DXR is a code search and navigation tool aimed at making sense of large projects. It supports full-text and regex searches as well as structural queries.

Mercurial (b6057e17f856)

VCS Links

Line Code
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208
/* -*- Mode: C++; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 4 -*-
 * vim: set ts=8 sts=4 et sw=4 tw=99:
 * This Source Code Form is subject to the terms of the Mozilla Public
 * License, v. 2.0. If a copy of the MPL was not distributed with this
 * file, You can obtain one at http://mozilla.org/MPL/2.0/. */

#ifndef js_CharacterEncoding_h
#define js_CharacterEncoding_h

#include "mozilla/NullPtr.h"
#include "mozilla/Range.h"

#include "js/TypeDecls.h"
#include "js/Utility.h"

namespace js {
struct ThreadSafeContext;
}

namespace JS {

/*
 * By default, all C/C++ 1-byte-per-character strings passed into the JSAPI
 * are treated as ISO/IEC 8859-1, also known as Latin-1. That is, each
 * byte is treated as a 2-byte character, and there is no way to pass in a
 * string containing characters beyond U+00FF.
 */
class Latin1Chars : public mozilla::Range<Latin1Char>
{
    typedef mozilla::Range<Latin1Char> Base;

  public:
    Latin1Chars() : Base() {}
    Latin1Chars(char* aBytes, size_t aLength) : Base(reinterpret_cast<Latin1Char*>(aBytes), aLength) {}
    Latin1Chars(const Latin1Char* aBytes, size_t aLength)
      : Base(const_cast<Latin1Char*>(aBytes), aLength)
    {}
    Latin1Chars(const char* aBytes, size_t aLength)
      : Base(reinterpret_cast<Latin1Char*>(const_cast<char*>(aBytes)), aLength)
    {}
};

/*
 * A Latin1Chars, but with \0 termination for C compatibility.
 */
class Latin1CharsZ : public mozilla::RangedPtr<Latin1Char>
{
    typedef mozilla::RangedPtr<Latin1Char> Base;

  public:
    Latin1CharsZ() : Base(nullptr, 0) {}

    Latin1CharsZ(char* aBytes, size_t aLength)
      : Base(reinterpret_cast<Latin1Char*>(aBytes), aLength)
    {
        MOZ_ASSERT(aBytes[aLength] == '\0');
    }

    Latin1CharsZ(Latin1Char* aBytes, size_t aLength)
      : Base(aBytes, aLength)
    {
        MOZ_ASSERT(aBytes[aLength] == '\0');
    }

    using Base::operator=;

    char* c_str() { return reinterpret_cast<char*>(get()); }
};

class UTF8Chars : public mozilla::Range<unsigned char>
{
    typedef mozilla::Range<unsigned char> Base;

  public:
    UTF8Chars() : Base() {}
    UTF8Chars(char* aBytes, size_t aLength)
      : Base(reinterpret_cast<unsigned char*>(aBytes), aLength)
    {}
    UTF8Chars(const char* aBytes, size_t aLength)
      : Base(reinterpret_cast<unsigned char*>(const_cast<char*>(aBytes)), aLength)
    {}
};

/*
 * SpiderMonkey also deals directly with UTF-8 encoded text in some places.
 */
class UTF8CharsZ : public mozilla::RangedPtr<unsigned char>
{
    typedef mozilla::RangedPtr<unsigned char> Base;

  public:
    UTF8CharsZ() : Base(nullptr, 0) {}

    UTF8CharsZ(char* aBytes, size_t aLength)
      : Base(reinterpret_cast<unsigned char*>(aBytes), aLength)
    {
        MOZ_ASSERT(aBytes[aLength] == '\0');
    }

    UTF8CharsZ(unsigned char* aBytes, size_t aLength)
      : Base(aBytes, aLength)
    {
        MOZ_ASSERT(aBytes[aLength] == '\0');
    }

    using Base::operator=;

    char* c_str() { return reinterpret_cast<char*>(get()); }
};

/*
 * SpiderMonkey uses a 2-byte character representation: it is a
 * 2-byte-at-a-time view of a UTF-16 byte stream. This is similar to UCS-2,
 * but unlike UCS-2, we do not strip UTF-16 extension bytes. This allows a
 * sufficiently dedicated JavaScript program to be fully unicode-aware by
 * manually interpreting UTF-16 extension characters embedded in the JS
 * string.
 */
class TwoByteChars : public mozilla::Range<jschar>
{
    typedef mozilla::Range<jschar> Base;

  public:
    TwoByteChars() : Base() {}
    TwoByteChars(jschar* aChars, size_t aLength) : Base(aChars, aLength) {}
    TwoByteChars(const jschar* aChars, size_t aLength) : Base(const_cast<jschar*>(aChars), aLength) {}
};

/*
 * A TwoByteChars, but \0 terminated for compatibility with JSFlatString.
 */
class TwoByteCharsZ : public mozilla::RangedPtr<jschar>
{
    typedef mozilla::RangedPtr<jschar> Base;

  public:
    TwoByteCharsZ() : Base(nullptr, 0) {}

    TwoByteCharsZ(jschar* chars, size_t length)
      : Base(chars, length)
    {
        MOZ_ASSERT(chars[length] == '\0');
    }

    using Base::operator=;
};

typedef mozilla::RangedPtr<const jschar> ConstCharPtr;

/*
 * Like TwoByteChars, but the chars are const.
 */
class ConstTwoByteChars : public mozilla::RangedPtr<const jschar>
{
  public:
    ConstTwoByteChars(const ConstTwoByteChars& s) : ConstCharPtr(s) {}
    MOZ_IMPLICIT ConstTwoByteChars(const mozilla::RangedPtr<const jschar>& s) : ConstCharPtr(s) {}
    ConstTwoByteChars(const jschar* s, size_t len) : ConstCharPtr(s, len) {}
    ConstTwoByteChars(const jschar* pos, const jschar* start, size_t len)
      : ConstCharPtr(pos, start, len)
    {}

    using ConstCharPtr::operator=;
};


/*
 * Convert a 2-byte character sequence to "ISO-Latin-1". This works by
 * truncating each 2-byte pair in the sequence to a 1-byte pair. If the source
 * contains any UTF-16 extension characters, then this may give invalid Latin1
 * output. The returned string is zero terminated. The returned string or the
 * returned string's |start()| must be freed with JS_free or js_free,
 * respectively. If allocation fails, an OOM error will be set and the method
 * will return a nullptr chars (which can be tested for with the ! operator).
 * This method cannot trigger GC.
 */
extern Latin1CharsZ
LossyTwoByteCharsToNewLatin1CharsZ(js::ThreadSafeContext* cx, TwoByteChars tbchars);

extern UTF8CharsZ
TwoByteCharsToNewUTF8CharsZ(js::ThreadSafeContext* cx, TwoByteChars tbchars);

uint32_t
Utf8ToOneUcs4Char(const uint8_t* utf8Buffer, int utf8Length);

/*
 * Inflate bytes in UTF-8 encoding to jschars.
 * - On error, returns an empty TwoByteCharsZ.
 * - On success, returns a malloc'd TwoByteCharsZ, and updates |outlen| to hold
 *   its length;  the length value excludes the trailing null.
 */
extern TwoByteCharsZ
UTF8CharsToNewTwoByteCharsZ(JSContext* cx, const UTF8Chars utf8, size_t* outlen);

/*
 * The same as UTF8CharsToNewTwoByteCharsZ(), except that any malformed UTF-8 characters
 * will be replaced by \uFFFD. No exception will be thrown for malformed UTF-8
 * input.
 */
extern TwoByteCharsZ
LossyUTF8CharsToNewTwoByteCharsZ(JSContext* cx, const UTF8Chars utf8, size_t* outlen);

} // namespace JS

inline void JS_free(JS::Latin1CharsZ& ptr) { js_free((void*)ptr.get()); }
inline void JS_free(JS::UTF8CharsZ& ptr) { js_free((void*)ptr.get()); }

#endif /* js_CharacterEncoding_h */