Elliott Hughes | 2faa5f1 | 2012-01-30 14:42:07 -0800 | [diff] [blame] | 1 | /* |
| 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 Hughes | 814e403 | 2011-08-23 12:07:56 -0700 | [diff] [blame] | 16 | |
Brian Carlstrom | fc0e321 | 2013-07-17 14:40:12 -0700 | [diff] [blame] | 17 | #ifndef ART_RUNTIME_UTF_H_ |
| 18 | #define ART_RUNTIME_UTF_H_ |
Elliott Hughes | 814e403 | 2011-08-23 12:07:56 -0700 | [diff] [blame] | 19 | |
Elliott Hughes | 7616005 | 2012-12-12 16:31:20 -0800 | [diff] [blame] | 20 | #include "base/macros.h" |
Ian Rogers | 719d1a3 | 2014-03-06 12:13:39 -0800 | [diff] [blame] | 21 | #include "base/mutex.h" |
Elliott Hughes | 7616005 | 2012-12-12 16:31:20 -0800 | [diff] [blame] | 22 | |
Elliott Hughes | 814e403 | 2011-08-23 12:07:56 -0700 | [diff] [blame] | 23 | #include <stddef.h> |
| 24 | #include <stdint.h> |
| 25 | |
Elliott Hughes | b465ab0 | 2011-08-24 11:21:21 -0700 | [diff] [blame] | 26 | /* |
| 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 Hughes | 814e403 | 2011-08-23 12:07:56 -0700 | [diff] [blame] | 32 | namespace art { |
Ian Rogers | a672490 | 2013-09-23 09:23:37 -0700 | [diff] [blame] | 33 | |
Ian Rogers | 2dd0e2c | 2013-01-24 12:42:14 -0800 | [diff] [blame] | 34 | namespace mirror { |
Ian Rogers | a672490 | 2013-09-23 09:23:37 -0700 | [diff] [blame] | 35 | template<class T> class PrimitiveArray; |
| 36 | typedef PrimitiveArray<uint16_t> CharArray; |
Ian Rogers | 2dd0e2c | 2013-01-24 12:42:14 -0800 | [diff] [blame] | 37 | } // namespace mirror |
Ian Rogers | 0cfe1fb | 2011-08-26 03:29:44 -0700 | [diff] [blame] | 38 | |
Elliott Hughes | 814e403 | 2011-08-23 12:07:56 -0700 | [diff] [blame] | 39 | /* |
| 40 | * Returns the number of UTF-16 characters in the given modified UTF-8 string. |
| 41 | */ |
| 42 | size_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 | */ |
| 48 | size_t CountUtf8Bytes(const uint16_t* chars, size_t char_count); |
| 49 | |
| 50 | /* |
Elliott Hughes | b465ab0 | 2011-08-24 11:21:21 -0700 | [diff] [blame] | 51 | * Convert from Modified UTF-8 to UTF-16. |
Elliott Hughes | 814e403 | 2011-08-23 12:07:56 -0700 | [diff] [blame] | 52 | */ |
Elliott Hughes | b465ab0 | 2011-08-24 11:21:21 -0700 | [diff] [blame] | 53 | void ConvertModifiedUtf8ToUtf16(uint16_t* utf16_out, const char* utf8_in); |
| 54 | |
| 55 | /* |
Ian Rogers | 0571d35 | 2011-11-03 19:51:38 -0700 | [diff] [blame] | 56 | * Compare two modified UTF-8 strings as UTF-16 code point values in a non-locale sensitive manner |
| 57 | */ |
Ian Rogers | 1ff3c98 | 2014-08-12 02:30:58 -0700 | [diff] [blame] | 58 | ALWAYS_INLINE int CompareModifiedUtf8ToModifiedUtf8AsUtf16CodePointValues(const char* utf8_1, |
| 59 | const char* utf8_2); |
Ian Rogers | 0571d35 | 2011-11-03 19:51:38 -0700 | [diff] [blame] | 60 | |
| 61 | /* |
Vladimir Marko | a48aef4 | 2014-12-03 17:53:53 +0000 | [diff] [blame] | 62 | * 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 Rogers | 637c65b | 2013-05-31 11:46:00 -0700 | [diff] [blame] | 64 | */ |
Vladimir Marko | a48aef4 | 2014-12-03 17:53:53 +0000 | [diff] [blame] | 65 | int CompareModifiedUtf8ToUtf16AsCodePointValues(const char* utf8, const uint16_t* utf16, |
| 66 | size_t utf16_length); |
Ian Rogers | 637c65b | 2013-05-31 11:46:00 -0700 | [diff] [blame] | 67 | |
| 68 | /* |
Elliott Hughes | b465ab0 | 2011-08-24 11:21:21 -0700 | [diff] [blame] | 69 | * 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 | */ |
| 74 | void ConvertUtf16ToModifiedUtf8(char* utf8_out, const uint16_t* utf16_in, size_t char_count); |
Elliott Hughes | 814e403 | 2011-08-23 12:07:56 -0700 | [diff] [blame] | 75 | |
| 76 | /* |
| 77 | * The java.lang.String hashCode() algorithm. |
| 78 | */ |
Ian Rogers | ef7d42f | 2014-01-06 12:55:46 -0800 | [diff] [blame] | 79 | int32_t ComputeUtf16Hash(mirror::CharArray* chars, int32_t offset, size_t char_count) |
Mathieu Chartier | 9044347 | 2015-07-16 20:32:27 -0700 | [diff] [blame] | 80 | SHARED_REQUIRES(Locks::mutator_lock_); |
Elliott Hughes | 814e403 | 2011-08-23 12:07:56 -0700 | [diff] [blame] | 81 | int32_t ComputeUtf16Hash(const uint16_t* chars, size_t char_count); |
| 82 | |
Mathieu Chartier | e7c9a8c | 2014-11-06 16:35:45 -0800 | [diff] [blame] | 83 | // 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. |
| 85 | size_t ComputeModifiedUtf8Hash(const char* chars); |
Ian Rogers | 68b5685 | 2014-08-29 20:19:11 -0700 | [diff] [blame] | 86 | |
Elliott Hughes | 814e403 | 2011-08-23 12:07:56 -0700 | [diff] [blame] | 87 | /* |
Narayan Kamath | a5afcfc | 2015-01-29 20:06:46 +0000 | [diff] [blame] | 88 | * 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 Kamath | 8508e37 | 2015-05-06 14:55:43 +0100 | [diff] [blame] | 90 | * 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 Hughes | 814e403 | 2011-08-23 12:07:56 -0700 | [diff] [blame] | 93 | * |
| 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 Kamath | a5afcfc | 2015-01-29 20:06:46 +0000 | [diff] [blame] | 97 | * of a multi byte sequence, you can end up overrunning the buffer with |
Elliott Hughes | 814e403 | 2011-08-23 12:07:56 -0700 | [diff] [blame] | 98 | * 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 Kamath | a5afcfc | 2015-01-29 20:06:46 +0000 | [diff] [blame] | 105 | uint32_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 | */ |
| 111 | ALWAYS_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 | */ |
| 117 | ALWAYS_INLINE uint16_t GetTrailingUtf16Char(uint32_t maybe_pair); |
Elliott Hughes | 814e403 | 2011-08-23 12:07:56 -0700 | [diff] [blame] | 118 | |
| 119 | } // namespace art |
| 120 | |
Brian Carlstrom | fc0e321 | 2013-07-17 14:40:12 -0700 | [diff] [blame] | 121 | #endif // ART_RUNTIME_UTF_H_ |