view release on metacpan or  search on metacpan

include/boost/wave/cpplexer/validate_universal_char.hpp  view on Meta::CPAN

/*=============================================================================
    Boost.Wave: A Standard compliant C++ preprocessor library

    Grammar for universal character validation (see C++ standard: Annex E)
    
    http://www.boost.org/

    Copyright (c) 2001-2005 Hartmut Kaiser. Distributed under the Boost
    Software License, Version 1.0. (See accompanying file
    LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)
=============================================================================*/
#if !defined(VALIDATE_UNIVERSAL_CHAR_HPP_55F1B811_CD76_4C72_8344_CBC69CF3B339_INCLUDED)
#define VALIDATE_UNIVERSAL_CHAR_HPP_55F1B811_CD76_4C72_8344_CBC69CF3B339_INCLUDED

#include <boost/assert.hpp>

#include <boost/wave/util/file_position.hpp>
#include <boost/wave/cpplexer/cpplexer_exceptions.hpp>

///////////////////////////////////////////////////////////////////////////////
namespace boost {
namespace wave {
namespace cpplexer {
namespace impl {

enum universal_char_type {
    universal_char_type_valid = 0,
    universal_char_type_invalid = 1,
    universal_char_type_base_charset = 2,
    universal_char_type_not_allowed_for_identifiers = 3
};

namespace {

    ///////////////////////////////////////////////////////////////////////////
    // 
    //  is_range is a helper function for the classification by brute force 
    //  below
    //
    ///////////////////////////////////////////////////////////////////////////
    inline bool 
    in_range(unsigned long ch, unsigned long l, unsigned long u)
    {
        return (l <= ch && ch <= u);
    }
}

///////////////////////////////////////////////////////////////////////////////
//
//  classify_universal_char
//
//      This function classifies an universal character value into 4 subranges:
//      universal_char_type_valid
//          the universal character value is valid for identifiers
//      universal_char_type_invalid
//          the universal character value is not valid for its usage inside
//          identifiers (see C++ Standard: 2.2.2 [lex.charset])
//      universal_char_type_base_charset
//          the universal character value designates a character from the base
//          character set
//      universal_char_type_not_allowed_for_identifiers
//          the universal character value is not allowed in an identifier
//
//  Implementation note:
//      This classification isn't implemented very effectively here. This
//      function should be rewritten with some range run matching algorithm.
//      
///////////////////////////////////////////////////////////////////////////////
inline universal_char_type 
classify_universal_char (unsigned long ch)
{
// test for invalid characters
    if (ch <= 0x0020 || in_range(ch, 0x007f, 0x009f))

include/boost/wave/cpplexer/validate_universal_char.hpp  view on Meta::CPAN

    }
}

///////////////////////////////////////////////////////////////////////////////
//
//  validate_literal
//
//      The validate_literal function tests a given string or character literal 
//      for its validity with regard to eventually contained universal 
//      characters. These should be in valid ranges (see the function 
//      classify_universal_char above).
//
//      If the string or character literal contains invalid or not allowed 
//      universal characters a corresponding lexing_exception is thrown.
//
///////////////////////////////////////////////////////////////////////////////
template <typename StringT>
inline void 
validate_literal (StringT const &name, int line, int column, 
    StringT const &file_name)
{
    using namespace std;    // some systems have strtoul in namespace std::
    
typename StringT::size_type pos = name.find_first_of('\\');

    while (StringT::npos != pos) {
    // the literal contains a backslash (may be universal char)
        if ('u' == name[pos+1] || 'U' == name[pos+1]) {
        StringT uchar_val(name.substr(pos+2, ('u' == name[pos+1]) ? 4 : 8));
        universal_char_type type = 
            classify_universal_char(strtoul(uchar_val.c_str(), 0, 16));  
            
            if (universal_char_type_valid != type && 
                universal_char_type_not_allowed_for_identifiers != type) 
            {
            // an invalid char was found, so throw an exception
            StringT error_uchar(name.substr(pos, ('u' == name[pos+1]) ? 6 : 10));
            
                if (universal_char_type_invalid == type) {
                    BOOST_WAVE_LEXER_THROW(lexing_exception, universal_char_invalid, 
                        error_uchar, line, column, file_name.c_str());
                }
                else {
                    BOOST_WAVE_LEXER_THROW(lexing_exception, universal_char_base_charset, 
                        error_uchar, line, column, file_name.c_str());
                }
            }
        }
                
    // find next universal char (if appropriate)
        pos = name.find_first_of('\\', pos+2);
    }
}

///////////////////////////////////////////////////////////////////////////////
}   // namespace impl           
}   // namespace cpplexer
}   // namespace wave
}   // namespace boost

#endif // !defined(VALIDATE_UNIVERSAL_CHAR_HPP_55F1B811_CD76_4C72_8344_CBC69CF3B339_INCLUDED)