0N/A/*
2273N/A * Copyright (c) 1997, 2011, Oracle and/or its affiliates. All rights reserved.
0N/A * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
0N/A *
0N/A * This code is free software; you can redistribute it and/or modify it
0N/A * under the terms of the GNU General Public License version 2 only, as
0N/A * published by the Free Software Foundation.
0N/A *
0N/A * This code is distributed in the hope that it will be useful, but WITHOUT
0N/A * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
0N/A * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
0N/A * version 2 for more details (a copy is included in the LICENSE file that
0N/A * accompanied this code).
0N/A *
0N/A * You should have received a copy of the GNU General Public License version
0N/A * 2 along with this work; if not, write to the Free Software Foundation,
0N/A * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
0N/A *
1472N/A * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
1472N/A * or visit www.oracle.com if you need additional information or have any
1472N/A * questions.
0N/A *
0N/A */
0N/A
1879N/A#include "precompiled.hpp"
1879N/A#include "utilities/utf8.hpp"
0N/A
0N/A// Assume the utf8 string is in legal form and has been
0N/A// checked in the class file parser/format checker.
0N/Achar* UTF8::next(const char* str, jchar* value) {
0N/A unsigned const char *ptr = (const unsigned char *)str;
0N/A unsigned char ch, ch2, ch3;
0N/A int length = -1; /* bad length */
0N/A jchar result;
0N/A switch ((ch = ptr[0]) >> 4) {
0N/A default:
0N/A result = ch;
0N/A length = 1;
0N/A break;
0N/A
0N/A case 0x8: case 0x9: case 0xA: case 0xB: case 0xF:
0N/A /* Shouldn't happen. */
0N/A break;
0N/A
0N/A case 0xC: case 0xD:
0N/A /* 110xxxxx 10xxxxxx */
0N/A if (((ch2 = ptr[1]) & 0xC0) == 0x80) {
0N/A unsigned char high_five = ch & 0x1F;
0N/A unsigned char low_six = ch2 & 0x3F;
0N/A result = (high_five << 6) + low_six;
0N/A length = 2;
0N/A break;
0N/A }
0N/A break;
0N/A
0N/A case 0xE:
0N/A /* 1110xxxx 10xxxxxx 10xxxxxx */
0N/A if (((ch2 = ptr[1]) & 0xC0) == 0x80) {
0N/A if (((ch3 = ptr[2]) & 0xC0) == 0x80) {
0N/A unsigned char high_four = ch & 0x0f;
0N/A unsigned char mid_six = ch2 & 0x3f;
0N/A unsigned char low_six = ch3 & 0x3f;
0N/A result = (((high_four << 6) + mid_six) << 6) + low_six;
0N/A length = 3;
0N/A }
0N/A }
0N/A break;
0N/A } /* end of switch */
0N/A
0N/A if (length <= 0) {
0N/A *value = ptr[0]; /* default bad result; */
0N/A return (char*)(ptr + 1); // make progress somehow
0N/A }
0N/A
0N/A *value = result;
0N/A
0N/A // The assert is correct but the .class file is wrong
0N/A // assert(UNICODE::utf8_size(result) == length, "checking reverse computation");
0N/A return (char *)(ptr + length);
0N/A}
0N/A
0N/Achar* UTF8::next_character(const char* str, jint* value) {
0N/A unsigned const char *ptr = (const unsigned char *)str;
0N/A /* See if it's legal supplementary character:
0N/A 11101101 1010xxxx 10xxxxxx 11101101 1011xxxx 10xxxxxx */
0N/A if (is_supplementary_character(ptr)) {
0N/A *value = get_supplementary_character(ptr);
0N/A return (char *)(ptr + 6);
0N/A }
0N/A jchar result;
0N/A char* next_ch = next(str, &result);
0N/A *value = result;
0N/A return next_ch;
0N/A}
0N/A
0N/A// Count bytes of the form 10xxxxxx and deduct this count
0N/A// from the total byte count. The utf8 string must be in
0N/A// legal form which has been verified in the format checker.
0N/Aint UTF8::unicode_length(const char* str, int len) {
0N/A int num_chars = len;
0N/A for (int i = 0; i < len; i++) {
0N/A if ((str[i] & 0xC0) == 0x80) {
0N/A --num_chars;
0N/A }
0N/A }
0N/A return num_chars;
0N/A}
0N/A
0N/A// Count bytes of the utf8 string except those in form
0N/A// 10xxxxxx which only appear in multibyte characters.
0N/A// The utf8 string must be in legal form and has been
0N/A// verified in the format checker.
0N/Aint UTF8::unicode_length(const char* str) {
0N/A int num_chars = 0;
0N/A for (const char* p = str; *p; p++) {
0N/A if (((*p) & 0xC0) != 0x80) {
0N/A num_chars++;
0N/A }
0N/A }
0N/A return num_chars;
0N/A}
0N/A
0N/A// Writes a jchar a utf8 and returns the end
0N/Astatic u_char* utf8_write(u_char* base, jchar ch) {
0N/A if ((ch != 0) && (ch <=0x7f)) {
0N/A base[0] = (u_char) ch;
0N/A return base + 1;
0N/A }
0N/A
0N/A if (ch <= 0x7FF) {
0N/A /* 11 bits or less. */
0N/A unsigned char high_five = ch >> 6;
0N/A unsigned char low_six = ch & 0x3F;
0N/A base[0] = high_five | 0xC0; /* 110xxxxx */
0N/A base[1] = low_six | 0x80; /* 10xxxxxx */
0N/A return base + 2;
0N/A }
0N/A /* possibly full 16 bits. */
0N/A char high_four = ch >> 12;
0N/A char mid_six = (ch >> 6) & 0x3F;
0N/A char low_six = ch & 0x3f;
0N/A base[0] = high_four | 0xE0; /* 1110xxxx */
0N/A base[1] = mid_six | 0x80; /* 10xxxxxx */
0N/A base[2] = low_six | 0x80; /* 10xxxxxx */
0N/A return base + 3;
0N/A}
0N/A
0N/Avoid UTF8::convert_to_unicode(const char* utf8_str, jchar* unicode_str, int unicode_length) {
0N/A unsigned char ch;
0N/A const char *ptr = (const char *)utf8_str;
0N/A int index = 0;
0N/A
0N/A /* ASCII case loop optimization */
0N/A for (; index < unicode_length; index++) {
0N/A if((ch = ptr[0]) > 0x7F) { break; }
0N/A unicode_str[index] = ch;
0N/A ptr = (const char *)(ptr + 1);
0N/A }
0N/A
0N/A for (; index < unicode_length; index++) {
0N/A ptr = UTF8::next(ptr, &unicode_str[index]);
0N/A }
0N/A}
0N/A
0N/A// Returns NULL if 'c' it not found. This only works as long
0N/A// as 'c' is an ASCII character
2062N/Aconst jbyte* UTF8::strrchr(const jbyte* base, int length, jbyte c) {
0N/A assert(length >= 0, "sanity check");
0N/A assert(c >= 0, "does not work for non-ASCII characters");
0N/A // Skip backwards in string until 'c' is found or end is reached
0N/A while(--length >= 0 && base[length] != c);
0N/A return (length < 0) ? NULL : &base[length];
0N/A}
0N/A
2062N/Abool UTF8::equal(const jbyte* base1, int length1, const jbyte* base2, int length2) {
0N/A // Length must be the same
0N/A if (length1 != length2) return false;
0N/A for (int i = 0; i < length1; i++) {
0N/A if (base1[i] != base2[i]) return false;
0N/A }
0N/A return true;
0N/A}
0N/A
0N/Abool UTF8::is_supplementary_character(const unsigned char* str) {
0N/A return ((str[0] & 0xFF) == 0xED) && ((str[1] & 0xF0) == 0xA0) && ((str[2] & 0xC0) == 0x80)
0N/A && ((str[3] & 0xFF) == 0xED) && ((str[4] & 0xF0) == 0xB0) && ((str[5] & 0xC0) == 0x80);
0N/A}
0N/A
0N/Ajint UTF8::get_supplementary_character(const unsigned char* str) {
0N/A return 0x10000 + ((str[1] & 0x0f) << 16) + ((str[2] & 0x3f) << 10)
0N/A + ((str[4] & 0x0f) << 6) + (str[5] & 0x3f);
0N/A}
0N/A
0N/A
0N/A//-------------------------------------------------------------------------------------
0N/A
0N/A
0N/Aint UNICODE::utf8_size(jchar c) {
0N/A if ((0x0001 <= c) && (c <= 0x007F)) return 1;
0N/A if (c <= 0x07FF) return 2;
0N/A return 3;
0N/A}
0N/A
0N/Aint UNICODE::utf8_length(jchar* base, int length) {
0N/A int result = 0;
0N/A for (int index = 0; index < length; index++) {
0N/A jchar c = base[index];
0N/A if ((0x0001 <= c) && (c <= 0x007F)) result += 1;
0N/A else if (c <= 0x07FF) result += 2;
0N/A else result += 3;
0N/A }
0N/A return result;
0N/A}
0N/A
0N/Achar* UNICODE::as_utf8(jchar* base, int length) {
0N/A int utf8_len = utf8_length(base, length);
0N/A u_char* result = NEW_RESOURCE_ARRAY(u_char, utf8_len + 1);
0N/A u_char* p = result;
0N/A for (int index = 0; index < length; index++) {
0N/A p = utf8_write(p, base[index]);
0N/A }
0N/A *p = '\0';
0N/A assert(p == &result[utf8_len], "length prediction must be correct");
0N/A return (char*) result;
0N/A}
0N/A
0N/Achar* UNICODE::as_utf8(jchar* base, int length, char* buf, int buflen) {
0N/A u_char* p = (u_char*)buf;
0N/A u_char* end = (u_char*)buf + buflen;
0N/A for (int index = 0; index < length; index++) {
0N/A jchar c = base[index];
0N/A if (p + utf8_size(c) >= end) break; // string is truncated
0N/A p = utf8_write(p, base[index]);
0N/A }
0N/A *p = '\0';
0N/A return buf;
0N/A}
0N/A
0N/Avoid UNICODE::convert_to_utf8(const jchar* base, int length, char* utf8_buffer) {
0N/A for(int index = 0; index < length; index++) {
0N/A utf8_buffer = (char*)utf8_write((u_char*)utf8_buffer, base[index]);
0N/A }
0N/A *utf8_buffer = '\0';
0N/A}