/*****************************************************************************
#   Copyright (C) 1994-2008 by David Gordon.
#   All rights reserved.                           
#                                                                           
#   This software is part of a beta-test version of the Consed/Autofinish
#   package.  It should not be redistributed or
#   used for any commercial purpose, including commercially funded
#   sequencing, without written permission from the author and the
#   University of Washington.
#   
#   This software is provided ``AS IS'' and any express or implied
#   warranties, including, but not limited to, the implied warranties of
#   merchantability and fitness for a particular purpose, are disclaimed.
#   In no event shall the authors or the University of Washington be
#   liable for any direct, indirect, incidental, special, exemplary, or
#   consequential damages (including, but not limited to, procurement of
#   substitute goods or services; loss of use, data, or profits; or
#   business interruption) however caused and on any theory of liability,
#   whether in contract, strict liability, or tort (including negligence
#   or otherwise) arising in any way out of the use of this software, even
#   if advised of the possibility of such damage.
#
#   Building Consed from source is error prone and not simple which is
#   why I provide executables.  Due to time limitations I cannot
#   provide any assistance in building Consed.  Even if you do not
#   modify the source, you may introduce errors due to using a
#   different version of the compiler, a different version of motif,
#   different versions of other libraries than I used, etc.  For this
#   reason, if you discover Consed bugs, I can only offer help with
#   those bugs if you first reproduce those bugs with an executable
#   provided by me--not an executable you have built.
# 
#   Modifying Consed is also difficult.  Although Consed is modular,
#   some modules are used by many other modules.  Thus making a change
#   in one place can have unforeseen effects on many other features.
#   It may takes months for you to notice these other side-effects
#   which may not seen connected at all.  It is not feasable for me to
#   provide help with modifying Consed sources because of the
#   potentially huge amount of time involved.
#
#*****************************************************************************/
#include    "readPHDOfSinglet.h"
#include    "readPHDAgainForHighQualitySegment.h"
#include    "filename.h"
#include    "singletInfo.h"
#include    <cstdio>
using namespace std;
#include    "mbt_exception.h"
#include    <ctype.h>
#include    "consed.h"
#include    "nGetReadTypeFromReadName.h"


// singlets are not read for the bases or qualities.  But we need the
// experiment id for Autofinish.  The chemistry, read type, and high
// quality segment are also read.  bIsThisUniversalPrimerReadInSinglets is 
// probably used so that Autofinish doesn't repeatedly suggest the same
// universal primer read.  The high quality segment is used by the 
// evaluation stage of Autofinish for determining how well Autofinish's
// suggestions worked.



#define PARSE_PANIC( szMessage ) \
{  ostringstream ost; \
   ost << "phd file error detected from source file " \
     << __FILE__ << " at " << __LINE__ <<endl; \
   ost << "Error in file " << filPHDFullPath << " at line " <<  \
       nLine << "  Line:\n" \
      << szLine << "\n" << szMessage << endl \
        << ends; \
   InputDataError ide(ost.str().c_str()); \
   throw ide; }
      

#define PARSE_PANIC_1_ARG( szMessage, arg1 ) \
{ \
     char szTemp[500]; \
     sprintf( szTemp, szMessage, arg1 ); \
     PARSE_PANIC( szTemp ); \
}

#define     FGETS_OR_PARSE_PANIC( szErrorMessage ) \
if ( fgets( szLine, nMaxLineSize, pFil ) == NULL ) \
{  ostringstream ost; \
   ost << "phd file error detected from source file " \
     << __FILE__ << " at " << __LINE__ <<endl; \
   ost << "Error in file " << filPHDFullPath << " at line " <<  \
       nLine << "  Line:\n" \
      << szLine << "\n" << szErrorMessage << endl \
        << ends; \
   InputDataError ide(ost.str().c_str()); \
   throw ide; } \
++nLine;




const int nMaxLineSize = 300;
static char  szLine[ nMaxLineSize + 1];
static int nLine;


void readPHDOfSinglet( bool& bFoundExpID,
                          singletInfo* pSingletInfo ) {

   FileName filPHDFullPath = ConsEd::pGetConsEd()->filGetPHDDir() + "/" +
     pSingletInfo->filPHD_;

   FILE* pFil = fopen( (char*) filPHDFullPath.data(), "r" );

   if ( pFil == NULL ) {
      printf( "could not open singlet phd file %s\n",
              (char*) filPHDFullPath.data() );
      bFoundExpID = false;
      return;
   }

   nLine = 0;


   // will be set by read name if not found in phd file
   pSingletInfo->nReadType_ = 0;


   pSingletInfo->soChemistry_ = "";


   // look for BEGIN_COMMENT

   bool bLookingForBEGIN_COMMENT = true;
   do {
      if ( fgets( szLine, nMaxLineSize, pFil ) == NULL ) {
         PARSE_PANIC( "premature end of file while looking for BEGIN_COMMENT" );
      }
      ++nLine;

      if ( szLine[0] == 'B' &&
           szLine[1] == 'E' &&
           szLine[2] == 'G' &&
           szLine[3] == 'I' &&
           szLine[4] == 'N' &&
           szLine[5] == '_' &&
           szLine[6] == 'C' &&
           szLine[7] == 'O' &&
           szLine[8] == 'M' &&
           szLine[9] == 'M' &&
           szLine[10] == 'E' &&
           szLine[11] == 'N' &&
           szLine[12] == 'T' ) {

         bLookingForBEGIN_COMMENT = false;
      }
   } while( bLookingForBEGIN_COMMENT );

   // look for CHEM:

   bool bContinueLookingForChemistry = true;

   do {
      if ( fgets( szLine, nMaxLineSize, pFil ) == NULL ) {
         PARSE_PANIC( "premature end of file while looking for END_COMMENT" );
      }
      ++nLine;

      if ( szLine[0] == 'E' &&
           szLine[1] == 'N' &&
           szLine[2] == 'D' &&
           szLine[3] == '_' &&
           szLine[4] == 'C' &&
           szLine[5] == 'O' &&
           szLine[6] == 'M' &&
           szLine[7] == 'M' &&
           szLine[8] == 'E' &&
           szLine[9] == 'N' &&
           szLine[10] == 'T' ) {

         bContinueLookingForChemistry = false;
      }
      else {
         
         if ( szLine[0] == 'C' &&
              szLine[1] == 'H' &&
              szLine[2] == 'E' &&
              szLine[3] == 'M' &&
              szLine[4] == ':' ) {

            pSingletInfo->soChemistry_ = "";
            
            // jump over whitespace

            int n = 5;
            while( isspace( szLine[n] ) )
               ++n;

            // so szLine[n] is the first character of the chemistry
            
            while( szLine[n] != '\n' &&
                   szLine[n] != '\0' &&
                   szLine[n] != ' ' ) {
               pSingletInfo->soChemistry_.append( szLine[n] );
               ++n;
            }

            bContinueLookingForChemistry = false;
         }
      }
   } while( bContinueLookingForChemistry );

   // look for END_DNA

   bool bContinueLookingForEndDNA = true;
   do {
      if ( fgets( szLine, nMaxLineSize, pFil ) == NULL ) {
         PARSE_PANIC( "premature end of file while looking for END_DNA" );
      }
      ++nLine;


      if ( szLine[0] == 'E' &&
           szLine[1] == 'N' &&
           szLine[2] == 'D' &&
           szLine[3] == '_' &&
           szLine[4] == 'D' &&
           szLine[5] == 'N' &&
           szLine[6] == 'A' )
         bContinueLookingForEndDNA = false;

   } while ( bContinueLookingForEndDNA );

   // try to find expid WR item

   while(1) {
      if ( fgets( szLine, nMaxLineSize, pFil ) == NULL ) {
         bFoundExpID = false;
         break;
      }

      ++nLine;

      if ( szLine[0] == 'W' &&
           szLine[1] == 'R' &&
           szLine[2] == '{' &&
           isspace( szLine[3] ) ) {
      
         FGETS_OR_PARSE_PANIC( "premature end of file while in WR item" );
      
         if ( szLine[0] == 'e' &&
              szLine[1] == 'x' &&
              szLine[2] == 'p' &&
              szLine[3] == 'i' &&
              szLine[4] == 'd' &&
              isspace( szLine[5] ) ) {
      
            // found expid whole read item
            // get the expid
      
            FGETS_OR_PARSE_PANIC( "premature end of file while in expid WR item" );
      
            int n = 0;
            int nExpID = 0;
            while( szLine[n] != '\n' ) {
               if ( '0' <= szLine[n] && szLine[n] <= '9' ) {
                  nExpID = 10*nExpID + ( szLine[n] - '0' );
                  ++n;
               }
               else {
                  PARSE_PANIC_1_ARG( "line should have been just a number but found char %c\n", 
                                     szLine[n] );
               }
            }
      
            bFoundExpID = true;
            pSingletInfo->nExpID_ = nExpID;
         } // if ( szLine[0] == 'e' && ...
         else if ( szLine[0] == 't' &&
                   szLine[1] == 'e' &&
                   szLine[2] == 'm' &&
                   szLine[3] == 'p' &&
                   szLine[4] == 'l' &&
                   szLine[5] == 'a' &&
                   szLine[6] == 't' &&
                   szLine[7] == 'e' &&
                   isspace( szLine[8] ) ) {
      
            // found the template WR item
            // now get the template name
            
            bool bEndOfWRItem = false;

            do {
            
               FGETS_OR_PARSE_PANIC( "premature end of file while in template WR item" );
      
               RWCString soDataLine( szLine );


               if ( soDataLine.length() >= 2 &&
                    soDataLine[0] == '}' &&
                    isspace( soDataLine[1] ) ) 
                  bEndOfWRItem = true;
               else if ( soDataLine.length() >= 5 &&
                         soDataLine( 0, 5 ) == "name:" ) {
                  if ( soDataLine.length() < 6 ) {
                     PARSE_PANIC( "In template WR item which should be of form name: (template name) but (template name) is missing" );
                  }
      
                  soDataLine = soDataLine( 5, soDataLine.length() - 5 );
      
                  // strip off leading and trailing whitespace
      
                  soDataLine = soDataLine.stripWhitespace( RWCString::BOTH );
                  pSingletInfo->soTemplate_ = soDataLine;
               }
               else if ( soDataLine.length() >= 5 &&
                         soDataLine( 0, 5 ) == "type:" ) {
                  if ( soDataLine.length() < 6 )
                     PARSE_PANIC( "In template WR item, type line which should be of the form type: (bac, cos, puc, pbc, pcr, etc.)" );
                  soDataLine = soDataLine( 5, soDataLine.length() - 5 );

                  soDataLine = soDataLine.stripWhitespace( RWCString::BOTH );
                  pSingletInfo->soTemplateType_ =  soDataLine;
               }
               // size isn't yet implemented

            } while( !bEndOfWRItem );
         } // if ( szLine[0] == 't' ... template
         else if ( szLine[0] == 'p' &&
                   szLine[1] == 'r' &&
                   szLine[2] == 'i' &&
                   szLine[3] == 'm' &&
                   szLine[4] == 'e' &&
                   szLine[5] == 'r' &&
                   isspace( szLine[6] ) ) {
            
            // found the primer WR item
            // now get the read type
            
            bool bEndOfWRItem = false;
            do {
               FGETS_OR_PARSE_PANIC( "premature end of file while in primer WR item" );

               RWCString soDataLine( szLine );

               if ( soDataLine.length() >= 2 && 
                    soDataLine[0] == '}' &&
                    isspace( soDataLine[1] ) )

                  bEndOfWRItem = true;
               else if ( soDataLine.length() >= 5 &&
                         soDataLine( 0, 5 ) == "type:" ) {
                  if ( soDataLine.length() < 6 )
                     PARSE_PANIC( "In primer WR item, type line which should be of the form type: univ fwd, univ rev, pcr end, or walk" );
                  
                  soDataLine = soDataLine( 5, soDataLine.length() - 5 );

                  soDataLine = soDataLine.stripWhitespace( RWCString::BOTH );

                  if ( soDataLine == "univ fwd" ) 
                     pSingletInfo->nReadType_ = nUniversalForward;
                  else if ( soDataLine == "univ rev" )
                     pSingletInfo->nReadType_ = nUniversalReverse;
                  else if ( soDataLine == "walk" )
                     pSingletInfo->nReadType_ = nWalk;
                  else if ( soDataLine == "pcr end" )
                     pSingletInfo->nReadType_ = nPCREnd;
                  else {
                     PARSE_PANIC( "after type: there must be one of univ fwd, univ rev, pcr end, or walk" )
                        }
               } // ...soDataLine( 0, 5 ) == "type:"
            } while( !bEndOfWRItem );
         }
      }  // if ( szLine[0] == 'W' ...
   } // while( 1 )


   if ( bFoundExpID ) {
      readPHDAgainForHighQualitySegment( pSingletInfo, pFil );
   }
   else {
      pSingletInfo->nHighQualitySegmentStart_ = -1;
      pSingletInfo->nHighQualitySegmentEnd_ = -1;
      pSingletInfo->nExpID_ = -1;
   }

   fclose( pFil );
}