void DeconvolutionRL_wide() {
   Int_t i;
   const Int_t nbins = 256;
   Double_t xmin  = 0;
   Double_t xmax  = nbins;
   Double_t source[nbins];
   Double_t response[nbins];
   gROOT->ForceStyle();

   TString dir  = gROOT->GetTutorialsDir();
   TString file = dir+"/spectrum/TSpectrum.root";
   TFile *f     = new TFile(file.Data());
   h = (TH1F*) f->Get("decon3");
   h->SetTitle("Deconvolution of closely positioned overlapping peaks using Richardson-Lucy deconvolution method");
   d = (TH1F*) f->Get("decon_response_wide");

   for (i = 0; i < nbins; i++) source[i]=h->GetBinContent(i + 1);
   for (i = 0; i < nbins; i++) response[i]=d->GetBinContent(i + 1);

   h->SetMaximum(30000);
   h->Draw("L");
   TSpectrum *s = new TSpectrum();
   s->DeconvolutionRL(source,response,256,10000,1,1);

   for (i = 0; i < nbins; i++) d->SetBinContent(i + 1,source[i]);
   d->SetLineColor(kRed);
   d->Draw("SAME L");
}

int e428ana1::MakePara(Double_t thr){
    int npeak,nlap;
    double threshol,thrmin,thrmax;
    TSpectrum *spec;
    for (int i=0; i<4; i++) {
        for (int j=0; j<127; j++) {
            npeak=0;nlap=0;
            threshol=thr;
            thrmax=1;thrmin=0;
            while (npeak!=4 && nlap<100){
                spec = new TSpectrum(10);
                npeak = spec->Search(hdssd[i][j],2,"",threshol);
                if (npeak >3) {
                    thrmin=threshol;
                }else thrmax=threshol;
                threshol=(thrmax-thrmin)/2;
                nlap++;
        //        printf("Npeak   %d   %d    %d    %f\n",i,j,npeak,threshol);
            }
            float *xpeak=spec->GetPositionX();
            float *ypeak=spec->GetPositionY();
            printf("Npeak   %d   %d    %d",i,j,npeak);
            for (int i1=0; i1<npeak; i1++) {
                printf("    %f   %f",xpeak[i1],ypeak[i1]);
            }
            printf("\n");
        }
    }    
    return 0;
}

void Deconvolution_wide_boost() {
   Int_t i;
   const Int_t nbins = 256;
   Double_t xmin     = 0;
   Double_t xmax     = nbins;
   Double_t source[nbins];
   Double_t response[nbins];
   gROOT->ForceStyle();

   TH1F *h = new TH1F("h","Deconvolution",nbins,xmin,xmax);
   TH1F *d = new TH1F("d","",nbins,xmin,xmax);

   TString dir  = gROOT->GetTutorialDir();
   TString file = dir+"/spectrum/TSpectrum.root";
   TFile *f     = new TFile(file.Data());
   h = (TH1F*) f->Get("decon3");
   h->SetTitle("Deconvolution of closely positioned overlapping peaks using boosted Gold deconvolution method");
   d = (TH1F*) f->Get("decon_response_wide");

   for (i = 0; i < nbins; i++) source[i]=h->GetBinContent(i + 1);
   for (i = 0; i < nbins; i++) response[i]=d->GetBinContent(i + 1);

   h->SetMaximum(200000);
   h->Draw("L");
   TSpectrum *s = new TSpectrum();
   s->Deconvolution(source,response,256,200,50,1.2);

   for (i = 0; i < nbins; i++) d->SetBinContent(i + 1,source[i]);
   d->SetLineColor(kRed);
   d->Draw("SAME L");
}

TGraph* autogain152(TH1 *hist) {

   hist->GetXaxis()->SetRangeUser(200.,16000.);
   TSpectrum *s = new TSpectrum();
   Int_t nfound = s->Search(hist,6,"",0.08); //This will be dependent on the source used.
   printf("Found %d candidate peaks to fit\n",nfound);
   if(nfound > 6)
      nfound = 6;

   std::vector<float> vec;
   for(int x=0;x<nfound;x++)
      vec.push_back(s->GetPositionX()[x]);

   std::sort(vec.begin(),vec.end());

   Float_t energies[] = {121.7830, 244.6920, 344.276, 778.903, 964.131, 1408.011};
   TGraph* slopefit = new TGraph(nfound, &(vec[0]), energies);

   printf("Now fitting: Be patient\n");
   slopefit->Fit("pol1");
   if(slopefit->GetFunction("pol1")->GetChisquare() > 10.) {
      slopefit->RemovePoint(slopefit->GetN()-1);
      slopefit->Fit("pol1");
   }
   TChannel *chan = 0;
   slopefit->Draw("AC*");

   return slopefit;
}

void peaks(Int_t np=10) {
   npeaks = TMath::Abs(np);
   TH1F *h = new TH1F("h","test",500,0,1000);
   //generate n peaks at random
   Double_t par[3000];
   par[0] = 0.8;
   par[1] = -0.6/1000;
   Int_t p;
   for (p=0;p<npeaks;p++) {
      par[3*p+2] = 1;
      par[3*p+3] = 10+gRandom->Rndm()*980;
      par[3*p+4] = 3+2*gRandom->Rndm();
   }
   TF1 *f = new TF1("f",fpeaks,0,1000,2+3*npeaks);
   f->SetNpx(1000);
   f->SetParameters(par);
   TCanvas *c1 = new TCanvas("c1","c1",10,10,1000,900);
   c1->Divide(1,2);
   c1->cd(1);
   h->FillRandom("f",200000);
   h->Draw();
   TH1F *h2 = (TH1F*)h->Clone("h2");
   //Use TSpectrum to find the peak candidates
   TSpectrum *s = new TSpectrum(2*npeaks);
   Int_t nfound = s->Search(h,2,"",0.10);
   printf("Found %d candidate peaks to fit\n",nfound);
   //Estimate background using TSpectrum::Background
   TH1 *hb = s->Background(h,20,"same");
   if (hb) c1->Update();
   if (np <0) return;

   //estimate linear background using a fitting method
   c1->cd(2);
   TF1 *fline = new TF1("fline","pol1",0,1000);
   h->Fit("fline","qn");
   //Loop on all found peaks. Eliminate peaks at the background level
   par[0] = fline->GetParameter(0);
   par[1] = fline->GetParameter(1);
   npeaks = 0;
   Double_t *xpeaks = s->GetPositionX();
   for (p=0;p<nfound;p++) {
      Double_t xp = xpeaks[p];
      Int_t bin = h->GetXaxis()->FindBin(xp);
      Double_t yp = h->GetBinContent(bin);
      if (yp-TMath::Sqrt(yp) < fline->Eval(xp)) continue;
      par[3*npeaks+2] = yp;
      par[3*npeaks+3] = xp;
      par[3*npeaks+4] = 3;
      npeaks++;
   }
   printf("Found %d useful peaks to fit\n",npeaks);
   printf("Now fitting: Be patient\n");
   TF1 *fit = new TF1("fit",fpeaks,0,1000,2+3*npeaks);
   //we may have more than the default 25 parameters
   TVirtualFitter::Fitter(h2,10+3*npeaks);
   fit->SetParameters(par);
   fit->SetNpx(1000);
   h2->Fit("fit");
}

void Background_order() {
   Int_t i;
   const Int_t nbins = 4096;
   Double_t xmin     = 0;
   Double_t xmax     = 4096;
   Double_t source[nbins];
   gROOT->ForceStyle();

   TH1F *d1 = new TH1F("d1","",nbins,xmin,xmax);
   TH1F *d2 = new TH1F("d2","",nbins,xmin,xmax);
   TH1F *d3 = new TH1F("d3","",nbins,xmin,xmax);
   TH1F *d4 = new TH1F("d4","",nbins,xmin,xmax);

   TString dir  = gROOT->GetTutorialsDir();
   TString file = dir+"/spectrum/TSpectrum.root";
   TFile *f     = new TFile(file.Data());
   TH1F *back = (TH1F*) f->Get("back2");
   back->SetTitle("Influence of clipping filter difference order on the estimated background");
   back->SetAxisRange(1220,1460);
   back->SetMaximum(3000);
   back->Draw("L");

   TSpectrum *s = new TSpectrum();

   for (i = 0; i < nbins; i++) source[i]=back->GetBinContent(i + 1);
   s->Background(source,nbins,40,TSpectrum::kBackDecreasingWindow,
                 TSpectrum::kBackOrder2,kFALSE,
                 TSpectrum::kBackSmoothing3,kFALSE);
   for (i = 0; i < nbins; i++) d1->SetBinContent(i + 1,source[i]);
   d1->SetLineColor(kRed);
   d1->Draw("SAME L");

   for (i = 0; i < nbins; i++) source[i]=back->GetBinContent(i + 1);
   s->Background(source,nbins,40,TSpectrum::kBackDecreasingWindow,
                 TSpectrum::kBackOrder4,kFALSE,
                 TSpectrum::kBackSmoothing3,kFALSE);
   for (i = 0; i < nbins; i++) d2->SetBinContent(i + 1,source[i]);
   d2->SetLineColor(kBlue);
   d2->Draw("SAME L");

   for (i = 0; i < nbins; i++) source[i]=back->GetBinContent(i + 1);
   s->Background(source,nbins,40,TSpectrum::kBackDecreasingWindow,
                 TSpectrum::kBackOrder6,kFALSE,
                 TSpectrum::kBackSmoothing3,kFALSE);
   for (i = 0; i < nbins; i++) d3->SetBinContent(i + 1,source[i]);
   d3->SetLineColor(kGreen);
   d3->Draw("SAME L");

   for (i = 0; i < nbins; i++) source[i]=back->GetBinContent(i + 1);
   s->Background(source,nbins,40,TSpectrum::kBackDecreasingWindow,
                 TSpectrum::kBackOrder8,kFALSE,
                 TSpectrum::kBackSmoothing3,kFALSE);
   for (i = 0; i < nbins; i++) d4->SetBinContent(i + 1,source[i]);
   d4->SetLineColor(kMagenta);
   d4->Draw("SAME L");
}

TGraph* autogain(TH1 *hist,TNucleus *nuc) {    //Display The fits on a TPad  

   if(!hist || !nuc)
      return 0;

   nuc->SetSourceData();

   if(nuc->GetA() == 152) {
      return autogain152(hist);
   }

// Search
   hist->GetXaxis()->SetRangeUser(200.,16000.);
   TSpectrum *s = new TSpectrum();
   Int_t nfound = s->Search(hist,6,"",0.1); //This will be dependent on the source used.
   printf("Found %d candidate peaks to fit\n",nfound);
// Match

   nuc->TransitionList.Sort();

   std::vector<float> engvec;
   TIter iter(&(nuc->TransitionList));
   TObject* obj;
   while(obj = iter.Next()) {
      if(!obj->InheritsFrom("TGRSITransition"))
         continue;
      TGRSITransition *tran = (TGRSITransition*)obj;

      engvec.push_back(static_cast<float>(tran->energy));
      if(engvec.size() == nfound)
         break;
   }

   if(nfound != engvec.size())
      return 0;

   Float_t *posPeaks = s->GetPositionX();
   Float_t *energies = &(engvec[0]);

   for(int x=0;x<nfound;x++) {
      printf("posPeaks[%i] = %f\t\tenrgies[%i] = %f\n",x,posPeaks[x],x,energies[x]);
   }

   TGraph *slopefit = new TGraph(nfound,posPeaks,energies ); 

   printf("Now fitting: Be patient\n");
   slopefit->Fit("pol1");
   slopefit->Draw("AC*");

   return slopefit;

}

//______________________________________________________________________________
void Fit(Int_t run)
{
    // Perform fit.
    
    Char_t tmp[256];
    
    // delete old function
    if (gFitFunc) delete gFitFunc;

    // create fitting function
    if (gFitFunc) delete gFitFunc;
    sprintf(tmp, "fGauss_%d", run);
    gFitFunc = new TF1(tmp, "expo(0)+gaus(2)");
    gFitFunc->SetLineColor(2);
    
    // estimate peak position
    TSpectrum s;
    s.Search(gH, 10, "goff nobackground", 0.05);
    Double_t peak = TMath::MaxElement(s.GetNPeaks(), s.GetPositionX());
    
    // prepare fitting function
    gFitFunc->SetRange(gMin, gMax);
    gFitFunc->SetParameter(2, gH->GetXaxis()->FindBin(peak));
    gFitFunc->SetParLimits(2, 0, 100000);
    gFitFunc->SetParameter(3, peak);
    gFitFunc->SetParameter(4, 20);
    gFitFunc->SetParLimits(4, 18, 100);
    
    for (Int_t i = 0; i < 10; i++)
        if (!gH->Fit(gFitFunc, "RBQ0")) break;
    
    // get peak
    peak = gFitFunc->GetParameter(3);

    // indicator line
    gLine->SetX1(peak);
    gLine->SetX2(peak);
    gLine->SetY1(0);
    gLine->SetY2(gH->GetMaximum());

    // draw 
    gCFit->cd();
    gH->GetXaxis()->SetRangeUser(0, 2000);
    gH->Draw();
    gFitFunc->Draw("same");
    gLine->Draw("same");

    // fill overview histogram
    gHOverview->SetBinContent(run+1, peak);
    gHOverview->SetBinError(run+1, 0.0001);
}

//Given an input histogram and TSpectrum returns a numeric value based on the input keyword; supported keywords are "AMPLITUDE,MEAN,PEAK,SIGMA"
float AnalyzeResponseUniformity::getValByKeyword(string strInputKeyword, shared_ptr<TH1F> hInput, TSpectrum &specInput){
    
    //Try to automatically assign a value
    if (0 == strInputKeyword.compare("AMPLITUDE") ) { //Case: Histo Amplitude
        return hInput->GetBinContent( hInput->GetMaximumBin() );
    } //End Case: Histo Amplitude
    else if (0 == strInputKeyword.compare("MEAN") ) { //Case: Histo Mean
        return hInput->GetMean();
    } //End Case: Histo Mean
    else if ( 0 == strInputKeyword.compare("PEAK") ){ //Case: Histo Peak
        Double_t *dPeakPos = specInput.GetPositionX();
        
        return dPeakPos[0];
    } //End Case: Histo Peak
    else if (0 == strInputKeyword.compare("SIGMA") ) { //Case: Histo RMS
        return hInput->GetRMS();
    } //End Case: Histo RMS
    else{ //Case: manual assignment
        printClassMethodMsg("AnalyzeResponseUniformity","getValByKeyword","Error! Input Keyword Not Recognized");
        printClassMethodMsg("AnalyzeResponseUniformity","getValByKeyword", ("\tGiven: " + strInputKeyword ).c_str() );
        printClassMethodMsg("AnalyzeResponseUniformity","getValByKeyword","\tRecognized Keywords:\n");
        
        for (int i=0; i < vec_strSupportedKeywords.size(); ++i) {
            printClassMethodMsg("AnalyzeResponseUniformity","getValByKeyword", vec_strSupportedKeywords[i].c_str() );
        }
        
        printClassMethodMsg("AnalyzeResponseUniformity","getValByKeyword","\tUndefined Behavior May Occur");
        
        return -1e12;
    } //End Case: manual assignment
} //End AnalyzeResponseUniformity::getValByKeyword()

void MSSM_spectrum_plotter::write_spectrum(const TSpectrum& spectrum, std::ofstream& filestr) const
{
   for (std::size_t s = 0; s < spectrum.size(); ++s) {
      if (!filestr.good()) {
         ERROR("MSSM_spectrum_plotter::write_spectrum: "
               "file stream is corrupted");
         break;
      }

      const std::string& name = spectrum[s].name;
      const std::string& latex_name = spectrum[s].latex_name;
      const std::valarray<double>& masses = spectrum[s].masses;
      const std::size_t multiplicity = masses.size();

      filestr << std::left << "# " << name << '\n';

      for (std::size_t i = 0; i < multiplicity; ++i) {
         std::string lname("$" + latex_name + "$");
         std::stringstream lname_with_index;
         lname_with_index << "$" << latex_name;
         if (multiplicity > 1)
            lname_with_index << "_{" << (i+1) << "}";
         lname_with_index  << "$";

         filestr << std::left << std::setw(width) << s
                 << std::left << std::setw(width) << masses[i]
                 << std::left << std::setw(width) << name
                 << std::left << std::setw(2*width) << lname
                 << std::left << std::setw(2*width) << lname_with_index.str()
                 << '\n';
      }
   }
}

void FindPeak(TH1 *hm, int * i, char * namefile){
  int np =5, p, max = 0;
  int npeaks = TMath::Abs(np);
  double par[3000];
  par[0] = 0.8;
  par[1] = -0.6/1000;

  for (p=0;p<npeaks;p++) {
    par[3*p+2] = 1;
    par[3*p+3] = 10+gRandom->Rndm()*980;
    par[3*p+4] = 3+2*gRandom->Rndm();
  }
 
  TSpectrum *s = new TSpectrum(2*npeaks,1);
  int nfound = s->Search(hm,2,"",0.10);
  printf("Found %d candidate peaks to fit\n",nfound);

  TH1 *hb = s->Background(hm,20,"same");
  if (np <0) return;

  // loope over peaks
  TF1 *fline = new TF1("fline","pol1",0,1000);
  hm->Fit("fline","qn");
  par[0] = fline->GetParameter(0);
  par[1] = fline->GetParameter(1);
  npeaks = 0;
  float *xpeaks = s->GetPositionX();
  for (p=0;p<nfound;p++) {
    float xp = xpeaks[p];
    int bin = hm->GetXaxis()->FindBin(xp);
    float yp = hm->GetBinContent(bin);
    if (yp-TMath::Sqrt(yp) < fline->Eval(xp)) continue;
    par[3*npeaks+2] = yp;
    par[3*npeaks+3] = xp;
    par[3*npeaks+4] = 3;
    npeaks++;
  }
  printf("Found %d useful peaks to fit\n",npeaks); 
  printf("Now fitting: Be patient\n");
  if (max < npeaks) max = npeaks;
  TF1 *fit = new TF1("fit",fpeaks,0,1000,2+3*npeaks);
  TVirtualFitter::Fitter(hm,10+3*npeaks);
  fit->SetParameters(par);
  fit->SetNpx(1000);
  hm->Fit("fit"); 
}

////    This version needs thought
void SiCalibrator::FindPeaks() {
	Load();
	int iterate = 25;
	int nbins;
	TSpectrum* spec = new TSpectrum(10);
	avesigma = 0;
	int nsigma = 0;
	for (int src=0; src < CalData.size(); src++) {
		if (CalData[src].hSource != 0) {
			for (int ch=0; ch<CalData[src].sourcedata.size(); ch++) {
				TH1D* hbi = CalData[src].hSource->ProjectionY("hb",ch+1,ch+1);
				nbins = hbi->GetNbinsX();
				TH1D* hbk1 = (TH1D*) spec->Background(hbi,iterate);
				hbi->Add(hbk1,-1);
				//Estimate parameters
				double lim = 5;
				double max = 0;
				double peak = 0;
				int i = nbins - 1;
				while (hbi->GetBinContent(i) < lim && i > 0) {
					max = i;
					i--;
				}
				double sigma = (max/CalibSource::sourcelist[src].betas.back().E)*2; //2 keV
				if (sigma > 1) {
					avesigma += sigma;
					nsigma++;
				}
				gErrorIgnoreLevel = kError;
				int npeaks = spec->Search(hbi,sigma,"nodraw ",0.001);
				float* adc = spec->GetPositionX();
				float* amp = spec->GetPositionY();
				for (int i=0; i<npeaks; i++) {
					CalData[src].sourcedata[ch].ADC.push_back(adc[i]);
					CalData[src].sourcedata[ch].Amp.push_back(amp[i]);
				}
			}
		}
	}
	avesigma = avesigma/nsigma;
}

/*
  Finds peak in timing difference spectrum
*/
double DiffPeak(TTree *&tree, TString pidcut = "PID>1000")
{
	TSpectrum *spec = new TSpectrum(10); // peak finder
	TH1D *diff = new TH1D("diff","diff",200,-5,5);

	tree->Project("diff","diff",pidcut.Data());
	spec->Search(diff,2,"nodraw",0.7);
	double diffpeak = spec->GetPositionX()[0];

	// testing
	//TCanvas *c_tmp = new TCanvas();
	//diff->Draw();
	//c_tmp->WaitPrimitive();
	//delete c_tmp;
	

	delete diff;
	delete spec;

	return diffpeak;
} // end DiffPeak()

void draw_ToF_Calib_Tower() {

  ofstream fout("ToF_Calib_Tower.txt");
  TFile* f = new TFile("/phenix/plhf/zji/taxi/Run13pp510MinBias/8328/data/total.root");
  TH2I* h_hitmap_tof_energy_PbSc = (TH2I*)f->Get("hitmap_tof_energy_PbSc");
  TH2I* h_hitmap_tof_energy_PbGl = (TH2I*)f->Get("hitmap_tof_energy_PbGl");
  TH2I* h_hitmap_tof_tower = (TH2I*)f->Get("hitmap_tof_tower");

  TCanvas* c = new TCanvas("c", "Canvas", 1800, 600);
  gStyle->SetOptStat(1);
  gStyle->SetOptFit(1);
  c->Divide(3,1);

  c->cd(1);
  h_hitmap_tof_energy_PbSc->Draw("colz");

  c->cd(2);
  h_hitmap_tof_energy_PbGl->Draw("colz");

  c->cd(3);
  fout << "towerid    ToF peak" << endl; 
  for(Int_t id=0; id<25000; id++) {
    TH1D* hp_hitmap_tof_tower = (TH1D*)h_hitmap_tof_tower->ProjectionY("_py",id+1,id+1);
    TSpectrum* peak = new TSpectrum();
    Int_t nfound = peak->Search(hp_hitmap_tof_tower,2.,"nodraw");
    if(nfound) {
      Float_t peakX = *peak->GetPositionX();
      hp_hitmap_tof_tower->Fit("gaus","Q0","",peakX-3.,peakX+3.);
      TF1* f_hitmap_tof_tower = hp_hitmap_tof_tower->GetFunction("gaus");
      Double_t mean_hitmap_tof_tower = f_hitmap_tof_tower->GetParameter(1);
      fout << id << "    " << mean_hitmap_tof_tower << endl;
      f_hitmap_tof_tower->Delete();
    }
    else {
      fout << id << "    " << "-100." << endl;
    }
    delete peak;
    hp_hitmap_tof_tower->Delete();
  }
  fout.close();

  TH1D* hp_hitmap_tof_tower = (TH1D*)h_hitmap_tof_tower->ProjectionY("_py",100,100);
  TSpectrum* peak = new TSpectrum();
  peak->Search(hp_hitmap_tof_tower,2.,"nodraw");
  Float_t peakX = *peak->GetPositionX();
  hp_hitmap_tof_tower->Fit("gaus","","",peakX-3.,peakX+3.);
  TF1* f_hitmap_tof_tower = hp_hitmap_tof_tower->GetFunction("gaus");
  Double_t mean_hitmap_tof_tower = f_hitmap_tof_tower->GetParameter(1);
  cout << "peakX=" << peakX << endl;
  cout << "ToF mean=" << mean_hitmap_tof_tower << endl;

  c->Print("ToF_Calib_Tower.pdf");

}

//________________________________________________________________________________
void peaks(TH1* h, Int_t np=10, Int_t ij=0) {
  if (! h) return;
  npeaks = TMath::Abs(np);
  if (! c1)  c1 = new TCanvas();
  else       c1->Clear();
  if (c2 && ij > 0) {c2->cd(ij); h->Draw(); c2->Update();}
  c1->Divide(1,2);
  c1->cd(1);
  h->Draw();
  TH1F *h2 = (TH1F*)h->Clone("h2");
  //Use TSpectrum to find the peak candidates
  TSpectrum *s = new TSpectrum(2*npeaks);
  Int_t nfound = s->Search(h,5,"",0.05);
  printf("Found %d candidate peaks to fit\n",nfound);
  if (! nfound) return;
  //Estimate background using TSpectrum::Background
  TH1 *hb = s->Background(h,20,"same");
  hb->Draw("same");
  c1->Update();
  if (c2 && ij > 0) {c2->cd(ij); h->Draw(); c2->Update();}
  if (np <0) return;

  //estimate linear background using a fitting method
  c1->cd(2);
  TF1 *fline = new TF1("fline","pol1",0,1000);
  fline->FixParameter(1,0.);
  h->Fit("fline","qnlw");
  if (c2 && ij > 0) {c2->cd(ij+1); h->Draw(); c2->Update(); c1->cd(2);}
  //Loop on all found peaks. Eliminate peaks at the background level
  Double_t par[3000];
  par[0] = fline->GetParameter(0);
  par[1] = fline->GetParameter(1);
  npeaks = 0;
  Float_t *xpeaks = s->GetPositionX();
  Float_t ymax = 0;
  for (Int_t p=0;p<nfound;p++) {
    Float_t xp = xpeaks[p];
    Int_t bin = h->GetXaxis()->FindBin(xp);
    Float_t yp = h->GetBinContent(bin);
    if (yp-3*TMath::Sqrt(yp) < fline->Eval(xp)) continue;
    par[3*npeaks+2] = yp;
    if (yp > ymax) ymax = yp;
    par[3*npeaks+3] = xp;
    par[3*npeaks+4] = 3;
    npeaks++;
  }
  cout << "Found " << npeaks << " useful peaks to fit" << endl;
  Int_t NP = 0;
  Int_t nbad = 0;
  TString LineH("  {\"");  LineH +=  h->GetName(); LineH += "\"";
  TString Line("");
  struct ParErr_t {Double_t par, err;};
  ParErr_t parErr[10];
  TF1 *fit = 0;
  if (ymax > 2*par[0]) {
    cout << "Now fitting: Be patient" << endl;
    fit = new TF1("fit",fpeaks,0,1000,2+3*npeaks);
    TVirtualFitter::Fitter(h2,10+3*npeaks); //we may have more than the default 25 parameters
    fit->SetParameter(0,par[0]);
    fit->FixParameter(1,0.);
    for (Int_t p = 0; p < npeaks; p++) {
      fit->SetParName(3*p+2,Form("A%i",p));
      fit->SetParLimits(3*p+2,0,1e6);
      fit->SetParameter(3*p+2,par[3*p+2]);
      fit->SetParName(3*p+3,Form("#mu%i",p));
      fit->SetParLimits(3*p+3,TMath::Max(0.,par[3*p+3]-2), TMath::Min(240.,par[3*p+3]+2));
      fit->SetParameter(3*p+3,par[3*p+3]);
      fit->SetParName(3*p+4,Form("#sigma%i",p));
      fit->SetParLimits(3*p+4,0.01,20);
      fit->SetParameter(3*p+4,par[3*p+4]);
    }
    fit->SetNpx(1000);
    h2->SetStats(0);
    h2->Fit("fit","l");
    if (c2 && ij > 0) {c2->cd(ij); h2->Draw("same"); c2->Update(); c1->cd(2);}
    fit->GetParameters(par);
    for (Int_t p = 0; p<np;p++) {
      if (p < npeaks && par[3*p+2] > 5*fit->GetParError(3*p+2) && 
	  par[3*p+2] > par[0]) {
	if (TMath::Abs(par[3*p+4]) > 5.0) nbad++;
	//      Line += Form(",%f,%f,%7.2f,%5.2f",par[3*p+2],fit->GetParError(3*p+2),par[3*p+3],TMath::Abs(par[3*p+4]));
	parErr[NP].par = par[3*p+3];
	parErr[NP].err = TMath::Abs(par[3*p+4]);
	for (Int_t i = 0; i < NP; i++) {
	  if (parErr[i].par > parErr[NP].par) {
	    ParErr_t temp = parErr[i];
	    parErr[i] = parErr[NP];
	    parErr[NP] = temp;
	  }
	}
	NP++;
      } 
    }
  }
  for (Int_t p = 0; p < np; p++) {
    if (p < NP) Line += Form(",%7.2f,%5.2f",parErr[p].par,parErr[p].err);
    else  Line += ",0,0";
  }
  Line += "},";
//.........这里部分代码省略.........

void fit_double_gaussian(TH1F*& hrsp)
{
  if (0==hrsp) {
    cout<<"ERROR: Empty pointer to fit_double_gaussian()"<<endl;return;
  }
  
  string histname = hrsp->GetName();
  double mean     = hrsp->GetMean();
  double rms      = hrsp->GetRMS();

  TSpectrum *spec = new TSpectrum(10);
  int nfound = spec->Search(hrsp,1,"new");
  if(nfound!=2)return;

  double *xpeaks = spec->GetPositionX();

  double peak1  = xpeaks[0];
  double bin1   = hrsp->FindBin(peak1);
  double norm1  = hrsp->GetBinContent(bin1);
  double sigma1 = 0.2;

  double peak2  = xpeaks[1];
  double bin2   = hrsp->FindBin(peak2);
  double norm2  = hrsp->GetBinContent(bin2);
  double sigma2 = 0.3;


  cout << " Mean  : "  << mean  << " \t  RMS  : " << rms    << endl;
  cout << " norm1 : "  << norm1 << " \t  norm2 : " << norm2 << endl;
  cout << " peak1 : "  << peak1 << " \t  sig1 : " << sigma1 << endl;
  cout << " peak2 : "  << peak2 << " \t  sig2 : " << sigma2 << endl;

  double fitrange_min = 0.2;
  double fitrange_max = 1.7;

  TF1* fitfnc(0); int fitstatus(-1);
  TF1 *fitg1(0), *fitg2(0);
  fitfnc = new TF1("fdgaus","gaus(0)+gaus(3)",fitrange_min,fitrange_max);
  fitfnc->SetLineColor(1);
  fitfnc->SetLineStyle(2);
  fitfnc->SetLineWidth(2);

  fitfnc->SetParNames("N_{1}", "#mu_{1}", "#sigma_{1}",
		      "N_{2}", "#mu_{2}", "#sigma_{2}");
  fitfnc->SetParameters(norm1, peak1, sigma1, 
   			norm2, peak2, sigma2); 


  //fitfnc->SetParLimits(0,0.0,2.0*norm1);
  //fitfnc->SetParLimits(1,peak1-3*sigma1,peak1+3*sigma1);
  //fitfnc->SetParLimits(2,0.1,1.0);

  //fitfnc->SetParLimits(3,0.0,1.5*norm2);
  //fitfnc->SetParLimits(4,peak2-3*sigma2,peak2+3*sigma2);
  //fitfnc->SetParLimits(5,0.01,1.0);

  fitstatus = hrsp->Fit(fitfnc,"RQ");
  // if (0!=fitstatus){
  //   fitfnc->SetParLimits(4,0.2,1.7);
  //   fitfnc->SetParLimits(5,2.0,10.0);
  //   //cout <<" Not able to Fit this pt bin " << hrsp->GetName() << endl;
  // }

  fitstatus = hrsp->Fit(fitfnc,"RQ");
  hrsp->SetMaximum(norm1+0.2*norm1);
  fitg1 = new TF1("fg1","gaus(0)",fitrange_min,fitrange_max);
  fitg1->SetParameters(fitfnc->GetParameter(0),
		       fitfnc->GetParameter(1),
		       fitfnc->GetParameter(2));
  fitg1->SetLineColor(2);
  fitg1->SetLineStyle(2);
  hrsp->GetListOfFunctions()->Add(fitg1);

  fitg2 = new TF1("fg2","gaus(0)",fitrange_min,fitrange_max);
  fitg2->SetParameters(fitfnc->GetParameter(3),
		       fitfnc->GetParameter(4),
		       fitfnc->GetParameter(5));
  fitg2->SetLineColor(4);
  fitg2->SetLineStyle(4);
  hrsp->GetListOfFunctions()->Add(fitg2);

  if(hrsp->GetFunction("fdgaus")==0){
    cout << "No function recorded in histogram " << hrsp->GetName() << endl;
  }
  if (0!=fitstatus){
    cout<<"fit_double_gaussian() to "<<hrsp->GetName()
        <<" failed. Fitstatus: "<<fitstatus
        <<" - FNC deleted."<<endl;
    hrsp->GetListOfFunctions()->Delete();
  }
}

I2GFvalues I2GFmainLoop(TH1F *htemp, int N_iter, float N_sigma_range, bool ShowFit)
//Arguments: (histo to be fit, N iterations to find peak using gaus fit, fit range param., do or do not plot fit on histo)
{
  I2GFvalues myI2GFvalues;

  //Set initial values...(in case fit fails)
  myI2GFvalues.rchi2 = -100;
  myI2GFvalues.mean = -100;
  myI2GFvalues.mean_err = -100;
  myI2GFvalues.sigma = -100;
  myI2GFvalues.sigma_err = -100;


  TSpectrum *s = new TSpectrum(); //TSpectrum(1,1)->Argument: (Number of peaks to find, Distance to neighboring peak: "1"-->3sigma)
  Int_t NPeaks;
  Float_t *Peak;                     //TSpectrum *s = new TSpectrum(); --> No warning message 
  Float_t *PeakAmp;                    
  float peak_pos = 0;
  float peak_pos_amp = 0;  //Initial value, assuming positive going peaks
  int peak_pos_bin = 0;
  
  int binMaxCnt = 0;
  float binMaxCnt_value = 0;
  int binMaxCnt_counts = 0;

  int Nbins = 0;
  Int_t zero_value_bin = 0;
  float low_limit = 0;
  float high_limit = 0;
  float peak_pos_count = 0;
  float zero_bin_value = 0;
  float max_bin_value = 0;
 

  TF1 *func;
  TF1 *func1;
  TF1 *func2;
  TF1 *func3;
  float Chi2;
  int NDF = 1;
   

  float f_RChi2 = 1; 
  float f_const = 1;
  float f_mean = 1;
  float f_mean_err = 1;
  float f_sigma = 1;
  float f_sigma_err = 1;
  
  float peak = 1;  

  float f_const2 = 1;
  float f_mean2 = 1;
  float f_mean_err2 = 1;
  float f_sigma2 = 1;
  float f_sigma_err2 = 1;

  //---------Basic Histo Peak Finding Parameters----------
  binMaxCnt = htemp->GetMaximumBin();  //Finds bin w/ max counts (can be dubious, so use TSpectrum)
  binMaxCnt_value = (Double_t) htemp->GetXaxis()->GetBinCenter(binMaxCnt); //if the bin number is known and the bin value (in x-axis units) is wanted
  binMaxCnt_counts = (Int_t) htemp->GetBinContent(binMaxCnt); //Finds counts within particular bin

  //---------TSpectrum Peak Finding Parameters--------
  if (ShowFit) NPeaks = s->Search(htemp, 2, "goff", 0.5); //opens a canvas (one time in a loop), even with:  s->Search(htemp, 2, "nodraw", 0.9);
  //else  NPeaks = s->Search(htemp, 2, "", 0.5);  //s->Search(htemp, 2, "nodraw", 0.9);

  //Npeaks = s->GetNPeaks(); //If using this, need pointer in declaration above: Int_t *NPeaks
  //N_peaks =  Npeaks[0];

  Peak = s->GetPositionX();
  PeakAmp = s->GetPositionY();

  for (int i=0; i<NPeaks; i++)
    {
      if (peak_pos_amp < PeakAmp[i])
	{
	    peak_pos_amp = PeakAmp[i]; //TSpectrum finds peak counts
	    peak_pos = Peak[i]; //TSpectrum finds pos. of peak in x-axis units
	}
    }
  peak_pos_bin = htemp->GetXaxis()->FindBin(peak_pos); //if the bin value (in x-axis units) is known and the bin number is wanted
  peak_pos_count = htemp->GetBinContent(peak_pos_bin);  //counts in particular bin

  //------------------------------------------------------------------------------------------------------------------  

  zero_value_bin = htemp->GetXaxis()->FindBin(0.0); //if the bin value (in x-axis units) is known and the bin number is wanted
  Nbins = htemp->GetSize() - 2; //total number of bins in histo
  zero_bin_value =  htemp->GetXaxis()->GetBinCenter(0); //if the bin number is known and the bin value (in x-axis units) is wanted.
  max_bin_value =  htemp->GetXaxis()->GetBinCenter(Nbins); //if the bin number is known and the bin value (in x-axis units) is wanted.
 
  int TS = 0;
  if (peak_pos >= zero_bin_value  &&  peak_pos <= max_bin_value)  //Make sure that TSpectrum peak is within histo range
    {                                                             //if not, use Par initial values from Basic Histo Peak Find
             
      TS=1; //for cout below                        
      low_limit = peak_pos - (0.1 * abs(max_bin_value-zero_bin_value)); //peakpos-10% of full range of histo
      high_limit = peak_pos + (0.1 * abs(max_bin_value-zero_bin_value)); //peakpos+10% of full range of histo
                                                       
      func = new TF1("func", "gaus");
      //func->FixParameter(1,0);
//.........这里部分代码省略.........

//.........这里部分代码省略.........
  //Rebinning the histograms based on the mean value...
  for (int p=0; p<nPMT; p++) {
    for (int i=0; i<nBinsXY; i++) {
      for (int j=0; j<nBinsXY; j++) {	

	hisxy[p][i][j]->GetXaxis()->SetRange(1,nBinHist);
	Double_t mean = hisxy[p][i][j]->GetMean();
	hisxy[p][i][j]->GetXaxis()->SetRange(0,nBinHist);
	double nGroup = 4.*mean/200.;
	nGroup = nGroup>1. ? nGroup : 1.;
	hisxy[p][i][j]->Rebin((int)nGroup);
	
      }
    }
  }

  // Extracting mean from 200 keV peak 

  // Define fit ranges
  double xLowBin[nPMT][nBinsXY][nBinsXY];
  double xHighBin[nPMT][nBinsXY][nBinsXY];
  double xLow[nPMT][nBinsXY][nBinsXY];
  double xHigh[nPMT][nBinsXY][nBinsXY];
  int maxBin[nPMT][nBinsXY][nBinsXY];
  double maxCounts[nPMT][nBinsXY][nBinsXY];
  double binCenterMax[nPMT][nBinsXY][nBinsXY];
  double meanVal[nPMT][nBinsXY][nBinsXY]; // Holds the mean of the distribution as defined by first fitting the peak
  double fitMean[nPMT][nBinsXY][nBinsXY]; // Holds the mean of the low energy peak
  double fitSigma[nPMT][nBinsXY][nBinsXY]; // Holds the sigma of the low energy peak
  double endpoint[nPMT][nBinsXY][nBinsXY]; // Holds the endpoint


  /////// First determine roughly where the low energy peak is
  TSpectrum *spec;

  for (int p=0; p<nPMT; p++) {
    for (int i=0; i<nBinsXY; i++) {
      for (int j=0; j<nBinsXY; j++) {	
	
	double r = sqrt(power(posmap.getBinCenter(j),2)+power(posmap.getBinCenter(i),2));
	
        // Find bin with maximum content
	hisxy[p][i][j]->GetXaxis()->SetRange(2,nBinHist);
        maxBin[p][i][j] = hisxy[p][i][j]->GetMaximumBin();
        maxCounts[p][i][j] = hisxy[p][i][j]->GetBinContent(maxBin[p][i][j]);
        binCenterMax[p][i][j] = hisxy[p][i][j]->GetBinCenter(maxBin[p][i][j]);
	
	  
	if (r<=(50.+2*xyBinWidth))
	      {
		spec = new TSpectrum(20);
		Int_t npeaks = spec->Search(hisxy[p][i][j],1.5,"",0.5);
		
		if (npeaks==0)
		  {
		    cout << "No peaks identified at PMT" << p << " position " << posmap.getBinCenter(i) << ", " << posmap.getBinCenter(j) << endl;
		  }
		else
		  {
		    Float_t *xpeaks = spec->GetPositionX(); // Note that newer versions of ROOT return a pointer to double...
		    TAxis *xaxis = (TAxis*)(hisxy[p][i][j]->GetXaxis());
		    Int_t peakBin=0;
		    Double_t BinSum=0.;
		    Double_t BinSumHold = 0.;
		    Int_t maxPeak=0.;
		    for (int pk=0;pk<npeaks;pk++) {

void fit(){
  int range = 1e4;

  TFile *f = new TFile("hgrroot.root");
  TCanvas *ca = new TCanvas("ca","ca",0,0,1400,900);
  ca->Divide(4,4);
  TH1F* h[7][4];
  TF1* fu[7][4][2];
  TF1* fus[7][4][2][3];
  double res[7*4];
  double det[7*4];
  double slope[7*4];
  double offset[7*4];
  double resolution =5;
  TEnv* output = new TEnv("corecal.dat");

  for(int d=0;d<7;d++){
    for(int c=0;c<4;c++){
      h[d][c] = (TH1F*)f->Get(Form("htraceen_b%d_c%d_cr%d_d%d",0,9,c,d+1));
      ca->cd(1+c*4);
      //h[d][c]->GetXaxis()->SetRangeUser(100,4000);
      TSpectrum *sp = new TSpectrum(2,resolution);
      sp->SetResolution(resolution);
      Int_t nfound = 0;
      nfound = sp->Search(h[d][c],resolution,"nobackground",0.5);
      Float_t *xpeaks = sp->GetPositionX();
      Float_t *ypeaks = sp->GetPositionY();
//       for(int p=0;p<nfound;p++){
// 	cout << xpeaks[p] << "\t" << ypeaks[p] << endl;
//       }
      if(nfound!=2){
	cout << "Found " << nfound << " peaks in spectrum, too many, aborting" << endl;
	continue;
      }
      h[d][c]->DrawCopy();
      //check if first peak is lower in energy, otherwise swap them
      if(xpeaks[0]>xpeaks[1]){
	Float_t temp = xpeaks[1];
	xpeaks[1] = xpeaks[0];
	xpeaks[0] = temp;
	temp = ypeaks[1];
	ypeaks[1] = ypeaks[0];
	ypeaks[0] = temp;
	
      }

      for(int p=0;p<nfound;p++){
	ca->cd(1+c*4+1+p);
	h[d][c]->GetXaxis()->SetRangeUser(xpeaks[p]-range,xpeaks[p]+range);
	h[d][c]->DrawCopy();
	fu[d][c][p] = new TF1(Form("fcore_d%d_c%d_p%d",d,c,p),fgammagaussbg,xpeaks[p]-range,xpeaks[p]+range,6);
	fu[d][c][p]->SetLineColor(3);
	fu[d][c][p]->SetLineWidth(1);
	fu[d][c][p]->SetParameter(0,0);//bg const
	fu[d][c][p]->SetParameter(1,0);//bg slope
	fu[d][c][p]->SetParameter(2,h[d][c]->Integral(xpeaks[p]-range,xpeaks[p]+range));//norm
	fu[d][c][p]->SetParameter(3,xpeaks[p]);//mean
	fu[d][c][p]->SetParLimits(3,xpeaks[p]-500,xpeaks[p]+500);//mean
	fu[d][c][p]->SetParameter(4,100);//sigma
	fu[d][c][p]->SetParLimits(4,0.001,1000);//sigma
	fu[d][c][p]->SetParameter(5,h[d][c]->GetBinContent(h[d][c]->FindBin(xpeaks[p]-range)));//step
	
	h[d][c]->Fit(fu[d][c][p],"Rqn");
	fu[d][c][p]->Draw("same");


	fus[d][c][p][0] = new TF1(Form("fcore_d%d_c%d_p%d_bg",d,c,p),fgammabg,xpeaks[p]-range,xpeaks[p]+range,6);
	fus[d][c][p][1] = new TF1(Form("fcore_d%d_c%d_p%d_st",d,c,p),fgammastep,xpeaks[p]-range,xpeaks[p]+range,