convert_string_literal.cpp

/*******************************************************************\

Module: C/C++ Language Conversion

Author: Daniel Kroening, kroening@kroening.com

\*******************************************************************/

/// \file
/// C/C++ Language Conversion

#include "convert_string_literal.h"

#include <cassert>

#include <util/arith_tools.h>
#include <util/c_types.h>
#include <util/unicode.h>
#include <util/string_constant.h>

#include "unescape_string.h"

std::basic_string<unsigned int> convert_one_string_literal(
  const std::string &src)
{
  assert(src.size()>=2);

  if(src[0]=='u' && src[1]=='8')
  {
    assert(src[src.size()-1]=='"');
    assert(src[2]=='"');

    std::basic_string<unsigned int> value=
      unescape_wide_string(std::string(src, 3, src.size()-4));

    // turn into utf-8
    const std::string utf8_value = utf32_native_endian_to_utf8(value);

    // pad into wide string
    value.resize(utf8_value.size());
    for(std::size_t i=0; i<utf8_value.size(); i++)
      value[i]=utf8_value[i];

    return value;
  }
  else if(src[0]=='L' || src[0]=='u' || src[0]=='U')
  {
    assert(src[src.size()-1]=='"');
    assert(src[1]=='"');

    return unescape_wide_string(std::string(src, 2, src.size()-3));
  }
  else
  {
    assert(src[0]=='"');
    assert(src[src.size()-1]=='"');

    std::string char_value=
      unescape_string(std::string(src, 1, src.size()-2));

    // pad into wide string
    std::basic_string<unsigned int> value;
    value.resize(char_value.size());
    for(std::size_t i=0; i<char_value.size(); i++)
      value[i]=char_value[i];

    return value;
  }
}

exprt convert_string_literal(const std::string &src)
{
  // note that 'src' could be a concatenation of string literals,
  // e.g., something like "asd" "xyz".
  // GCC allows "asd" L"xyz"!

  std::basic_string<unsigned int> value;

  char wide=0;

  for(std::size_t i=0; i<src.size(); i++)
  {
    char ch=src[i];

    // skip whitespace/newline
    if(ch!='L' && ch!='u' && ch!='U' && ch!='"')
      continue;

    if(ch=='L')
      wide=ch;
    if((ch=='u' || ch=='U') && i+1<src.size() && src[i+1]=='"')
      wide=ch;

    // find start of sequence
    std::size_t j=src.find('"', i);
    if(j==std::string::npos)
      throw "invalid string constant `"+src+"'";

    // find end of sequence, considering escaping
    for(++j; j<src.size() && src[j]!='"'; ++j)
      if(src[j]=='\\') // skip next character
        ++j;

    assert(j<=src.size());
    if(j==src.size())
      throw "non-terminated string constant `"+src+"'";

    std::string tmp_src=std::string(src, i, j-i+1);
    std::basic_string<unsigned int> tmp_value=
      convert_one_string_literal(tmp_src);
    value.append(tmp_value);
    i=j;
  }

  if(wide!=0)
  {
    // add implicit trailing zero
    value.push_back(0);

    // L is wchar_t, u is char16_t, U is char32_t.
    typet subtype;

    switch(wide)
    {
    case 'L': subtype=wchar_t_type(); break;
    case 'u': subtype=char16_t_type(); break;
    case 'U': subtype=char32_t_type(); break;
    default: assert(false);
    }

    exprt result=exprt(ID_array);
    result.set(ID_C_string_constant, true);
    result.type()=typet(ID_array);
    result.type().subtype()=subtype;
    result.type().set(ID_size, from_integer(value.size(), size_type()));

    result.operands().resize(value.size());
    for(std::size_t i=0; i<value.size(); i++)
      result.operands()[i]=from_integer(value[i], subtype);

    return result;
  }
  else
  {
    std::string char_value;

    char_value.resize(value.size());

    for(std::size_t i=0; i<value.size(); i++)
    {
      // Loss of data here if value[i]>255.
      // gcc issues a warning in this case.
      char_value[i]=value[i];
    }

    return string_constantt(char_value);
  }
}