/* massXpert - the true massist's program.
   --------------------------------------
   Copyright(C) 2006,2007 Filippo Rusconi

   http://www.massxpert.org/massXpert

   This file is part of the massXpert project.

   The massxpert project is the successor to the "GNU polyxmass"
   project that is an official GNU project package(see
   www.gnu.org). The massXpert project is not endorsed by the GNU
   project, although it is released ---in its entirety--- under the
   GNU General Public License. A huge part of the code in massXpert
   is actually a C++ rewrite of code in GNU polyxmass. As such
   massXpert was started at the Centre National de la Recherche
   Scientifique(FRANCE), that granted me the formal authorization to
   publish it under this Free Software License.

   This software is free software; you can redistribute it and/or
   modify it under the terms of the GNU  General Public
   License version 3, as published by the Free Software Foundation.
   

   This software is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
   General Public License for more details.

   You should have received a copy of the GNU General Public License
   along with this software; if not, write to the

   Free Software Foundation, Inc.,

   51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA.
*/


/////////////////////// Qt includes
#include <QObject>


/////////////////////// Local includes
#include "monomerDictionary.hpp"
#include "sequence.hpp"


namespace massXpert
{

  MonomerDictionary::MonomerDictionary(QString filePath,
				       const QStringList &inputChainStringList,
				       int inputCodeLength,
				       int outputCodeLength)
    : m_filePath(filePath), m_inputChainStringList(inputChainStringList), 
      m_inputCodeLength(inputCodeLength), m_outputCodeLength(outputCodeLength)
  {
  }

  
  //! Destroys the cleavage rule.
  MonomerDictionary::~MonomerDictionary()
  {
  }
  
  void 
  MonomerDictionary::setFilePath(QString &filePath)
  {
    m_filePath = filePath;
  }
  
  void 
  MonomerDictionary::setInputChainStringList(const QStringList &inputChainStringList)
  {
    m_inputChainStringList = inputChainStringList;
  }
  
  void 
  MonomerDictionary::setInputCodeLength(int codeLength)
  {
    m_inputCodeLength = codeLength;
  }
  
  void 
  MonomerDictionary::setOutputCodeLength(int codeLength)
  {
    m_outputCodeLength = codeLength;
  }
  
  bool 
  MonomerDictionary::isLineProperSectionDivider(const QString &line)
  {
    // Section dividers in the monomer dictionary file format are
    // lines containing the following syntax: X>Y, that is for example
    // 3>1. This means that the following conversion rules (like
    // ILE>I) should convert 3-letter codes into 1-letter codes.

    // However, this line should only be considered proper if X is
    // actually the value of m_inputCodeLength and Y the value of
    // m_outputCodeLength.

//     qDebug() << __FILE__ << __LINE__
// 	     << "Checking if line is proper section divider :" << line;
    
    if (line.contains(QRegExp ("[0-9]+>[0-9]+")))
      {
	// We are opening a new section, get the input/output code
	// lengths and if they math what we expect, then set the
	// current stream position and call the section parser.

	int greaterThanIndex = line.indexOf('>');
	    
	QString codeLengthString = line.left(greaterThanIndex);

// 	qDebug() << __FILE__ << __LINE__
// 		 << "Left codeLengthString:" << codeLengthString
// 		 << "m_inputCodeLength:" << m_inputCodeLength;

	bool ok = false;
	int codeLength = codeLengthString.toInt(&ok, 10);
	    
	if (!codeLength && !ok)
	  {
	    qDebug() << __FILE__ << __LINE__
		     << "Monomer dictionary"
		     << "Failed to parse file " << m_filePath
		     << "at line " << line;
		
	    return false;
	  }
	    
	if (codeLength != m_inputCodeLength)
	  {
	    return false;
	  }
	    
	codeLengthString = line.mid(greaterThanIndex + 1, -1);

// 	qDebug() << __FILE__ << __LINE__
// 		 << "Right codeLengthString:" << codeLengthString
// 		 << "m_outputCodeLength:" << m_outputCodeLength;
	
	ok = false;
	codeLength = codeLengthString.toInt(&ok, 10);
	    
	if (!codeLength && !ok)
	  {
	    qDebug() << __FILE__ << __LINE__
		     << "Monomer dictionary"
		     << "Failed to parse file " << m_filePath
		     << "at line " << line;
		
	    return false;
	  }

	if (codeLength != m_outputCodeLength)
	  {
	    return false;
	  }
	    
	// At this point, it seems we are in the proper
	// section. 

	return true;
      }

    // If we are here, that means that the section is not for us.

//     qDebug() << __FILE__ << __LINE__
// 	     << "Line is no proper section divider.";
    
    return false;
  }
  

  void 
  MonomerDictionary::skipSection(QTextStream *stream)
  {
    // We have entered a section, all we have to do is go through it
    // and return when we have found either the end of the stream or
    // the {END} marker.

    qint64 lineLength = 1024;
    QString line;
    
    while (!stream->atEnd())
      {
	line = stream->readLine(lineLength);
	
	if (!line.contains("{END}"))
	  {
	    line = stream->readLine(lineLength);
	  }
	else
	  return;
      }
  }
  

  int 
  MonomerDictionary::parseSection(QTextStream *stream)
  {
    Q_ASSERT(stream);
    
    qint64 lineLength = 1024;
    QString line;

    // Iterate in the file using the stream and for each line create
    // an item to insert into the dictionary hash.

    while (!stream->atEnd())
      {
	line = stream->readLine(lineLength);
            
	// We might encounter the end of the section, that is a line
	// having {END} as its sole content.
	
	if (line.contains("{END}"))
	  break;
	
	QStringList stringList = line.split('>');

	QString inputCode = stringList.first ();
	QString outputCode = stringList.last ();
	
	// Check that the monomer codes have the proper length.
	
	if (inputCode.length() != m_inputCodeLength ||
	    outputCode.length() != m_outputCodeLength)
	  {
	    qDebug() << __FILE__ << __LINE__
		     << QObject::tr("Monomer dictionary:")
		     << QObject::tr("Failed to load dictionary.")
		     << QObject::tr("Monomer code lengths do not match:")
		     << QObject::tr("inputCode:") << inputCode
		     << QObject::tr("outputCode:") << outputCode;
	    
	
	    // We have to empty the hash
	    m_dictionaryHash.clear();
	    
	    break;
	  }
	
	m_dictionaryHash.insert (inputCode, outputCode);

	// 	qDebug() << __FILE__ << __LINE__
	// 		 << stringList.first () << stringList.last ();	
      }
    
    // At this point the parsing is finished, either because we
    // encountered the {END} section-ending delimiter, or because we
    // reached the en of file.

    int hashSize = m_dictionaryHash.size();
    
    if (hashSize)
      m_dictionaryLoaded = true;
    else
      {
	qDebug() << __FILE__ << __LINE__
		 << QObject::tr ("Monomer dictionary:")
		 << QObject::tr ("Failed to load dictionary.");
	
	m_dictionaryLoaded = false;
      }
    
    return hashSize;
  }
  
  
  bool
  MonomerDictionary::loadDictionary ()
  {
    // Load the file and for each line deconstruct the item into two
    // QString objects that are used to make a QHash entry in
    // QHash<QString, QString> m_dictionaryHash.
    bool success = true;
    qint64 lineLength = 1024;
    QString line;
  
    QFile file(m_filePath);

    if (!file.open(QIODevice::ReadOnly)) 
      {
	
	m_dictionaryLoaded = false;
	
	qDebug() << __FILE__ << __LINE__
		 << "Monomer dictionary:"
		 << "Failed to open file" << m_filePath
		 << "for writing.";
	
	return false;
      }

    if (m_inputCodeLength < 1 || m_outputCodeLength < 1)
      {
	qDebug() << __FILE__ << __LINE__
		 << "Monomer dictionary:"
		 << "Failed to parse file " << m_filePath
		 << "Please, set the m_inputCodeLength and "
	  "m_ouputCodeLength variables first.";
	
	return false;
      }
    
    QTextStream *stream = new QTextStream(&file);
    stream->setCodec("UTF-8");

    while (!stream->atEnd())
      {
	line = stream->readLine(lineLength);

	// 	  qDebug() << __FILE__ << __LINE__
	// 		   << "line: " << line;
 
	// Remove spaces from start and end of line.
	line = line.simplified();
	
	if (line.startsWith ('#') || line.isEmpty())
	  {
	    line = stream->readLine(lineLength);
	    continue;
	  }

	// There might be any number of sections in the file, all
	// delimited with a X>Y directive, indicating how many
	// characters are allowed for the input code and for the
	// output code.

	if (!isLineProperSectionDivider(line))
	  {
// 	    qDebug() << __FILE__ << __LINE__
// 		     << "skipping line:" << line;

	    line = stream->readLine(lineLength);
	    continue;
	  }
	else
	  {
// 	    qDebug() << __FILE__ << __LINE__
// 		     << "parsing section: " << line;

	    if (parseSection(stream) < 1)
	      {
		qDebug() << __FILE__ << __LINE__
			 << "Monomer dictionary:"
			 << "Failed to parse file " << m_filePath;
		  
		success = false;
		break;
	      }
	    else
	      {
		// We successfully parsed the section. Our work is done.
		
		success = true;
		break;
	      }
	      
	  }
      }
      
    delete stream;
      
    return success;
  }


  QStringList *
  MonomerDictionary::translate(const QStringList &chainStringList)
  {
    // The string in sequence is a space-separated list of monomer
    // codes in the original monomer code format. We have to translate
    // that to the proper monomer code format using the hash in this
    // dictionary.

    QStringList *outputChainStringList = new QStringList();
    
    if (!chainStringList.isEmpty())
      m_inputChainStringList = chainStringList;
    
    // If there is nothing to do return an empty string list so that
    // caller knows nothing is actually wrong, only there is no
    // sequence to translate.
    if (m_inputChainStringList.isEmpty())
      return outputChainStringList;
    
    // Iterate in each chain string of the list and perform the
    // translation.

    for (int iter = 0; iter < m_inputChainStringList.size(); ++iter)
      {
	QString iterString = chainStringList.at(iter);
	
// 	qDebug() << __FILE__ << __LINE__
// 		 << "translating sequence:" << iterString;
		
	QStringList codeList =
	  iterString.split(QRegExp("\\s+"), QString::SkipEmptyParts);
	
// 	qDebug() << __FILE__ << __LINE__ 
// 		 << "codeList:" << codeList;

// 	qDebug() << __FILE__ << __LINE__ 
// 		 << "hash:"
// 		 << m_dictionaryHash;
	
	for (int jter = 0; jter < codeList.size(); ++jter)
	  {
	    QString code = codeList.at(jter);

	    QHash<QString, QString>::const_iterator hashIter = 
	      m_dictionaryHash.find(code);
	
	    if (hashIter != m_dictionaryHash.end())
	      codeList.replace(jter, hashIter.value());
	    else
	      {
		// Delete the string list, set the pointer to 0 and
		// return that pointer so that caller knows something
		// has gone wrong.
		
		qDebug() << __FILE__ << __LINE__
			 << "Monomer dictionary:"
			 << "Failed to convert monomer code "
			 << code;
	    
		outputChainStringList->clear();

		delete outputChainStringList;
		outputChainStringList = 0;

		return outputChainStringList;
	      }
	  }
 
	// At this point the sequence codes have been translated. Join all
	// the item of the codeList into one single string.

	outputChainStringList->append(codeList.join(QString("")));
      }
    
    // End of
    // for (int iter = 0; iter < chainStringList.size(); ++iter)
    
    // If no translation could be performed, return a n
    
    if (!outputChainStringList->size())
      {
	outputChainStringList->clear();

	delete outputChainStringList;
	outputChainStringList = 0;
      }
    
    return outputChainStringList;
  }
  
  
} // namespace massXpert