
822 |
public long estimateSize() { return (long)(fence - index); } |
|
823 |
||
824 |
@Override |
|
825 |
public int characteristics() { |
|
826 |
return cs; |
|
827 |
} |
|
828 |
} |
|
829 |
||
830 |
static class CodePointsSpliterator implements Spliterator.OfInt { |
|
831 |
private final byte[] array; |
|
832 |
private int index; // current index, modified on advance/split |
|
833 |
private final int fence; // one past last index |
|
834 |
private final int cs; |
|
835 |
||
836 |
CodePointsSpliterator(byte[] array, int acs) { |
|
837 |
this(array, 0, array.length >> 1, acs); |
|
838 |
} |
|
839 |
||
840 |
CodePointsSpliterator(byte[] array, int origin, int fence, int acs) { |
|
841 |
this.array = array; |
|
842 |
this.index = origin; |
|
843 |
this.fence = fence; |
|
844 |
this.cs = acs | Spliterator.ORDERED; |
|
845 |
} |
|
846 |
||
847 |
@Override |
|
848 |
public OfInt trySplit() { |
|
849 |
int lo = index, mid = (lo + fence) >>> 1; |
|
850 |
if (lo >= mid) |
|
851 |
return null; |
|
852 |
||
853 |
int midOneLess; |
|
854 |
// If the mid-point intersects a surrogate pair |
|
855 |
if (Character.isLowSurrogate(getChar(array, mid)) && |
|
856 |
Character.isHighSurrogate(getChar(array, midOneLess = (mid -1)))) { |
|
857 |
// If there is only one pair it cannot be split |
|
858 |
if (lo >= midOneLess) |
|
859 |
return null; |
|
860 |
// Shift the mid-point to align with the surrogate pair |
|
861 |
return new CodePointsSpliterator(array, lo, index = midOneLess, cs); |
|
862 |
} |
|
863 |
return new CodePointsSpliterator(array, lo, index = mid, cs); |
|
864 |
} |
|
865 |
||
866 |
@Override |
|
867 |
public void forEachRemaining(IntConsumer action) { |
|
868 |
byte[] a; int i, hi; // hoist accesses and checks from loop |
|
869 |
if (action == null) |
|
870 |
throw new NullPointerException(); |
|
871 |
if (((a = array).length >> 1) >= (hi = fence) && |
|
872 |
(i = index) >= 0 && i < (index = hi)) { |
|
873 |
do { |
|
874 |
i = advance(a, i, hi, action); |
|
875 |
} while (i < hi); |
|
876 |
} |
|
877 |
} |
|
878 |
||
879 |
@Override |
|
880 |
public boolean tryAdvance(IntConsumer action) { |
|
881 |
if (action == null) |
|
882 |
throw new NullPointerException(); |
|
883 |
if (index >= 0 && index < fence) { |
|
884 |
index = advance(array, index, fence, action); |
|
885 |
return true; |
|
886 |
} |
|
887 |
return false; |
|
888 |
} |
|
889 |
||
890 |
// Advance one code point from the index, i, and return the next |
|
891 |
// index to advance from |
|
892 |
private static int advance(byte[] a, int i, int hi, IntConsumer action) { |
|
893 |
char c1 = getChar(a, i++); |
|
894 |
int cp = c1; |
|
895 |
if (Character.isHighSurrogate(c1) && i < hi) { |
|
896 |
char c2 = getChar(a, i); |
|
897 |
if (Character.isLowSurrogate(c2)) { |
|
898 |
i++; |
|
899 |
cp = Character.toCodePoint(c1, c2); |
|
900 |
} |
|
901 |
} |
|
902 |
action.accept(cp); |
|
903 |
return i; |
|
904 |
} |
|
905 |
||
906 |
@Override |
|
907 |
public long estimateSize() { return (long)(fence - index); } |
|
908 |
||
909 |
@Override |
|
910 |
public int characteristics() { |
|
911 |
return cs; |
|
912 |
} |
|
913 |
} |
|
914 |
||
915 |
//////////////////////////////////////////////////////////////// |
|
916 |
||
917 |
public static void putCharSB(byte[] val, int index, int c) { |
|
918 |
checkIndex(index, val.length >> 1); |
|
919 |
putChar(val, index, c); |
|
920 |
} |
|
921 |
||
922 |
public static void putCharsSB(byte[] val, int index, char[] ca, int off, int end) { |
|
923 |
checkOffset(index + end - off, val.length >> 1); |
|
924 |
putChars(val, index, ca, off, end); |
|
925 |
} |
|
926 |
||
927 |
public static void putCharsSB(byte[] val, int index, CharSequence s, int off, int end) { |
|
928 |
checkOffset(index + end - off, val.length >> 1); |
|
929 |
for (int i = off; i < end; i++) { |
|
930 |
putChar(val, index++, s.charAt(i)); |
|
931 |
} |
|
932 |
} |
|
933 |
||
934 |
public static int codePointAtSB(byte[] val, int index, int end) { |
|
935 |
checkOffset(end, val.length >> 1); |
|
936 |
return codePointAt(val, index, end); |
|
937 |
} |
|
938 |
||
939 |
public static int codePointBeforeSB(byte[] val, int index) { |
|
940 |
checkOffset(index, val.length >> 1); |
|
941 |
return codePointBefore(val, index); |
|
942 |
} |
|
943 |
||
944 |
public static int codePointCountSB(byte[] val, int beginIndex, int endIndex) { |
|
945 |
checkOffset(endIndex, val.length >> 1); |
|
946 |
return codePointCount(val, beginIndex, endIndex); |
|
947 |
} |
|
948 |
||
949 |
//////////////////////////////////////////////////////////////// |
|
950 |
||
951 |
private static native boolean isBigEndian(); |
|
952 |
||
953 |
static final int HI_BYTE_SHIFT; |
|
954 |
static final int LO_BYTE_SHIFT; |
|
955 |
static { |
|
956 |
if (isBigEndian()) { |
|
957 |
HI_BYTE_SHIFT = 8; |
|
958 |
LO_BYTE_SHIFT = 0; |
|
959 |
} else { |
|
960 |
HI_BYTE_SHIFT = 0; |
|
961 |
LO_BYTE_SHIFT = 8; |
|
962 |
} |
|
963 |
} |
|
964 |
||
965 |
static final int MAX_LENGTH = Integer.MAX_VALUE >> 1; |
|
966 |
} |