cleanEntries.cc

Go to the documentation of this file.

/*
 SPDX-FileCopyrightText: © 2014-2015,2022 Siemens AG
 Author: Johannes Najjar
 
 SPDX-License-Identifier: GPL-2.0-only
*/
 
#include "cleanEntries.hpp"
#include <sstream>
#include <iterator>
using std::stringstream;
using std::ostream_iterator;
 
string cleanGeneral(string::const_iterator sBegin, string::const_iterator sEnd)
{
  stringstream ss;
  rx::regex_replace(ostream_iterator<char>(ss), sBegin, sEnd, rx::regex("[[:space:]\\x0-\\x1f]{2,}"), " ");
  string s = ss.str();
  string::size_type len = s.length();
  if (len > 1)
  {
    char cBegin = s[0];
    char cEnd = s[len - 1];
    if (cBegin == ' ' && cEnd == ' ')
      return s.substr(1, len - 2);
    else if (cBegin == ' ')
      return s.substr(1);
    else if (cEnd == ' ')
      return s.substr(0, len - 1);
  }
  // Only one character/space??? Should not be possible
  return s == " " ? "" : s;
}
 
string cleanSpdxStatement(string::const_iterator sBegin, string::const_iterator sEnd)
{
  stringstream ss;
  rx::regex_replace(ostream_iterator<char>(ss), sBegin, sEnd, rx::regex("spdx-filecopyrighttext:", rx::regex_constants::icase), " ");
  string s = ss.str();
  return cleanGeneral(s.begin(), s.end());
}
 
string cleanStatement(string::const_iterator sBegin, string::const_iterator sEnd)
{
  stringstream ss;
  rx::regex_replace(ostream_iterator<char>(ss), sBegin, sEnd, rx::regex("\n[[:space:][:punct:]]*"), " ");
  string s = ss.str();
  return cleanSpdxStatement(s.begin(), s.end());
}
 
string cleanNonPrint(string::const_iterator sBegin, string::const_iterator sEnd)
{
  string s(sBegin, sEnd);
  const unsigned char *in = reinterpret_cast<const unsigned char*>(s.c_str());
  int len = s.length();
 
  icu::UnicodeString out;
  for (int i = 0; i < len;)
  {
    UChar32 uniChar;
    size_t lastPos = i;
    U8_NEXT(in, i, len, uniChar);   // Get next UTF-8 char
    if (uniChar > 0)
    {
      out.append(uniChar);
    }
    else
    {
      i = lastPos;  // Rest pointer
      U16_NEXT(in, i, len, uniChar); // Try to get failed input as UTF-16
      if (U_IS_UNICODE_CHAR(uniChar) && uniChar > 0)
      {
        out.append(uniChar);
      }
    }
  }
  out.trim();
 
  string ret;
  out.toUTF8String(ret);
  return ret;
}
 
string cleanMatch(const string& sText, const match& m)
{
  string::const_iterator it = sText.begin();
  icu::UnicodeString unicodeStr = fo::recodeToUnicode(string(it + m.start,
    it + m.end));
  string utfCompatibleText;
 
  unicodeStr.toUTF8String(utfCompatibleText);
 
  if (m.type == "statement")
    return cleanStatement(utfCompatibleText.begin(), utfCompatibleText.end());
  else
    return cleanGeneral(utfCompatibleText.begin(), utfCompatibleText.end());
}