blob: 7f05248c29d4efd0fe2ebf8805e039c8476a6b67 [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 */
Ian Rogers1ff3c982014-08-12 02:30:58 -070058ALWAYS_INLINE int CompareModifiedUtf8ToModifiedUtf8AsUtf16CodePointValues(const char* utf8_1,
59 const char* utf8_2);
Ian Rogers0571d352011-11-03 19:51:38 -070060
61/*
Vladimir Markoa48aef42014-12-03 17:53:53 +000062 * Compare a null-terminated modified UTF-8 string with a UTF-16 string (not null-terminated)
63 * as code point values in a non-locale sensitive manner.
Ian Rogers637c65b2013-05-31 11:46:00 -070064 */
Vladimir Markoa48aef42014-12-03 17:53:53 +000065int CompareModifiedUtf8ToUtf16AsCodePointValues(const char* utf8, const uint16_t* utf16,
66 size_t utf16_length);
Ian Rogers637c65b2013-05-31 11:46:00 -070067
68/*
Elliott Hughesb465ab02011-08-24 11:21:21 -070069 * Convert from UTF-16 to Modified UTF-8. Note that the output is _not_
70 * NUL-terminated. You probably need to call CountUtf8Bytes before calling
71 * this anyway, so if you want a NUL-terminated string, you know where to
72 * put the NUL byte.
73 */
74void ConvertUtf16ToModifiedUtf8(char* utf8_out, const uint16_t* utf16_in, size_t char_count);
Elliott Hughes814e4032011-08-23 12:07:56 -070075
76/*
77 * The java.lang.String hashCode() algorithm.
78 */
Ian Rogersef7d42f2014-01-06 12:55:46 -080079int32_t ComputeUtf16Hash(mirror::CharArray* chars, int32_t offset, size_t char_count)
Ian Rogersb726dcb2012-09-05 08:57:23 -070080 SHARED_LOCKS_REQUIRED(Locks::mutator_lock_);
Elliott Hughes814e4032011-08-23 12:07:56 -070081int32_t ComputeUtf16Hash(const uint16_t* chars, size_t char_count);
82
Mathieu Chartiere7c9a8c2014-11-06 16:35:45 -080083// Compute a hash code of a modified UTF-8 string. Not the standard java hash since it returns a
84// size_t and hashes individual chars instead of codepoint words.
85size_t ComputeModifiedUtf8Hash(const char* chars);
Ian Rogers68b56852014-08-29 20:19:11 -070086
Elliott Hughes814e4032011-08-23 12:07:56 -070087/*
Narayan Kamatha5afcfc2015-01-29 20:06:46 +000088 * Retrieve the next UTF-16 character or surrogate pair from a UTF-8 string.
89 * single byte, 2-byte and 3-byte UTF-8 sequences result in a single UTF-16
Narayan Kamath8508e372015-05-06 14:55:43 +010090 * character (possibly one half of a surrogate) whereas 4-byte UTF-8 sequences
91 * result in a surrogate pair. Use GetLeadingUtf16Char and GetTrailingUtf16Char
92 * to process the return value of this function.
Elliott Hughes814e4032011-08-23 12:07:56 -070093 *
94 * Advances "*utf8_data_in" to the start of the next character.
95 *
96 * WARNING: If a string is corrupted by dropping a '\0' in the middle
Narayan Kamatha5afcfc2015-01-29 20:06:46 +000097 * of a multi byte sequence, you can end up overrunning the buffer with
Elliott Hughes814e4032011-08-23 12:07:56 -070098 * reads (and possibly with the writes if the length was computed and
99 * cached before the damage). For performance reasons, this function
100 * assumes that the string being parsed is known to be valid (e.g., by
101 * already being verified). Most strings we process here are coming
102 * out of dex files or other internal translations, so the only real
103 * risk comes from the JNI NewStringUTF call.
104 */
Narayan Kamatha5afcfc2015-01-29 20:06:46 +0000105uint32_t GetUtf16FromUtf8(const char** utf8_data_in);
106
107/**
108 * Gets the leading UTF-16 character from a surrogate pair, or the sole
109 * UTF-16 character from the return value of GetUtf16FromUtf8.
110 */
111ALWAYS_INLINE uint16_t GetLeadingUtf16Char(uint32_t maybe_pair);
112
113/**
114 * Gets the trailing UTF-16 character from a surrogate pair, or 0 otherwise
115 * from the return value of GetUtf16FromUtf8.
116 */
117ALWAYS_INLINE uint16_t GetTrailingUtf16Char(uint32_t maybe_pair);
Elliott Hughes814e4032011-08-23 12:07:56 -0700118
119} // namespace art
120
Brian Carlstromfc0e3212013-07-17 14:40:12 -0700121#endif // ART_RUNTIME_UTF_H_