blob: 29f84997e62c34edaa184044dae1bf11321e14e7 [file] [log] [blame]
Elliott Hughes2faa5f12012-01-30 14:42:07 -08001/*
2 * Copyright (C) 2011 The Android Open Source Project
3 *
4 * Licensed under the Apache License, Version 2.0 (the "License");
5 * you may not use this file except in compliance with the License.
6 * You may obtain a copy of the License at
7 *
8 * http://www.apache.org/licenses/LICENSE-2.0
9 *
10 * Unless required by applicable law or agreed to in writing, software
11 * distributed under the License is distributed on an "AS IS" BASIS,
12 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13 * See the License for the specific language governing permissions and
14 * limitations under the License.
15 */
Elliott Hughes814e4032011-08-23 12:07:56 -070016
Brian Carlstromfc0e3212013-07-17 14:40:12 -070017#ifndef ART_RUNTIME_UTF_H_
18#define ART_RUNTIME_UTF_H_
Elliott Hughes814e4032011-08-23 12:07:56 -070019
Elliott Hughes76160052012-12-12 16:31:20 -080020#include "base/macros.h"
Ian Rogers719d1a32014-03-06 12:13:39 -080021#include "base/mutex.h"
Elliott Hughes76160052012-12-12 16:31:20 -080022
Elliott Hughes814e4032011-08-23 12:07:56 -070023#include <stddef.h>
24#include <stdint.h>
25
Elliott Hughesb465ab02011-08-24 11:21:21 -070026/*
27 * All UTF-8 in art is actually modified UTF-8. Mostly, this distinction
28 * doesn't matter.
29 *
30 * See http://en.wikipedia.org/wiki/UTF-8#Modified_UTF-8 for the details.
31 */
Elliott Hughes814e4032011-08-23 12:07:56 -070032namespace art {
Ian Rogersa6724902013-09-23 09:23:37 -070033
Ian Rogers2dd0e2c2013-01-24 12:42:14 -080034namespace mirror {
Ian Rogersa6724902013-09-23 09:23:37 -070035 template<class T> class PrimitiveArray;
36 typedef PrimitiveArray<uint16_t> CharArray;
Ian Rogers2dd0e2c2013-01-24 12:42:14 -080037} // namespace mirror
Ian Rogers0cfe1fb2011-08-26 03:29:44 -070038
Elliott Hughes814e4032011-08-23 12:07:56 -070039/*
40 * Returns the number of UTF-16 characters in the given modified UTF-8 string.
41 */
42size_t CountModifiedUtf8Chars(const char* utf8);
43
44/*
45 * Returns the number of modified UTF-8 bytes needed to represent the given
46 * UTF-16 string.
47 */
48size_t CountUtf8Bytes(const uint16_t* chars, size_t char_count);
49
50/*
Elliott Hughesb465ab02011-08-24 11:21:21 -070051 * Convert from Modified UTF-8 to UTF-16.
Elliott Hughes814e4032011-08-23 12:07:56 -070052 */
Elliott Hughesb465ab02011-08-24 11:21:21 -070053void ConvertModifiedUtf8ToUtf16(uint16_t* utf16_out, const char* utf8_in);
54
55/*
Ian Rogers0571d352011-11-03 19:51:38 -070056 * Compare two modified UTF-8 strings as UTF-16 code point values in a non-locale sensitive manner
57 */
58int CompareModifiedUtf8ToModifiedUtf8AsUtf16CodePointValues(const char* utf8_1, const char* utf8_2);
59
60/*
Ian Rogers637c65b2013-05-31 11:46:00 -070061 * Compare a modified UTF-8 string with a UTF-16 string as code point values in a non-locale
62 * sensitive manner.
63 */
64int CompareModifiedUtf8ToUtf16AsCodePointValues(const char* utf8_1, const uint16_t* utf8_2);
65
66/*
Elliott Hughesb465ab02011-08-24 11:21:21 -070067 * Convert from UTF-16 to Modified UTF-8. Note that the output is _not_
68 * NUL-terminated. You probably need to call CountUtf8Bytes before calling
69 * this anyway, so if you want a NUL-terminated string, you know where to
70 * put the NUL byte.
71 */
72void ConvertUtf16ToModifiedUtf8(char* utf8_out, const uint16_t* utf16_in, size_t char_count);
Elliott Hughes814e4032011-08-23 12:07:56 -070073
74/*
75 * The java.lang.String hashCode() algorithm.
76 */
Ian Rogersef7d42f2014-01-06 12:55:46 -080077int32_t ComputeUtf16Hash(mirror::CharArray* chars, int32_t offset, size_t char_count)
Ian Rogersb726dcb2012-09-05 08:57:23 -070078 SHARED_LOCKS_REQUIRED(Locks::mutator_lock_);
Elliott Hughes814e4032011-08-23 12:07:56 -070079int32_t ComputeUtf16Hash(const uint16_t* chars, size_t char_count);
80
81/*
82 * Retrieve the next UTF-16 character from a UTF-8 string.
83 *
84 * Advances "*utf8_data_in" to the start of the next character.
85 *
86 * WARNING: If a string is corrupted by dropping a '\0' in the middle
87 * of a 3-byte sequence, you can end up overrunning the buffer with
88 * reads (and possibly with the writes if the length was computed and
89 * cached before the damage). For performance reasons, this function
90 * assumes that the string being parsed is known to be valid (e.g., by
91 * already being verified). Most strings we process here are coming
92 * out of dex files or other internal translations, so the only real
93 * risk comes from the JNI NewStringUTF call.
94 */
95uint16_t GetUtf16FromUtf8(const char** utf8_data_in);
96
97} // namespace art
98
Brian Carlstromfc0e3212013-07-17 14:40:12 -070099#endif // ART_RUNTIME_UTF_H_