[1] | 1 | /* $Id: utf-16-printf.cpp 98103 2023-01-17 14:15:46Z vboxsync $ */
|
---|
| 2 | /** @file
|
---|
[83837] | 3 | * IPRT - String Formatters, Outputting UTF-16.
|
---|
[1] | 4 | */
|
---|
| 5 |
|
---|
| 6 | /*
|
---|
[98103] | 7 | * Copyright (C) 2006-2023 Oracle and/or its affiliates.
|
---|
[1] | 8 | *
|
---|
[96407] | 9 | * This file is part of VirtualBox base platform packages, as
|
---|
| 10 | * available from https://www.virtualbox.org.
|
---|
[5999] | 11 | *
|
---|
[96407] | 12 | * This program is free software; you can redistribute it and/or
|
---|
| 13 | * modify it under the terms of the GNU General Public License
|
---|
| 14 | * as published by the Free Software Foundation, in version 3 of the
|
---|
| 15 | * License.
|
---|
| 16 | *
|
---|
| 17 | * This program is distributed in the hope that it will be useful, but
|
---|
| 18 | * WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 19 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
---|
| 20 | * General Public License for more details.
|
---|
| 21 | *
|
---|
| 22 | * You should have received a copy of the GNU General Public License
|
---|
| 23 | * along with this program; if not, see <https://www.gnu.org/licenses>.
|
---|
| 24 | *
|
---|
[5999] | 25 | * The contents of this file may alternatively be used under the terms
|
---|
| 26 | * of the Common Development and Distribution License Version 1.0
|
---|
[96407] | 27 | * (CDDL), a copy of it is provided in the "COPYING.CDDL" file included
|
---|
| 28 | * in the VirtualBox distribution, in which case the provisions of the
|
---|
[5999] | 29 | * CDDL are applicable instead of those of the GPL.
|
---|
| 30 | *
|
---|
| 31 | * You may elect to license modified versions of this file under the
|
---|
| 32 | * terms and conditions of either the GPL or the CDDL or both.
|
---|
[96407] | 33 | *
|
---|
| 34 | * SPDX-License-Identifier: GPL-3.0-only OR CDDL-1.0
|
---|
[1] | 35 | */
|
---|
| 36 |
|
---|
| 37 |
|
---|
[57358] | 38 | /*********************************************************************************************************************************
|
---|
| 39 | * Header Files *
|
---|
| 40 | *********************************************************************************************************************************/
|
---|
[83837] | 41 | #include <iprt/utf16.h>
|
---|
[21337] | 42 | #include "internal/iprt.h"
|
---|
| 43 |
|
---|
[1] | 44 | #include <iprt/assert.h>
|
---|
[83837] | 45 | #include <iprt/string.h>
|
---|
| 46 | #include <iprt/uni.h>
|
---|
[1] | 47 |
|
---|
| 48 |
|
---|
[57358] | 49 | /*********************************************************************************************************************************
|
---|
| 50 | * Structures and Typedefs *
|
---|
| 51 | *********************************************************************************************************************************/
|
---|
[83837] | 52 | /** rtUtf16PrintfOutput() argument structure. */
|
---|
| 53 | typedef struct UTF16PRINTFOUTPUTARGS
|
---|
[1] | 54 | {
|
---|
| 55 | /** Pointer to current buffer position. */
|
---|
[83837] | 56 | PRTUTF16 pwszCur;
|
---|
| 57 | /** Number of RTUTF16 units left in the buffer (including the trailing zero). */
|
---|
| 58 | size_t cwcLeft;
|
---|
[64340] | 59 | /** Set if we overflowed. */
|
---|
[83837] | 60 | bool fOverflowed;
|
---|
| 61 | } UTF16PRINTFOUTPUTARGS;
|
---|
| 62 | /** Pointer to a rtUtf16PrintfOutput() argument structure. */
|
---|
| 63 | typedef UTF16PRINTFOUTPUTARGS *PUTF16PRINTFOUTPUTARGS;
|
---|
[1] | 64 |
|
---|
| 65 |
|
---|
| 66 | /**
|
---|
| 67 | * Output callback.
|
---|
| 68 | *
|
---|
[83837] | 69 | * @returns Number of RTUTF16 units we (would have) outputted.
|
---|
[64340] | 70 | *
|
---|
[1] | 71 | * @param pvArg Pointer to a STRBUFARG structure.
|
---|
| 72 | * @param pachChars Pointer to an array of utf-8 characters.
|
---|
| 73 | * @param cbChars Number of bytes in the character array pointed to by pachChars.
|
---|
| 74 | */
|
---|
[83837] | 75 | static DECLCALLBACK(size_t) rtUtf16PrintfOutput(void *pvArg, const char *pachChars, size_t cbChars)
|
---|
[1] | 76 | {
|
---|
[83837] | 77 | PUTF16PRINTFOUTPUTARGS pArgs = (PUTF16PRINTFOUTPUTARGS)pvArg;
|
---|
| 78 | size_t cwcRet = 0;
|
---|
[1] | 79 |
|
---|
[83837] | 80 | size_t cwcLeft = pArgs->cwcLeft;
|
---|
| 81 | if (cwcLeft > 1)
|
---|
[1] | 82 | {
|
---|
[83837] | 83 | Assert(!pArgs->fOverflowed);
|
---|
[64340] | 84 |
|
---|
[83837] | 85 | PRTUTF16 pwszCur = pArgs->pwszCur;
|
---|
| 86 | for (;;)
|
---|
[64340] | 87 | {
|
---|
[83837] | 88 | if (cbChars > 0)
|
---|
| 89 | {
|
---|
| 90 | RTUNICP uc;
|
---|
| 91 | int rc = RTStrGetCpNEx(&pachChars, &cbChars, &uc);
|
---|
[83943] | 92 | AssertRCStmt(rc, uc = 0xfffd /* REPLACEMENT */);
|
---|
[83837] | 93 |
|
---|
| 94 | /* Simple: */
|
---|
| 95 | if (RTUniCpIsBMP(uc))
|
---|
| 96 | {
|
---|
| 97 | cwcRet += 1;
|
---|
| 98 | if (RT_LIKELY(cwcLeft > 1))
|
---|
| 99 | *pwszCur++ = uc;
|
---|
| 100 | else
|
---|
| 101 | break;
|
---|
| 102 | cwcLeft--;
|
---|
| 103 | }
|
---|
| 104 | /* Surrogate pair: */
|
---|
| 105 | else if (uc >= 0x10000 && uc <= 0x0010ffff)
|
---|
| 106 | {
|
---|
| 107 | cwcRet += 2;
|
---|
| 108 | if (RT_LIKELY(cwcLeft > 2))
|
---|
| 109 | *pwszCur++ = 0xd800 | (uc >> 10);
|
---|
| 110 | else
|
---|
| 111 | {
|
---|
| 112 | if (cwcLeft > 1)
|
---|
| 113 | {
|
---|
| 114 | cwcLeft = 1;
|
---|
| 115 | pwszCur[1] = '\0';
|
---|
| 116 | }
|
---|
| 117 | break;
|
---|
| 118 | }
|
---|
| 119 | *pwszCur++ = 0xdc00 | (uc & 0x3ff);
|
---|
| 120 | cwcLeft -= 2;
|
---|
| 121 | }
|
---|
| 122 | else
|
---|
| 123 | {
|
---|
| 124 | AssertMsgFailed(("uc=%#x\n", uc));
|
---|
| 125 | cwcRet += 1;
|
---|
| 126 | if (RT_LIKELY(cwcLeft > 1))
|
---|
[83943] | 127 | *pwszCur++ = 0xfffd; /* REPLACEMENT */
|
---|
[83837] | 128 | else
|
---|
| 129 | break;
|
---|
| 130 | cwcLeft--;
|
---|
| 131 | }
|
---|
| 132 | }
|
---|
| 133 | else
|
---|
| 134 | {
|
---|
| 135 | *pwszCur = '\0';
|
---|
| 136 | pArgs->pwszCur = pwszCur;
|
---|
| 137 | pArgs->cwcLeft = cwcLeft;
|
---|
| 138 | return cwcRet;
|
---|
| 139 | }
|
---|
[64340] | 140 | }
|
---|
[83837] | 141 |
|
---|
| 142 | /*
|
---|
| 143 | * We only get here if we run out of buffer space.
|
---|
| 144 | */
|
---|
| 145 | Assert(cwcLeft == 1);
|
---|
| 146 | *pwszCur = '\0';
|
---|
| 147 | pArgs->pwszCur = pwszCur;
|
---|
| 148 | pArgs->cwcLeft = cwcLeft;
|
---|
[1] | 149 | }
|
---|
[83837] | 150 | /*
|
---|
| 151 | * We get a special zero byte call at the end for the formatting operation.
|
---|
| 152 | *
|
---|
| 153 | * Make sure we don't turn that into an overflow and that we'll terminate
|
---|
| 154 | * empty result strings.
|
---|
| 155 | */
|
---|
| 156 | else if (cbChars == 0 && cwcLeft > 0)
|
---|
[64340] | 157 | {
|
---|
[83837] | 158 | *pArgs->pwszCur = '\0';
|
---|
| 159 | return 0;
|
---|
| 160 | }
|
---|
| 161 |
|
---|
| 162 | /*
|
---|
| 163 | * Overflow handling. Calc needed space.
|
---|
| 164 | */
|
---|
| 165 | pArgs->fOverflowed = true;
|
---|
| 166 |
|
---|
| 167 | while (cbChars > 0)
|
---|
| 168 | {
|
---|
| 169 | RTUNICP uc;
|
---|
| 170 | int rc = RTStrGetCpNEx(&pachChars, &cbChars, &uc);
|
---|
[83943] | 171 | AssertRCStmt(rc, uc = 0xfffd /* REPLACEMENT */);
|
---|
[83837] | 172 |
|
---|
| 173 | if (RTUniCpIsBMP(uc))
|
---|
| 174 | cwcRet += 1;
|
---|
| 175 | else if (uc >= 0x10000 && uc <= 0x0010ffff)
|
---|
| 176 | cwcRet += 2;
|
---|
| 177 | else
|
---|
[64340] | 178 | {
|
---|
[83837] | 179 | AssertMsgFailed(("uc=%#x\n", uc));
|
---|
| 180 | cwcRet += 1;
|
---|
[64340] | 181 | }
|
---|
| 182 | }
|
---|
[1] | 183 |
|
---|
[83837] | 184 | return cwcRet;
|
---|
[1] | 185 | }
|
---|
| 186 |
|
---|
| 187 |
|
---|
[83837] | 188 | RTDECL(ssize_t) RTUtf16Printf(PRTUTF16 pwszBuffer, size_t cwcBuffer, const char *pszFormat, ...)
|
---|
[1] | 189 | {
|
---|
[83837] | 190 | /* Explicitly inline RTStrPrintfV + RTStrPrintfExV here because this is a frequently use API. */
|
---|
| 191 | UTF16PRINTFOUTPUTARGS Args;
|
---|
| 192 | size_t cwcRet;
|
---|
[64340] | 193 | va_list args;
|
---|
[83837] | 194 | AssertMsg(cwcBuffer > 0, ("Excellent idea! Format a string with no space for the output!\n"));
|
---|
[7942] | 195 |
|
---|
[83837] | 196 | Args.pwszCur = pwszBuffer;
|
---|
| 197 | Args.cwcLeft = cwcBuffer;
|
---|
[64340] | 198 | Args.fOverflowed = false;
|
---|
[1] | 199 |
|
---|
[64340] | 200 | va_start(args, pszFormat);
|
---|
[83837] | 201 | cwcRet = RTStrFormatV(rtUtf16PrintfOutput, &Args, NULL, NULL, pszFormat, args);
|
---|
[64340] | 202 | va_end(args);
|
---|
| 203 |
|
---|
[83837] | 204 | return !Args.fOverflowed ? (ssize_t)cwcRet : -(ssize_t)cwcRet - 1;
|
---|
[1] | 205 | }
|
---|
[64340] | 206 | RT_EXPORT_SYMBOL(RTStrPrintf2);
|
---|
[1] | 207 |
|
---|
| 208 |
|
---|
[83837] | 209 | RTDECL(ssize_t) RTUtf16PrintfExV(PFNSTRFORMAT pfnFormat, void *pvArg, PRTUTF16 pwszBuffer, size_t cwcBuffer,
|
---|
| 210 | const char *pszFormat, va_list args)
|
---|
[1] | 211 | {
|
---|
[83837] | 212 | UTF16PRINTFOUTPUTARGS Args;
|
---|
| 213 | size_t cwcRet;
|
---|
| 214 | AssertMsg(cwcBuffer > 0, ("Excellent idea! Format a string with no space for the output!\n"));
|
---|
[64340] | 215 |
|
---|
[83837] | 216 | Args.pwszCur = pwszBuffer;
|
---|
| 217 | Args.cwcLeft = cwcBuffer;
|
---|
[64340] | 218 | Args.fOverflowed = false;
|
---|
[83837] | 219 | cwcRet = RTStrFormatV(rtUtf16PrintfOutput, &Args, pfnFormat, pvArg, pszFormat, args);
|
---|
| 220 | return !Args.fOverflowed ? (ssize_t)cwcRet : -(ssize_t)cwcRet - 1;
|
---|
[1] | 221 | }
|
---|
[83837] | 222 | RT_EXPORT_SYMBOL(RTUtf16PrintfExV);
|
---|
[1] | 223 |
|
---|
[21337] | 224 |
|
---|
[83837] | 225 | RTDECL(ssize_t) RTUtf16PrintfV(PRTUTF16 pwszBuffer, size_t cwcBuffer, const char *pszFormat, va_list args)
|
---|
[1] | 226 | {
|
---|
[83837] | 227 | return RTUtf16PrintfExV(NULL, NULL, pwszBuffer, cwcBuffer, pszFormat, args);
|
---|
[1] | 228 | }
|
---|
[83837] | 229 | RT_EXPORT_SYMBOL(RTUtf16Printf2V);
|
---|
[1] | 230 |
|
---|
[21337] | 231 |
|
---|
[83837] | 232 | RTDECL(ssize_t) RTUtf16PrintfEx(PFNSTRFORMAT pfnFormat, void *pvArg, PRTUTF16 pwszBuffer, size_t cwcBuffer,
|
---|
| 233 | const char *pszFormat, ...)
|
---|
[1] | 234 | {
|
---|
| 235 | va_list args;
|
---|
[64340] | 236 | ssize_t cbRet;
|
---|
[1] | 237 | va_start(args, pszFormat);
|
---|
[83837] | 238 | cbRet = RTUtf16PrintfExV(pfnFormat, pvArg, pwszBuffer, cwcBuffer, pszFormat, args);
|
---|
[1] | 239 | va_end(args);
|
---|
| 240 | return cbRet;
|
---|
| 241 | }
|
---|
[83837] | 242 | RT_EXPORT_SYMBOL(RTUtf16PrintfEx);
|
---|
[1] | 243 |
|
---|