/**********************************************************************
 *
 * browsefilter.cpp -- 
 * Copyright (C) 1999  The New Zealand Digital Library Project
 *
 * A component of the Greenstone digital library software
 * from the New Zealand Digital Library Project at the
 * University of Waikato, New Zealand.
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * This program 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 program; if not, write to the Free Software
 * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
 *
 *********************************************************************/

#include "browsefilter.h"
#include "fileutil.h"


browsefilterclass::browsefilterclass () {
  db_ptr = NULL;

  // -- onePerQuery StartResults   integer
  FilterOption_t filtopt;
  filtopt.name = "StartResults";
  filtopt.type = FilterOption_t::integert;
  filtopt.repeatable = FilterOption_t::onePerQuery;
  filtopt.defaultValue = "1";
  filtopt.validValues.push_back("1");
  filtopt.validValues.push_back("10000");
  filterOptions["StartResults"] = filtopt;

  // -- onePerQuery EndResults     integer
  filtopt.clear();
  filtopt.name = "EndResults";
  filtopt.type = FilterOption_t::integert;
  filtopt.repeatable = FilterOption_t::onePerQuery;
  filtopt.defaultValue = "-1";
  filtopt.validValues.push_back("-1");
  filtopt.validValues.push_back("10000");
  filterOptions["EndResults"] = filtopt;

  // -- onePerQuery ParentNode     string ("" will return the browsing available)
  filtopt.clear();
  filtopt.name = "ParentNode";
  filtopt.type = FilterOption_t::stringt;
  filtopt.repeatable = FilterOption_t::onePerQuery;
  filtopt.defaultValue = g_EmptyText;
  filterOptions["ParentNode"] = filtopt;
}

browsefilterclass::~browsefilterclass () {
}

void browsefilterclass::configure (const text_t &key, const text_tarray &cfgline) {
  filterclass::configure (key, cfgline);
  if (key == "indexstem") {
    indexstem = cfgline[0];
  }
}

bool browsefilterclass::init (ostream &logout) {
  outconvertclass text_t2ascii;

  if (!filterclass::init(logout)) return false;

  if (db_ptr == NULL) {
    // most likely a configuration problem
    logout << text_t2ascii 
	   << "configuration error: browsefilter contains a null dbclass\n\n";
    return false;
  }

  if (indexstem.empty()) {
    indexstem = collection;
  }

  db_filename = resolve_db_filename(indexstem,db_ptr->getfileextension());
  if (!file_exists(db_filename)) {
    logout << text_t2ascii
	   << "warning: database \"" << db_filename << "\" does not exist\n\n";
    //    return false;
  }

  return true;
}

void browsefilterclass::filter (const FilterRequest_t &request,
				FilterResponse_t &response,
				comerror_t &err, ostream &logout) {
  int numDocs = 0;
  outconvertclass text_t2ascii;

  response.clear ();
  err = noError;
  if (db_ptr == NULL) {
    // most likely a configuration problem
    logout << text_t2ascii 
	   << "configuration error: browsefilter contains a null dbclass\n\n";
    err = configurationError;
    return;
  }

  // open the database
  db_ptr->setlogout(&logout);
  if (!db_ptr->opendatabase (db_filename, DB_READER, 100, false)) {
    // most likely a system problem (we have already checked that the database exists)
    logout << text_t2ascii
	   << "system problem: open on database \""
	   << db_filename << "\" failed\n\n";
    err = systemProblem;
    return;
  }

  // get the browse parameters
  int startresults = filterOptions["StartResults"].defaultValue.getint();
  int endresults = filterOptions["EndResults"].defaultValue.getint();
  text_t parentnode = filterOptions["ParentNode"].defaultValue;
  OptionValue_tarray::const_iterator options_here = request.filterOptions.begin();
  OptionValue_tarray::const_iterator options_end = request.filterOptions.end();
  while (options_here != options_end) {
    if ((*options_here).name == "StartResults") 
      startresults = (*options_here).value.getint();
    else if ((*options_here).name == "EndResults")
      endresults = (*options_here).value.getint();
    else if ((*options_here).name == "ParentNode")
      parentnode = (*options_here).value;
    else {
      logout << text_t2ascii
	     << "warning: unknown browsefilter option \""
	     << (*options_here).name
	     << "\" ignored.\n\n";
    }

    ++options_here;
  }

  infodbclass info;

  // translate any ".fc", ".pr" etc. stuff in the parentnode
  parentnode = db_ptr->translate_OID (parentnode, info);

  // adjust topmost browsing node
  if (parentnode.empty()) parentnode = "browse";

  // get the node
  if ((request.filterResultOptions & FROID) ||
      (request.filterResultOptions & FRmetadata)) {
    if (!db_ptr->getinfo(parentnode, info)) {
      // didn't find the node
      logout << text_t2ascii
	     << "warning: lookup for node \"" << parentnode 
	     << "\" failed for browsefilter.\n\n";
    } else {
      // found the node

      // replace " with the parent node name and split the contains string
      // into the result set
      text_tarray resultset;
      text_t tmptext;
      text_t &contains = info["contains"];
      text_t::iterator contains_here = contains.begin();
      text_t::iterator contains_end = contains.end();
      while (contains_here != contains_end) {
	if (*contains_here == '"') tmptext += parentnode;
	else if (*contains_here == ';') {
	  if (!tmptext.empty()) resultset.push_back (tmptext);
	  tmptext.clear();
	} else tmptext.push_back(*contains_here);
	
	++contains_here;
      }
      // insert the last result in the set
      if (!tmptext.empty()) resultset.push_back (tmptext);

      text_tarray offset_resultset;
      text_t &md_type = info["mdtype"];
      if (!md_type.empty())
	{
	  text_t &md_offset = info["mdoffset"];
	  if (!md_offset.empty())
	    {
	      text_t offsettext;

	      text_t::iterator offset_here = md_offset.begin();
	      text_t::iterator offset_end = md_offset.end();
	      while (offset_here != offset_end) 
		{
		  if (*offset_here == ';')
		    {
		      if (offsettext.empty())
			{
			  offset_resultset.push_back ("0");
			}
		      else
			{
			  offset_resultset.push_back (offsettext);
			}
		      offsettext.clear();
		    } 
		  else 
		    {
		      offsettext.push_back(*offset_here);
		    }
		  
		  ++offset_here;
		}
	      // insert the last result in the set
	      if (offsettext.empty())
		{
		  offset_resultset.push_back ("0");
		}
	      else
		{
		  offset_resultset.push_back (offsettext);
		}
	    }
	  else
	    {
	      // add 0 offset for each 'contains' entry
	      text_tarray::iterator result_here = resultset.begin();
	      text_tarray::iterator result_end = resultset.end();
	      while (result_here != result_end) {
		offset_resultset.push_back("0");
		++result_here;
	      }
	    }

	  // do an intersection with the input set
	  if (!request.docSet.empty()) {

	    text_tarray intersect_resultset;
	    text_tarray intersect_offset_resultset;
	    
	    text_tarray::const_iterator resultset_here = resultset.begin();
	    text_tarray::const_iterator resultset_end = resultset.end();
	    text_tarray::const_iterator offset_resultset_here = offset_resultset.begin();
	    
	    while (resultset_here != resultset_end) {
	      if (in_set (request.docSet, *resultset_here))
		{
		  intersect_resultset.push_back (*resultset_here);
		  intersect_offset_resultset.push_back (*offset_resultset_here);
		}
	      ++resultset_here;
	      ++offset_resultset_here;
	    }
	    resultset = intersect_resultset;
	    offset_resultset = intersect_offset_resultset;
	  }
	}
      else
	{
	  // do an intersection with the input set
	  if (!request.docSet.empty()) {
	    intersect (resultset, request.docSet);
	  }

	  // add 0 offset for each 'contains' entry
	  text_tarray::iterator result_here = resultset.begin();
	  text_tarray::iterator result_end = resultset.end();
	  while (result_here != result_end) {
	    offset_resultset.push_back("0");
	    ++result_here;
	  }
	}

      // create the response
      numDocs = resultset.size();
      int resultnum = 1;
      ResultDocInfo_t resultdoc;
      text_tarray::iterator result_here = resultset.begin();
      text_tarray::iterator result_end = resultset.end();
      text_tarray::iterator offset_result_here = offset_resultset.begin();

      while (result_here != result_end) {
	// if endresults is -1 get all results
	if ((endresults != -1) && (resultnum > endresults)) break;
	if (resultnum >= startresults) {
	  resultdoc.OID = (*result_here);
	  if (!md_type.empty())
	    {
	      resultdoc.classifier_metadata_type = md_type;
	      resultdoc.classifier_metadata_offset = offset_result_here->getint();
	    }
	  response.docInfo.push_back(resultdoc);
	}

	++resultnum;
	++result_here;
	if (!md_type.empty()) ++offset_result_here;
      }
    }
  }

  db_ptr->closedatabase();  // Important that local library doesn't leave any files open
  response.numDocs = numDocs;
  response.isApprox = Exact;
}