nsSBCSGroupProber.cpp

Go to the documentation of this file.

/* -*- Mode: C; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/*  -*- C++ -*-
*  Copyright (C) 1998 <developer@mozilla.org>
*
*
*  Permission is hereby granted, free of charge, to any person obtaining
*  a copy of this software and associated documentation files (the
*  "Software"), to deal in the Software without restriction, including
*  without limitation the rights to use, copy, modify, merge, publish,
*  distribute, sublicense, and/or sell copies of the Software, and to
*  permit persons to whom the Software is furnished to do so, subject to
*  the following conditions:
*
*  The above copyright notice and this permission notice shall be included 
*  in all copies or substantial portions of the Software.
*
*  THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
*  EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
*  MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
*  NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
*  LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
*  OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
*  WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*/

#include "nsSBCSGroupProber.h"

#include "nsSBCharSetProber.h"
#include "nsHebrewProber.h"
#include "UnicodeGroupProber.h"

#include <stdio.h>
#include <stdlib.h>

namespace kencodingprober {
nsSBCSGroupProber::nsSBCSGroupProber()
{
  mProbers[0] = new nsSingleByteCharSetProber(&Win1251Model);
  mProbers[1] = new nsSingleByteCharSetProber(&Koi8rModel);
  mProbers[2] = new nsSingleByteCharSetProber(&Latin5Model);
  mProbers[3] = new nsSingleByteCharSetProber(&MacCyrillicModel);
  mProbers[4] = new nsSingleByteCharSetProber(&Ibm866Model);
  mProbers[5] = new nsSingleByteCharSetProber(&Ibm855Model);
  mProbers[6] = new nsSingleByteCharSetProber(&Latin7Model);
  mProbers[7] = new nsSingleByteCharSetProber(&Win1253Model);
  mProbers[8] = new nsSingleByteCharSetProber(&Latin5BulgarianModel);
  mProbers[9] = new nsSingleByteCharSetProber(&Win1251BulgarianModel);

  nsHebrewProber *hebprober = new nsHebrewProber();
  // Notice: Any change in these indexes - 10,11,12 must be reflected
  // in the code below as well.
  mProbers[10] = hebprober;
  mProbers[11] = new nsSingleByteCharSetProber(&Win1255Model, false, hebprober); // Logical Hebrew
  mProbers[12] = new nsSingleByteCharSetProber(&Win1255Model, true, hebprober); // Visual Hebrew
  mProbers[13] = new UnicodeGroupProber();
  
  // Tell the Hebrew prober about the logical and visual probers
  if (mProbers[10] && mProbers[11] && mProbers[12]) // all are not null
  {
    hebprober->SetModelProbers(mProbers[11], mProbers[12]);
  }
  else // One or more is null. avoid any Hebrew probing, null them all
  {
    for (unsigned int i = 10; i <= 12; ++i)
    { 
      delete mProbers[i]; 
      mProbers[i] = 0; 
    }
  }

  // disable latin2 before latin1 is available, otherwise all latin1 
  // will be detected as latin2 because of their similarity.
  //mProbers[10] = new nsSingleByteCharSetProber(&Latin2HungarianModel);
  //mProbers[11] = new nsSingleByteCharSetProber(&Win1250HungarianModel);

  Reset();
}

nsSBCSGroupProber::~nsSBCSGroupProber()
{
  for (unsigned int i = 0; i < NUM_OF_SBCS_PROBERS; i++)
  {
    delete mProbers[i];
  }
}


const char* nsSBCSGroupProber::GetCharSetName()
{
  //if we have no answer yet
  if (mBestGuess == -1)
  {
    GetConfidence();
    //no charset seems positive
    if (mBestGuess == -1)
      //we will use default.
      mBestGuess = 0;
  }
  return mProbers[mBestGuess]->GetCharSetName();
}

void  nsSBCSGroupProber::Reset(void)
{
  mActiveNum = 0;
  for (unsigned int i = 0; i < NUM_OF_SBCS_PROBERS; i++)
  {
    if (mProbers[i]) // not null
    {
      mProbers[i]->Reset();
      mIsActive[i] = true;
      ++mActiveNum;
    }
    else
      mIsActive[i] = false;
  }
  mBestGuess = -1;
  mState = eDetecting;
}


nsProbingState nsSBCSGroupProber::HandleData(const char* aBuf, unsigned int aLen)
{
  nsProbingState st;
  unsigned int i;
  char *newBuf1 = 0;
  unsigned int newLen1 = 0;

  //apply filter to original buffer, and we got new buffer back
  //depend on what script it is, we will feed them the new buffer 
  //we got after applying proper filter
  //this is done without any consideration to KeepEnglishLetters
  //of each prober since as of now, there are no probers here which
  //recognize languages with English characters.
  if (!FilterWithoutEnglishLetters(aBuf, aLen, &newBuf1, newLen1))
    goto done;
  
  if (newLen1 == 0)
    goto done; // Nothing to see here, move on.

  for (i = 0; i < NUM_OF_SBCS_PROBERS; ++i)
  {
     if (!mIsActive[i])
       continue;
     st = mProbers[i]->HandleData(newBuf1, newLen1);
     if (st == eFoundIt)
     {
       mBestGuess = i;
       mState = eFoundIt;
       break;
     }
     else if (st == eNotMe)
     {
       mIsActive[i] = false;
       mActiveNum--;
       if (mActiveNum <= 0)
       {
         mState = eNotMe;
         break;
       }
     }
  }

done:
  free(newBuf1);

  return mState;
}

float nsSBCSGroupProber::GetConfidence(void)
{
  unsigned int i;
  float bestConf = 0.0, cf;

  switch (mState)
  {
  case eFoundIt:
    return (float)0.99; //sure yes
  case eNotMe:
    return (float)0.01;  //sure no
  default:
    for (i = 0; i < NUM_OF_SBCS_PROBERS; ++i)
    {
      if (!mIsActive[i])
        continue;
      cf = mProbers[i]->GetConfidence();
      if (bestConf < cf)
      {
        bestConf = cf;
        mBestGuess = i;
      }
    }
  }
  return bestConf;
}

#ifdef DEBUG_PROBE
void nsSBCSGroupProber::DumpStatus()
{
  unsigned int i;
  float cf;
  
  cf = GetConfidence();
  printf(" SBCS Group Prober --------begin status \r\n");
  for (i = 0; i < NUM_OF_SBCS_PROBERS; i++)
  {
    if (!mIsActive[i])
      printf("  inactive: [%s] (i.e. confidence is too low).\r\n", mProbers[i]->GetCharSetName());
    else
      mProbers[i]->DumpStatus();
  }
  printf(" SBCS Group found best match [%s] confidence %f.\r\n",  
         mProbers[mBestGuess]->GetCharSetName(), cf);
}
#endif
}