EWP-BplusToKstMuMu-AngAna/Code/Selection/ComparisonTool/Utils.py

from ROOT import gROOT, gDirectory, gStyle, TChain, TTree, TAxis, TH1D
import subprocess   
import time
import numpy as np
import re #for splitting with multiple deliminers
from array import array
import numexpr
import argparse
import sys
import os
import re 
from Globals import *

'''
Tools used in compareUlitmate.py
'''
#Define a list of variables that are considered integers
intVarsList = ["nTracks"
                ,"nLongTracks"
                ,"nVeloTracks"
                ,"nTTracks"
                ,"nDownstreamTracks"
                ,"nUpstreamTracks"
                ,"nCandidate"
                ,"totCandidates"
                ,"nSPDHits"                
                ,"nPV"              
                ,"B_plus_ID"        
                ,"B_plus_TRUEID"          
                ,"K_star_plus_ID"  
                ,"K_plus_ID"
                ,"pi_zero_resolved_ID"
                ,"mu_plus_ID"
                ,"mu_minus_ID"
                ,"Polarity"
                ,"nDiMuonMassBin"
                ,"TMedBKGCAT"
            ]

def expressionList():
    return ["Sqrt"
            ,"Abs"
            ,"Sin"
            ,"Cos"
            ,"Exp"
            ,"Log"
            ]


#Define a function that returns true for variables that are integers
def isInt(variable=""):
    if variable in intVarsList:
        return True
    else:
        return False


def treeName(MC=False, TM=False, Preselected=False):
    if (TM or (MC and Preselected)): #TODO
        return "DecayTreeTruthMatched"
    elif (not Preselected):
        return "b2KstKs0pimumu_Tuple" if KshortChannel() else "b2KstKpi0mumuResolvedTuple/DecayTree"
    else:
        return "DecayTree"

def getOptionsDictionary(year, Run, magnet, MC, TM, ReferenceChannel, PHSP, Preselected,
                            BDTed, sWeighted, bWeighted, b2Dweighted, weightBranch, KshortDecaysInVelo, UseLowQ2Range):
    dictionaryTmp = { 
        'year': year,
        'Run' : Run, 
        'magnet' : magnet, 
        'MC' : int(MC), 
        'TM' : int(TM), 
        'ReferenceChannel' : int(ReferenceChannel), 
        'PHSP' : int(PHSP), 
        'Preselected' : int(Preselected),
        'BDTed' : BDTed, 
        'sWeighted' : sWeighted, 
        'bWeighted' : bWeighted, 
        'b2Dweighted' : b2Dweighted, 
        'weightBranch' : weightBranch, 
        'KshortDecaysInVelo' : int(KshortDecaysInVelo), 
        'UseLowQ2Range' : UseLowQ2Range

    }
    return dictionaryTmp     

def checkYearSample(optionsDict):  #If wrong year/MC, skip
    if optionsDict['year'] not in yearList(): return False, "Wrong Year!"
    if (not optionsDict['MC']): return True, ""
    else:
        if (KshortChannel()):
            if (optionsDict['PHSP']): return True, ""
            elif (optionsDict['ReferenceChannel']):
                if (optionsDict['year'] < 2013 or optionsDict['year'] == 2016): return True, "" #Referenche channel available only for Run I
                else: return False, "RefChannel only available for Run I!"
            else: return True, "" 
        else:
            if (optionsDict['PHSP']): 
                if (optionsDict['year'] < 2018): return True, ""
                else: return False, ""
            elif (optionsDict['ReferenceChannel']):
                if (optionsDict['year'] < 2017): return True, "" #Referenche channel available only for Run I
                else: return False, "RefChannel only available for < 2017!"
            else:
                if (optionsDict['year'] > 2017):
                    return False, "No MC for 2018!"
                else: return True, ""

def getTreePath(optionsDict, verbose):
    #getPathForPython usage:
    #   getPathForPython command year Run magnet Preselected MC ReferenceChannel PHSP KshortDecayInVelo Selection UseLowQ2Range
    #It has to have everything to keep it easy
    #Get first file name
    filePath2 = ""
    command = "/home/lhcb/kopecna/B2KstarMuMu/code/ewp-Bplus2Kstmumu-AngAna/CodeForTests/CompareUltimate/getPathForPython " #TODO maybe replace by a current path
    if(optionsDict['BDTed']):
        options = "BDToutput %(year)i %(Run)i %(magnet)s 0 %(MC)r %(ReferenceChannel)r %(PHSP)r %(KshortDecaysInVelo)r 0 %(UseLowQ2Range)r" %optionsDict
        if verbose: print "[DEBUG]\t\tRunning a c++ script to get the name of the file to open: ", command + options
        #Read second line from the c++ script (first one is the Hello...)
        filePath = subprocess.check_output(command + options, shell=True,universal_newlines=True).split('\n')[1]

    elif (optionsDict['sWeighted'] or optionsDict['bWeighted'] or optionsDict['b2Dweighted']):
        options = "BDTinput %(year)i %(Run)i %(magnet)s 0 %(MC)r %(ReferenceChannel)r %(PHSP)r %(KshortDecaysInVelo)r 0 %(UseLowQ2Range)r" %optionsDict
        if verbose: print "[DEBUG]\t\tRunning a c++ script to get the name of the file to open: ", command + options
        #Read second line from the c++ script (first one is the Hello...)
        filePath = subprocess.check_output(command + options, shell=True,universal_newlines=True).split('\n')[1]
    else:         

        if (optionsDict['magnet']=="both"):
            #down
            options = "input %(year)i %(Run)i down %(Preselected)r %(MC)r %(ReferenceChannel)r %(PHSP)r %(KshortDecaysInVelo)r 0 %(UseLowQ2Range)r" %optionsDict
            if verbose: print "[DEBUG]\t\tRunning a c++ script to get the name of the file to open: ", command + options
            filePath = subprocess.check_output(command + options, shell=True,universal_newlines=True).split('\n')[1]
            #up
            options = "input %(year)i %(Run)i up %(Preselected)r %(MC)r %(ReferenceChannel)r %(PHSP)r %(KshortDecaysInVelo)r 0 %(UseLowQ2Range)r" %optionsDict
            if verbose: print "[DEBUG]\t\tRunning a c++ script to get the name of the file to open: ", command + options
            filePath2 = subprocess.check_output(command + options, shell=True,universal_newlines=True).split('\n')[1]
        else:    
            options = "input %(year)i %(Run)i %(magnet)s %(Preselected)r %(MC)r %(ReferenceChannel)r %(PHSP)r %(KshortDecaysInVelo)r 0 %(UseLowQ2Range)r" %optionsDict              
            if verbose: print "[DEBUG]\t\tRunning a c++ script to get the name of the file to open: ", command + options
            filePath = subprocess.check_output(command + options, shell=True,universal_newlines=True).split('\n')[1]
    return filePath, filePath2
     
def addToTChain(tree, optionsDict, verbose):
    filePath, filePath2 =  getTreePath(optionsDict=optionsDict, verbose=verbose)

    #Get first file name
    if(optionsDict['BDTed'] or optionsDict['sWeighted'] or optionsDict['bWeighted'] or optionsDict['b2Dweighted']):         
        print "[INFO]\tOpening", filePath
        tree.Add(filePath)
    else:
        if (optionsDict['magnet']=="both"): 
            print "[INFO]\tOpening", filePath
            tree.Add(filePath)
            print "[INFO]\tOpening", filePath2
            tree.Add(filePath2)
        else: 
            print "[INFO]\tOpening", filePath
            tree.Add(filePath)   
         

def addTMathTags(varName):    
    for expr in expressionList():
        varName = varName.replace(expr,"TMath::"+expr)  
    return varName        

def list_of_DTF_vars ():
  return [
        "_PT",
        "_PX",
        "_PY",
        "_PZ",
        "_PE",
        "_P"
        "_ETA" ,
        "_MM",
        "B_plus_M" ,
        "K_star_plus_M" ,
        "pi_zero_resolved_plus_M"    
    ]

def replace_variables_to_DTF(variable = ""):
        
    #Append "_DTF" to everything in the list above by 
    for var in list_of_DTF_vars():
        if (variable.find(var) != -1):
            positions = [i.start() for i in re.finditer(var, variable)]
            #print positions
            for i in range(len(positions)):
                pos=positions[i]
                variable =  variable[:pos+len(var)] + "_DTF" + variable[pos+len(var):]
                positions = [x+4 for x in positions]

    
    #Remove false DTF variales, such as MinIP (searching for _M)
    removeList = re.findall('DTF[A-Z]', variable) 
    #print removeList
    for remove in removeList:
        positions = [i.start() for i in re.finditer(remove, variable)]
        for i in range(len(positions)):
            pos=positions[i]
            variable =  variable[:pos-1] + variable[pos+3:]
            positions = [x-4 for x in positions]
        
    return variable
    

def getListOfUsedVariables(variable, cut):
    tmp = str(variable) + "|" + str(cut) #| is there as a separator

    tmp = tmp.replace("gamma1","gammaa") 
    tmp = tmp.replace("gamma2","gammab")  #Protect from removing numbers

    tmp = re.sub(r'[0-9]+', '', tmp) #remove numbers 

    tmp = tmp.replace("gammaa","gamma1") 
    tmp = tmp.replace("gammab","gamma2")  #Protect from removing numbers

    tmp = tmp.replace("CHI","CHI2")  #check CHI2 numbers  
    tmp = tmp.replace("DOCA","DOCA1")  #check CHI2 numbers     
    tmp = tmp.replace("Q","Q2")  #check Q2 numbers    #TODO
    tmp = tmp.replace(".","") #remove decimals left from numbers
    tmp = tmp.replace("-","|") #remove minus from negative numbers, keep an operator

    tmp = tmp.replace("TMath::","") #remove functions
    
    for expr in expressionList():
        tmp = tmp.replace(expr,"") #Remove expressions
        tmp = tmp.replace(expr.lower(),"") #remove expressions without capital

    tmp = tmp.replace("<=","<")
    tmp = tmp.replace(">=",">")
    tmp = tmp.replace("==","=")
    tmp = tmp.replace("&&","&")
    tmp = tmp.replace("||","|")
    tmp = tmp.replace("!","")

    #tmp = re.sub(r"\s+", "", tmp) #Removes whitespaces, s matches all whitespaces
    branchList = re.split('[+ - * / ( ) < > = & | ]', tmp)
    branchList = filter(None, branchList) #removes empty strings
    branchList = np.unique(branchList) #Removes duplicates
   
    return branchList


def evaluateCut(cut,variablesDict):
     #TODO: should be easy to let it be able take operations into account
    if (cut == ""): return True
    tmp = str(cut)
    for varName,varValue in variablesDict.items():
        tmp = tmp.replace(str(varName),str(varValue[0]))
    tmp = tmp.replace ("||"," or ")
    tmp = tmp.replace ("&&"," and ")
    tmp = tmp.replace ("="," == ")
    tmp = tmp.replace ("! =="," != ") #in case 
    #print "[DEBUG]\t\t", tmp
    return eval(tmp)

def makeVariablePythonFriendly(variable):
    tmp = variable
    tmp = tmp.replace("TMath::","") #remove ROOT tags
    tmp = tmp.lower() #replaces e.g. Log by log and so on
    return tmp

def evaluateVariable(variable,variablesDict):
    tmp = variable
    for varName,varValue in variablesDict.items():
        tmp = tmp.replace(str(varName),str(varValue[0]))
    tmp = makeVariablePythonFriendly(tmp)
    return numexpr.evaluate(tmp).item()

def evaluateWeight(variablesDict,optionsDict): #TODO
    if (optionsDict["sWeighted"]): return variablesDict["N_Bplus_sw"][0]
    elif (optionsDict["bWeighted"]): return variablesDict["weight_nLongTracks"][0]
    elif (optionsDict["b2Dweighted"]): return variablesDict["weight2D_nLongTracks"][0]
    else: return 1.0


def datasetTag(optionsDict):
    tag = ""
    if (optionsDict['MC']):
        if (optionsDict['ReferenceChannel']):
           tag = "RefChannel"
        elif (optionsDict['PHSP']):
            tag = "PHSP"
        else:       
            tag = "MC"
        if (optionsDict['TM']):
            tag = tag + "_TM"    
    else:
        tag = "data"
    if not (optionsDict['Preselected']):
        tag = tag + "_Strip"
    return tag      

def weightTag(optionsDict):
    tag = ""
    if (optionsDict['sWeighted']): tag = "_sWeighted"
    if (optionsDict['bWeighted']): tag = "_1DWeight"
    if (optionsDict['b2Dweighted']): tag = "_2DWeight"
    return tag

def KshortDecaysInVeloTag(KshortDecaysInVelo=False):
    return "" if not KshortChannel() else ("_LL" if KshortDecaysInVelo else "_DD" )

def variableTag(variable=""):
    name = variable.replace("TMath::","")
    name = name.replace("(","")
    name = name.replace(")","")
    name = name.replace("|","")
    name = name.replace("/","_over_")
    name = name.replace("*","_x_")
    return name


def histName(variable, optionsDict):
    name = variableTag(variable)
    name = name.replace("TMath::","") #remove functions
    for expr in (expressionList()): #For whatever reason I have to but the brackets there
        name = name.replace(expr.lower(),expr) #Add capitals to functions
    
    name = "Hist_" + str(optionsDict['year']) + "_" + datasetTag(optionsDict) 
    name = name + KshortDecaysInVeloTag(optionsDict) + weightTag(optionsDict)
    return name

    
def stopWatch(value):
    '''From seconds to Days;Hours:Minutes;Seconds'''

    valueD = (((value/365)/24)/60)
    Days = int (valueD)

    valueH = (valueD-Days)*365
    Hours = int(valueH)

    valueM = (valueH - Hours)*24
    Minutes = int(valueM)

    valueS = (valueM - Minutes)*60
    Seconds = int(valueS)


    print Days,";",Hours,":",Minutes,";",Seconds


class ShowArgumentsParser(argparse.ArgumentParser):
    def error(self, message):
        sys.stderr.write('error: %s\n\n' %message)
        parser.print_usage(sys.stderr)
        sys.stderr.write('\n'+self.description)
        sys.exit(2)

def checkMCyear(year, ReferenceChannel, PHSP):
    if ((not KshortChannel()) and (not ReferenceChannel) and (not PHSP) and year == 2015): return 2016
    else: return year