sal/textenc/handleundefinedunicodetotextchar.cxx

   1 /* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
   2 /*
   3  * Version: MPL 1.1 / GPLv3+ / LGPLv3+
   4  *
   5  * The contents of this file are subject to the Mozilla Public License Version
   6  * 1.1 (the "License"); you may not use this file except in compliance with
   7  * the License or as specified alternatively below. You may obtain a copy of
   8  * the License at http://www.mozilla.org/MPL/
   9  *
  10  * Software distributed under the License is distributed on an "AS IS" basis,
  11  * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
  12  * for the specific language governing rights and limitations under the
  13  * License.
  14  *
  15  * Major Contributor(s):
  16  * [ Copyright (C) 2012 Red Hat, Inc., Stephan Bergmann <sbergman@redhat.com>
  17  *   (initial developer) ]
  18  *
  19  * All Rights Reserved.
  20  *
  21  * For minor contributions see the git repository.
  22  *
  23  * Alternatively, the contents of this file may be used under the terms of
  24  * either the GNU General Public License Version 3 or later (the "GPLv3+"), or
  25  * the GNU Lesser General Public License Version 3 or later (the "LGPLv3+"),
  26  * in which case the provisions of the GPLv3+ or the LGPLv3+ are applicable
  27  * instead of those above.
  28  */
  29
  30 #include "sal/config.h"
  31
  32 #include "rtl/textcvt.h"
  33 #include "sal/types.h"
  34
  35 #include "handleundefinedunicodetotextchar.hxx"
  36 #include "tenchelp.hxx"
  37
  38 namespace {
  39
  40 bool ImplIsUnicodeIgnoreChar(sal_Unicode c, sal_uInt32 nFlags)
  41 {
  42     return
  43         ((nFlags & RTL_UNICODETOTEXT_FLAGS_NONSPACING_IGNORE) != 0
  44          && ImplIsZeroWidth(c))
  45         || ((nFlags & RTL_UNICODETOTEXT_FLAGS_CONTROL_IGNORE) != 0
  46             && ImplIsControlOrFormat(c))
  47         || ((nFlags & RTL_UNICODETOTEXT_FLAGS_PRIVATE_IGNORE) != 0
  48             && ImplIsPrivateUse(c));
  49 }
  50
  51 bool ImplGetUndefinedAsciiMultiByte(sal_uInt32 nFlags,
  52                                         char * pBuf,
  53                                         sal_Size nMaxLen)
  54 {
  55     if (nMaxLen == 0)
  56         return false;
  57     switch (nFlags & RTL_UNICODETOTEXT_FLAGS_UNDEFINED_MASK)
  58     {
  59     case RTL_UNICODETOTEXT_FLAGS_UNDEFINED_0:
  60         *pBuf = 0x00;
  61         break;
  62
  63     case RTL_UNICODETOTEXT_FLAGS_UNDEFINED_QUESTIONMARK:
  64     default: /* RTL_UNICODETOTEXT_FLAGS_UNDEFINED_DEFAULT */
  65         *pBuf = 0x3F;
  66         break;
  67
  68     case RTL_UNICODETOTEXT_FLAGS_UNDEFINED_UNDERLINE:
  69         *pBuf = 0x5F;
  70         break;
  71     }
  72     return true;
  73 }
  74
  75 bool ImplGetInvalidAsciiMultiByte(sal_uInt32 nFlags,
  76                                       char * pBuf,
  77                                       sal_Size nMaxLen)
  78 {
  79     if (nMaxLen == 0)
  80         return false;
  81     switch (nFlags & RTL_UNICODETOTEXT_FLAGS_UNDEFINED_MASK)
  82     {
  83     case RTL_UNICODETOTEXT_FLAGS_INVALID_0:
  84         *pBuf = 0x00;
  85         break;
  86
  87     case RTL_UNICODETOTEXT_FLAGS_INVALID_QUESTIONMARK:
  88     default: /* RTL_UNICODETOTEXT_FLAGS_INVALID_DEFAULT */
  89         *pBuf = 0x3F;
  90         break;
  91
  92     case RTL_UNICODETOTEXT_FLAGS_INVALID_UNDERLINE:
  93         *pBuf = 0x5F;
  94         break;
  95     }
  96     return true;
  97 }
  98
  99 }
 100
 101 bool sal::detail::textenc::handleUndefinedUnicodeToTextChar(
 102     sal_Unicode const ** ppSrcBuf, sal_Unicode const * pEndSrcBuf,
 103     char ** ppDestBuf, char const * pEndDestBuf, sal_uInt32 nFlags,
 104     sal_uInt32 * pInfo)
 105 {
 106     sal_Unicode c = **ppSrcBuf;
 107
 108     /* Should the private character map to one byte */
 109     if ( (c >= RTL_TEXTCVT_BYTE_PRIVATE_START) && (c <= RTL_TEXTCVT_BYTE_PRIVATE_END) )
 110     {
 111         if ( nFlags & RTL_UNICODETOTEXT_FLAGS_PRIVATE_MAPTO0 )
 112         {
 113             **ppDestBuf = (char)(sal_uChar)(c-RTL_TEXTCVT_BYTE_PRIVATE_START);
 114             (*ppDestBuf)++;
 115             (*ppSrcBuf)++;
 116             return true;
 117         }
 118     }
 119
 120     /* Should this character ignored (Private, Non Spacing, Control) */
 121     if ( ImplIsUnicodeIgnoreChar( c, nFlags ) )
 122     {
 123         (*ppSrcBuf)++;
 124         return true;
 125     }
 126
 127     /* Surrogates Characters should result in */
 128     /* one replacement character */
 129     if (ImplIsHighSurrogate(c))
 130     {
 131         if ( *ppSrcBuf == pEndSrcBuf )
 132         {
 133             *pInfo |= RTL_UNICODETOTEXT_INFO_ERROR | RTL_UNICODETOTEXT_INFO_SRCBUFFERTOSMALL;
 134             return false;
 135         }
 136
 137         c = *((*ppSrcBuf)+1);
 138         if (ImplIsLowSurrogate(c))
 139             (*ppSrcBuf)++;
 140         else
 141         {
 142             *pInfo |= RTL_UNICODETOTEXT_INFO_INVALID;
 143             if ( (nFlags & RTL_UNICODETOTEXT_FLAGS_INVALID_MASK) == RTL_UNICODETOTEXT_FLAGS_INVALID_ERROR )
 144             {
 145                 *pInfo |= RTL_UNICODETOTEXT_INFO_ERROR;
 146                 return false;
 147             }
 148             else if ( (nFlags & RTL_UNICODETOTEXT_FLAGS_INVALID_MASK) == RTL_UNICODETOTEXT_FLAGS_INVALID_IGNORE )
 149             {
 150                 (*ppSrcBuf)++;
 151                 return true;
 152             }
 153             else if (ImplGetInvalidAsciiMultiByte(nFlags,
 154                                                   *ppDestBuf,
 155                                                   pEndDestBuf - *ppDestBuf))
 156             {
 157                 ++*ppSrcBuf;
 158                 ++*ppDestBuf;
 159                 return true;
 160             }
 161             else
 162             {
 163                 *pInfo |= RTL_UNICODETOTEXT_INFO_ERROR
 164                               | RTL_UNICODETOTEXT_INFO_DESTBUFFERTOSMALL;
 165                 return false;
 166             }
 167         }
 168     }
 169
 170     *pInfo |= RTL_UNICODETOTEXT_INFO_UNDEFINED;
 171     if ( (nFlags & RTL_UNICODETOTEXT_FLAGS_UNDEFINED_MASK) == RTL_UNICODETOTEXT_FLAGS_UNDEFINED_ERROR )
 172     {
 173         *pInfo |= RTL_UNICODETOTEXT_INFO_ERROR;
 174         return false;
 175     }
 176     else if ( (nFlags & RTL_UNICODETOTEXT_FLAGS_UNDEFINED_MASK) == RTL_UNICODETOTEXT_FLAGS_UNDEFINED_IGNORE )
 177         (*ppSrcBuf)++;
 178     else if (ImplGetUndefinedAsciiMultiByte(nFlags,
 179                                             *ppDestBuf,
 180                                             pEndDestBuf - *ppDestBuf))
 181     {
 182         ++*ppSrcBuf;
 183         ++*ppDestBuf;
 184     }
 185     else
 186     {
 187         *pInfo |= RTL_UNICODETOTEXT_INFO_ERROR
 188                       | RTL_UNICODETOTEXT_INFO_DESTBUFFERTOSMALL;
 189         return false;
 190     }
 191
 192     return true;
 193 }