l10ntools/source/propmerge.cxx

   1 /* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
   2 /*
   3  * This file is part of the LibreOffice project.
   4  *
   5  * This Source Code Form is subject to the terms of the Mozilla Public
   6  * License, v. 2.0. If a copy of the MPL was not distributed with this
   7  * file, You can obtain one at http://mozilla.org/MPL/2.0/.
   8  */
   9
  10 #include <memory>
  11 #include <cstdlib>
  12 #include <cassert>
  13 #include <iostream>
  14 #include <fstream>
  15 #include <iomanip>
  16 #include <string_view>
  17
  18 #include <export.hxx>
  19 #include <common.hxx>
  20 #include <propmerge.hxx>
  21
  22 namespace
  23 {
  24     //Find ascii escaped unicode
  25     sal_Int32 lcl_IndexOfUnicode(
  26         const OString& rSource, const sal_Int32 nFrom = 0 )
  27     {
  28         const OString sHexDigits = "0123456789abcdefABCDEF";
  29         sal_Int32 nIndex = rSource.indexOf( "\\u", nFrom );
  30         if( nIndex == -1 )
  31         {
  32             return -1;
  33         }
  34         bool bIsUnicode = true;
  35         for( short nDist = 2; nDist <= 5; ++nDist )
  36         {
  37             if( sHexDigits.indexOf( rSource[nIndex + nDist] ) == -1 )
  38             {
  39                 bIsUnicode = false;
  40             }
  41         }
  42         return bIsUnicode ? nIndex : -1;
  43     }
  44
  45     //Convert ascii escaped unicode to utf-8
  46     OString lcl_ConvertToUTF8( const OString& rText )
  47     {
  48         OString sResult = rText;
  49         sal_Int32 nIndex = lcl_IndexOfUnicode( sResult );
  50         while( nIndex != -1 && nIndex < rText.getLength() )
  51         {
  52             const OString sHex = sResult.copy( nIndex + 2, 4 );
  53             const sal_Unicode cDec =
  54                 static_cast<sal_Unicode>( strtol( sHex.getStr(), nullptr, 16 ) );
  55             const OString sNewChar( &cDec, 1, RTL_TEXTENCODING_UTF8 );
  56             sResult = sResult.replaceAll( "\\u" + sHex, sNewChar );
  57             nIndex = lcl_IndexOfUnicode( sResult, nIndex );
  58         }
  59         return sResult;
  60     }
  61
  62     //Escape unicode characters
  63     void lcl_PrintJavaStyle( std::string_view rText, std::ofstream &rOfstream )
  64     {
  65         const OUString sTemp =
  66             OStringToOUString( rText, RTL_TEXTENCODING_UTF8 );
  67         for ( sal_Int32 nIndex = 0; nIndex < sTemp.getLength(); ++nIndex )
  68         {
  69             sal_Unicode cUniCode = sTemp[nIndex];
  70             if( cUniCode < 128 )
  71             {
  72                 rOfstream << static_cast<char>( cUniCode );
  73             }
  74             else
  75             {
  76                 rOfstream
  77                     << "\\u"
  78                     << std::setfill('0') << std::setw(2) << std::uppercase
  79                     << std::hex << (cUniCode >> 8)
  80                     << std::setfill('0') << std::setw(2) << std::uppercase
  81                     << std::hex << (cUniCode & 0xFF);
  82             }
  83         }
  84     }
  85 }
  86
  87 //Open source file and store its lines
  88 PropParser::PropParser(
  89     const OString& rInputFile, const OString& rLang,
  90     const bool bMergeMode )
  91     : m_vLines( std::vector<OString>() )
  92     , m_sSource( rInputFile )
  93     , m_sLang( rLang )
  94     , m_bIsInitialized( false )
  95 {
  96     std::ifstream aIfstream( m_sSource.getStr() );
  97     if( aIfstream.is_open() )
  98     {
  99         std::string s;
 100         std::getline( aIfstream, s );
 101         while( !aIfstream.eof() )
 102         {
 103             OString sLine( s.data(), s.length() );
 104             if( bMergeMode ||
 105                 ( !sLine.startsWith(" *") && !sLine.startsWith("/*") ) )
 106             {
 107                 m_vLines.push_back( sLine );
 108             }
 109             std::getline( aIfstream, s );
 110         }
 111     }
 112     else
 113     {
 114         std::cerr
 115             << "Propex error: Cannot open source file: "
 116             << m_sSource << std::endl;
 117         return;
 118     }
 119     m_bIsInitialized = true;
 120 }
 121
 122 PropParser::~PropParser()
 123 {
 124 }
 125
 126 //Extract strings form source file
 127 void PropParser::Extract( const OString& rPOFile )
 128 {
 129     assert( m_bIsInitialized );
 130     PoOfstream aPOStream( rPOFile, PoOfstream::APP );
 131     if( !aPOStream.isOpen() )
 132     {
 133         std::cerr
 134             << "Propex error: Cannot open pofile for extract: "
 135             << rPOFile << std::endl;
 136         return;
 137     }
 138
 139     for( size_t nIndex = 0; nIndex < m_vLines.size(); ++nIndex )
 140     {
 141         const OString sLine = m_vLines[nIndex];
 142         const sal_Int32 nEqualSign = sLine.indexOf('=');
 143         if( nEqualSign != -1 )
 144         {
 145             OString sID = sLine.copy( 0, nEqualSign ).trim();
 146             OString sText = lcl_ConvertToUTF8( sLine.copy( nEqualSign + 1 ).trim() );
 147
 148             common::writePoEntry(
 149                 "Propex", aPOStream, m_sSource, "property",
 150                 sID, OString(), OString(), sText);
 151         }
 152     }
 153
 154     aPOStream.close();
 155 }
 156
 157 //Merge strings to source file
 158 void PropParser::Merge( const OString &rMergeSrc, const OString &rDestinationFile )
 159 {
 160     assert( m_bIsInitialized );
 161     std::ofstream aDestination(
 162         rDestinationFile.getStr(), std::ios_base::out | std::ios_base::trunc );
 163     if( !aDestination.is_open() ) {
 164         std::cerr
 165             << "Propex error: Cannot open source file for merge: "
 166             << rDestinationFile << std::endl;
 167         return;
 168     }
 169
 170     std::unique_ptr<MergeDataFile> pMergeDataFile;
 171     if( m_sLang != "qtz" )
 172     {
 173         pMergeDataFile.reset( new MergeDataFile( rMergeSrc, m_sSource, false, false ) );
 174
 175         const std::vector<OString> vLanguages = pMergeDataFile->GetLanguages();
 176         if( !vLanguages.empty() && vLanguages[0] != m_sLang )
 177         {
 178             std::cerr
 179                 << ("Propex error: given language conflicts with language of"
 180                     " Mergedata file: ")
 181                 << m_sLang << " - "
 182                 << vLanguages[0] << std::endl;
 183             return;
 184         }
 185     }
 186
 187     for( size_t nIndex = 0; nIndex < m_vLines.size(); ++nIndex )
 188     {
 189         const OString sLine = m_vLines[nIndex];
 190         const sal_Int32 nEqualSign = sLine.indexOf('=');
 191         if( !sLine.startsWith(" *") && !sLine.startsWith("/*") &&
 192             nEqualSign != -1 )
 193         {
 194             const OString sID( sLine.copy( 0, sLine.indexOf('=') ).trim() );
 195             ResData  aResData( sID, m_sSource );
 196             aResData.sResTyp = "property";
 197             OString sNewText;
 198             if( m_sLang == "qtz" )
 199             {
 200                 const OString sOriginText = lcl_ConvertToUTF8(sLine.copy( nEqualSign + 1 ).trim());
 201                 sNewText = MergeEntrys::GetQTZText(aResData, sOriginText);
 202             }
 203             else if( pMergeDataFile )
 204             {
 205                 MergeEntrys* pEntrys = pMergeDataFile->GetMergeEntrys( &aResData );
 206                 if( pEntrys )
 207                 {
 208                     pEntrys->GetText( sNewText, m_sLang );
 209                 }
 210             }
 211             if( !sNewText.isEmpty() )
 212             {
 213                 aDestination << OString(sID + "=");
 214                 lcl_PrintJavaStyle( sNewText, aDestination );
 215                 aDestination << std::endl;
 216             }
 217             else
 218             {
 219                 aDestination << sLine << std::endl;
 220             }
 221         }
 222         else
 223         {
 224             aDestination << sLine << std::endl;
 225         }
 226     }
 227     aDestination.close();
 228 }
 229
 230 /* vim:set shiftwidth=4 softtabstop=4 expandtab: */