sampler | Python Fiddle

# This code compares two lists of tokens and gets a sample of the largest one
# Javier Aguado-Orea

# First of all one has to say how many sampling processes are going to be run
numberIterations = raw_input('Number of iterations? ')
iteration = int(float(numberIterations))
print iteration
iterationb = iteration

for i in range(iteration):
    itestr = str(iterationb)
    outputdadfile = 'output_dad_'+itestr+'.txt'
    outputchifile = 'output_chi_'+itestr+'.txt'
    iterationb = iterationb - 1

outputdad = open(outputdadfile, 'w')
    outputchi = open(outputchifile, 'w')

# Then, it will open the input file as many interactions as indicated

input = open('input.txt', 'r')

# The input file is very restrictive for syntax
# A plain text file with the name input.txt is required
# The first line is a number: the amount of VERBS in the sample (e.g. 5)
# Then, a series of four lines has to be included per verb:
# First: number of tokens provided by the child with FIVE DIGITS (e.g. 00010)
# Second: number of tokens provided by the adult with FIVE DIGITS (e.g. 00100)
# Third: name of the verb (e.g. abrir)
# Fourth: twelve numbers, all five digits, six for child: 1s, 2s, 3s, 1p, 2p, 3p and six  for the adult: 1s, 2s, 3s, 1p, 2p, 3p (e.g. 00000 00000 00012 00000 00000 00000 00002 00000 00016 00001 00000 00001) " " "

# Now, the following code reads the 1st line (amount of verbs) and prints it to the console
# It's just to check that things are going fine with the input.txt file

verbos = input.readline() 
    print "verbos= "+ verbos
    verbos = int(verbos)

# -----

while (verbos > 0):
        verbos = verbos - 1

# Now, the following code reads the amount of tokens by adult and child
# it prints the difference, since sampling will change depending on that

totalchi = ""
        totaldad = ""
        childstr = input.readline(6)
        dadstr = input.readline(6)
        child = int(childstr)
        dad = int(dadstr)
        diferencia = dad - child
    
    # ATTENTION HERE, WHAT IF THE CHILD'S SAMPLE IS LARGER THAN THE ADULT ONE
    
        diferenciastr = str(diferencia)

# -----

# Now, it reads the name of the verb (e.g. abrir)

verbo = input.readline()
        verbo = verbo.rstrip('\n')
        msg = diferenciastr + " for " + verbo
        print msg

# -----

# Now, it reads all 12 numbers for the amount of tokens per form (1s, 2s, 3s, 1p, 2p, 3p)

# SIX FOR THE CHILD
# And creates a list of tokens per form, and adds them to "totalchi"

J1s = input.readline(6)
        J1s = int(J1s)
        list = ""
        if J1s > 0:
            VJ1s = verbo + "1s"
            for n in range(0, J1s):
                list = VJ1s + "," + list
        totalchi = totalchi + list

J2s = input.readline(6)
        J2s = int(J2s)
        list = ""
        if J2s > 0:
            VJ2s = verbo + "2s"
            for n in range(0, J2s):
                list =     VJ2s + "," + list
        totalchi = totalchi + list

J3s = input.readline(6)
        J3s = int(J3s)
        list = ""
        if J3s > 0:
            VJ3s = verbo + "3s"
            for n in range(0, J3s):
                list =     VJ3s + "," + list
        totalchi = totalchi + list

J1p = input.readline(6)
        J1p = int(J1p)
        list = ""
        if J1p > 0:
            VJ1p = verbo + "1p"
            for n in range(0, J1p):
                list =     VJ1p + "," + list
        totalchi = totalchi + list

J2p = input.readline(6)
        J2p = int(J2p)
        list = ""
        if J2p > 0:
            VJ2p = verbo + "2p"
            for n in range(0, J2p):
                list =     VJ2p + "," + list
        totalchi = totalchi + list   
    
        J3p = input.readline(6)
        J3p = int(J3p)
        list = ""
        if J3p > 0:
            VJ3p = verbo + "3p"
            for n in range(0, J3p):
                list =     VJ3p + "," + list
        totalchi = totalchi + list

totalchi = totalchi[:-1]

# SIX FOR THE ADULT (DAD)
# And creates a list of tokens per form, and adds them to "totaldad"

JD1s = input.readline(6)
        JD1s = int(JD1s)
        list = ""
        if JD1s > 0:
            VJD1s = verbo + "1s"
            for n in range(0, JD1s):
                list = VJD1s + "," + list
        totaldad = totaldad + list

JD2s = input.readline(6)
        JD2s = int(JD2s)
        list = ""
        if JD2s > 0:
            VJD2s = verbo + "2s"
            for n in range(0, JD2s):
                list =     VJD2s + "," + list
        totaldad = totaldad + list

JD3s = input.readline(6)
        JD3s = int(JD3s)
        list = ""
        if JD3s > 0:
            VJD3s = verbo + "3s"
            for n in range(0, JD3s):
                list =     VJD3s + "," + list
        totaldad = totaldad + list

JD1p = input.readline(6)
        JD1p = int(JD1p)
        list = ""
        if JD1p > 0:
            VJD1p = verbo + "1p"
            for n in range(0, JD1p):
                list =     VJD1p + "," + list
        totaldad = totaldad + list

JD2p = input.readline(6)
        JD2p = int(JD2p)
        list = ""
        if JD2p > 0:
            VJD2p = verbo + "2p"
            for n in range(0, JD2p):
                list =     VJD2p + "," + list
        totaldad = totaldad + list

JD3p = input.readline(6)
        JD3p = int(JD3p)
        list = ""
        if JD3p > 0:
            VJD3p = verbo + "3p"
            for n in range(0, JD3p):
                list =     VJD3p + "," + list
        totaldad = totaldad + list

totaldad = totaldad[:-1]

# It follows a bit of general information for the output file

verbo = verbo+'\n'
#In a previous version I was printing the verb before the actual tokens
#   outputdad.write(verbo)
#   outputchi.write(verbo)

# it converts strings to lists
        listatotaldad = totaldad.split(',')
        listatotalchi = totalchi.split(',')

# And it randomises the largest to the smallest size

import random

# Expected situation: more tokens in the adult sample
        if diferencia > 0 :

# First it writes the total items for the child
            totalchi = totalchi+'\n'
            outputchi.write(totalchi)
# Then it samples the adult tokens
            listacortadad = random.sample(xrange(len(listatotaldad)), diferencia)
            listacortadad = [listatotaldad[i] for i in listacortadad]
            cortadad = ','.join(listacortadad)
            outputdad.write(cortadad)

# Unexpected but possible: more tokens in the adult sample
        if diferencia < 0 :

# First it writes the total items for the adult
            totaldad = totaldad+'\n'
            outputdad.write(totaldad)
# Then it samples the adult tokens
            diferenciab = abs(diferencia)
            listacortachi = random.sample(xrange(len(listatotalchi)), diferenciab)
            listacortachi = [listatotalchi[i] for i in listacortachi]
            cortachi = ','.join(listacortachi)
            outputchi.write(cortachi)

# Same amount of tokens in both samples
        if diferencia == 0 :
            totalchi = totalchi+'\n'
            outputchi.write(totalchi)
            totaldad = totaldad+'\n'
            outputdad.write(totaldad)

espacio = '\n'+'\n'
        outputchi.write(espacio)
        outputdad.write(espacio)

else:
        print "END OF FILE"
print "END OF ITERATIONS"

Python Fiddle

Python Cloud IDE