maj

Stat/TP3/Exercice3_Substitution/c.py

+
+import math
+import string
+import pandas as pd
+import matplotlib.pyplot as plt
+import numpy as np
+import json
+import unidecode
+
+alphabet=string.ascii_lowercase
+ALPHABET=string.ascii_uppercase
+
+def is_upper(x):
+    for i in ALPHABET:
+        if x==i:
+            return True
+    return False
+
+def caesar_crypt(msg,decal):
+
+    new_alphabet=alphabet[decal:]+alphabet[:decal]
+    new_ALPHABET=ALPHABET[decal:]+ALPHABET[:decal]
+    msg_crypt=""
+    for i in msg :
+        if is_upper(i):
+            for y in range (0, len(ALPHABET)):
+                if i== ALPHABET[y]:
+                    msg_crypt+=new_ALPHABET[y]
+        else: 
+            for y in range (0, len(alphabet)):
+                if i== alphabet[y]:
+                    msg_crypt+=new_alphabet[y]
+    return msg_crypt
+
+
+def filetofreq(text,fichier):
+    new_file = open(fichier,"w")
+    dictionnary ={}
+    Alphabet=['a','b','c','d','e','f','g','h','i','j','k','l','m','n','o','p','q','r','s','t','u','v','w','x','y','z',' ']
+    with open(text, "r",encoding="UTF-8") as txt:
+        lignes = unidecode.unidecode(txt.read().lower()) 
+        for i in range(0,26):
+            freqlettre=lignes.count(Alphabet[i])/(len(lignes)-lignes.count(Alphabet[26]))*100
+            dictionnary[Alphabet[i]]=freqlettre
+    
+    json.dump(dictionnary,new_file)
+    new_file.close()   
+
+def bar_freq(fichier):
+    dictionary = json.load(open(fichier, 'r'))
+    xAxis = [key for key, value in dictionary.items()]
+    yAxis = [value for key, value in dictionary.items()]
+
+    fig, ax = plt.subplots()
+    ax.bar(xAxis,yAxis)
+    plt.tight_layout()
+    ax.set_title("Fréquence d'apparition d'une lettre")
+    plt.show()
+    
+
+def bar_freq1_freq2(fichier1,fichier2):
+    dictionary = json.load(open(fichier1, 'r'))
+    xAxis = [key for key, value in dictionary.items()]
+    yAxis = [value for key, value in dictionary.items()]
+    dictionary2 = json.load(open(fichier2, 'r'))
+    xAxis2 = [key for key,value in dictionary2.items()]
+    yAxis2 = [value for key, value in dictionary2.items()]
+
+    fig, ax = plt.subplots()
+    ax.bar(np.arange(len(xAxis))+0.2,yAxis,0.4,label="Freq 1")
+    ax.bar(np.arange(len(xAxis2))-0.2,yAxis2,0.4,label="Freq 2")
+    plt.tight_layout()
+    plt.legend()
+    ax.set_title("Fréquence d'apparition d'une lettre")
+    plt.show()
+
+def bar_compare_freq(fichier1,fichier2,decal):
+    dictionary = json.load(open(fichier1, 'r'))
+    xAxis = [key for key, value in dictionary.items()]
+    yAxis = [value for key, value in dictionary.items()]
+    dictionary2 = json.load(open(fichier2, 'r'))
+    xAxis2 = [key for key,value in dictionary2.items()]
+    yAxis2 = [value for key, value in dictionary2.items()]
+    yAxis2=yAxis2[decal:]+yAxis2[:decal]
+
+    fig, ax = plt.subplots()
+    ax.bar(np.arange(len(xAxis))+0.2,yAxis,0.4,label="Freq 1")
+    ax.bar(np.arange(len(xAxis2))-0.2,yAxis2,0.4,label="Freq 2")
+    plt.tight_layout()
+    plt.legend()
+    ax.set_title("Fréquence d'apparition d'une lettre")
+    plt.show()
+
+def manualdecrypt(fichier1,fichier2):
+    for i in range(1,26):
+        print("décalage :",i) 
+        bar_compare_freq(fichier1,fichier2,i)
+
+
+def moy(x):
+    add=0
+    for e in x:
+        add+=e
+    return add/len(x)
+
+def covar(X,Y):
+    som=0
+    if len(X) == len(Y):
+        for i in range(0,len(X)):
+            a=(X[i]-moy(X)) * (Y[i]-moy(Y))
+            a/=len(X)
+            som+=a
+            a=0
+        return som
+    else:
+        return "error"
+
+def variance(X):
+    a=0
+    for i in range(0,len(X)):
+        a+=(X[i]-moy(X))**2
+    return a*(1/len(X))
+
+def coeff_corr(X,Y):
+    return covar(X,Y)/(math.sqrt(variance(X)*variance(Y)))
+
+def correlation_freqs(fichier1,fichier2):
+    dictionary = json.load(open(fichier1, 'r'))
+    yAxis = [value for key, value in dictionary.items()]
+    dictionary2 = json.load(open(fichier2, 'r'))
+    yAxis2 = [value for key, value in dictionary2.items()]
+
+    corr=-1
+    decal=0
+
+    for i in range(1,26):
+        yAxis2=yAxis2[1:]+yAxis2[:1]
+        print("décalage :",i," crypt correlation :",coeff_corr(yAxis,yAxis2))
+        if corr<coeff_corr(yAxis,yAxis2):
+            decal=i
+            corr=coeff_corr(yAxis,yAxis2)
+    print("Décalage de décryptage :", decal)
+    return decal
+
+def decrpt(fichier):
+    filetofreq(fichier,"freq.json")
+    with open(fichier,"r",encoding="UTF-8") as txt:
+        new_file = open("decrypted_with_caesar.txt","w")
+        str=txt.read()
+        strdecry=caesar_crypt(str,-correlation_freqs("freq_lettre_fr.json","freq.json"))
+        new_file.write(strdecry)
+        new_file.close()
+
+def main():
+    filetofreq("encrypted_with_caesar.txt","freq1.json")
+    
+    #bar_freq("freq_lettre_fr.json")
+    #bar_freq1_freq2("freq_lettre_fr.json","freq1.json")
+    #bar_compare_freq("freq_lettre_fr.json","freq1.json",3)
+    #manualdecrypt("freq_lettre_fr.json","freq1.json") #ici c'est 17 de décalage
+    correlation_freqs("freq_lettre_fr.json","freq1.json")
+    decrpt("encrypted_with_caesar.txt")
+        

Stat/TP3/Exercice3_Substitution/decrypted_with_caesar.txt

+OCEMCJMQRCOIIPJOPEFONOTMISPIPRORRASONMCBOLSOIVAOPESPCAPEJPCWEPRFMANNOACFIMKMCTNMAFPINONNPETOCNNMCRPENOFFAQOCENAINMCFMRMGYMCPRFIPEACIMSOCPRRMAJIPEPRRAESOPNPRCQNOPNPNNOPEVAACPSZIMWRPMGMCKANPRZAEOPEOEMAEPCNEOCENAIACSMETMCEPRPWCMIOPERONPWCPKPGOEPMCPROZOCTMCCONMCIMSOCNAINMCRPEPRORROONMCROJOZMPRSMAPRROACWOCEVAPRFONNONAINMCKIMCENAINMCGMANMCFMARNZOEEOPEEIMFKMIEPROJOPEGYOATPRMAJIPENMCJONPNEONNGIAEOROCAPEPRKOPNOPETMAUACZIAPEPCTPNEPCGESMCEOPETAKOAZMAIWACGOIPRRMCFRANRMAITVAACWRONFRANNMAITVAACEMGNPCFRANFIMKMCTVAACZMAITMCCMCRMPCNMCCOEIMPNGMAFNTAGOCORNOPCESOIEPCACGROFMEPNFROPCEPKNPWCOROPEACGYOROCTVAPFONNOPENAIROZOEEOCETAJONPNEONACOCPSOROAEYMIOUPCTPWMOROPWAPRRMCNOKIOCCPACGOKOITCPACGYOIOCGMCSOPNFRAEMEACOIEPNMCNOJOCGOPEEIOPCOCEACZIPCTORKOPRNOFFIMGYOJMAROCEROFROEPITACGMAFJPKSOPNROCPSORFIPENMCJMRTPNFOIOPNNOCETOCNROCAPEOJOCEVAPROPEFARONNOPRRPIPREOFMEOTACTMPWEACOPISOIEPORNAIRMZRMCWGYONNPNTAJONPNEONPRMAJIPENMCKIPWMSAIORPRFIPETAROPEKIMPTPRZAEACWIOCTZMRPRNOFOPNOPEPRNONNPENAINMCGMNQPRFIPEACBMAICORVAPRFOIGMAIAETACOPITPNEIOPEPRORRASOACGPWOIPRRMVAPRKASOBANVAOAZMAEVAMPVAPREIMAJOENMCFOIKASPIIPEOCEPREMANNOPRSPEROIOTPMACOPIOKIMGAZOPCKAENAPJPTACZMNEMCFAPNACEOCWMFAPNACKMUEIMEFAPNACGMEPRRMCSPNOAWMAETABMAITAEIMCGGYOCEOTAROCLSOCCZOIZOIOACSOTIPWORTOIOWMCNEPGYIOCTORRACOPITOPTOPRTAENONNMAFPIACPCNEOCEGOIPRNAINOAEONMATOPCROIOTPMOCCMCGOPEJMPGPCMNPCKMISOEPMCNPRCQOJOPEOAGACKOPEPSFMIEOCEOJORFOIOPNMRPCOAWAIOEPMCTACFMCEOJOPEKOPEJPCWEGPCVSMIENOLAIPGYCMIMTMSNPYOCMAXKOPNOPENOJMPIVAPRCPIOPEFONOHONYPCWEMCOSOEPWCMCFMSFPTMAFIMFMNOPEOAUNQCTPGOENRMIWOCPNOEPMCTACNEOEAVAMNMGPORSOPNKOPNOPEGYMAZROCGOAZPOKIOGMCKRPENIOGPOAUOGMCOXIQMCFOIROPETACFAENGYACEQFYMCNOZOEEOPENAICOWONOXPEOCTPNVAACMAIOWOCOABMRPNAICMSTOSOCTONOCCMCGOPENAIEIPNEOCTOGACYOTMCEMCIOFOEIPOPEROFMFAROEPMCFOIOJPMCNGOIWMNOIMROCTWOIIMNFMAIKPCPITOCNACSOEGYGMSFEOCEFMAIROTOJPNGAFNOCEOCOOJOPEZOEEATOISMCNPUEIMPNACNPUEIMPNNPUTPUYAPEYAPENPUPRGMAFOROIOTPM
\ No newline at end of file

Stat/TP3/Exercice3_Substitution/freq.json

+{"a": 2.614727854855923, "b": 0.05336179295624333, "c": 6.189967982924227, "d": 1.44076840981857, "e": 1.4941302027748131, "f": 0.16008537886872998, "g": 8.164354322305229, "h": 7.790821771611527, "i": 13.927427961579507, "j": 11.419423692636073, "k": 0.42689434364994666, "l": 6.08324439701174, "m": 1.9743863393810033, "n": 3.2017075773745995, "o": 0.5336179295624333, "p": 0.9071504802561365, "q": 1.4941302027748131, "r": 0.16008537886872998, "s": 1.0138740661686232, "t": 1.1739594450373532, "u": 8.591248665955176, "v": 0.26680896478121663, "w": 9.818569903948772, "x": 0.0, "y": 8.057630736392744, "z": 3.04162219850587}
\ No newline at end of file

Stat/TP3/Exercice3_Substitution/freq_lettre_fr.json

+{"a": 8.4, "b": 1.1, "c": 3.0, "d": 4.2, "e": 17.3, "f": 1.1, "g": 1.3, "h": 0.9, "i": 7.3, "j": 0.3, "k": 0.1, "l": 6.0, "m": 3.0, "n": 7.1, "o": 5.3, "p": 3.0, "q": 1.0, "r": 6.6, "s": 8.1, "t": 7.1, "u": 5.7, "v": 1.3, "w": 0.1, "x": 0.4, "y": 0.3, "z": 0.1}

Stat/TP3/Exercice3_Substitution/s.py

+import json
+import string
+import random
+import numpy as np
+import unidecode
+import c 
+
+   
+
+def crypt(msg) :
+    new=msg.replace(' ','')
+    new = unidecode.unidecode(new)
+    new=new.upper()
+    c=list(new)
+    random.shuffle(c)
+    return ''.join(c)
+
+def chiffre(msg,key):
+    ALPHABET=string.ascii_uppercase
+    new=msg.lower()
+    new=new.replace(' ','')
+    new = unidecode.unidecode(new)
+    new=new.upper()
+    for i in range(0,len(new)):
+        for y in range(0,len(ALPHABET)):
+            if new[i]==ALPHABET[y]:
+                new = new[:i] + key[y] + new[i+1:]
+                break
+    return new
+
+def dictquad(text):
+    new_file = open("dictquad.json","w")
+    with open(text, "r",encoding="UTF-8") as txt:
+        lignes = txt.readlines()
+        for ligne in lignes:
+            DICT_QUAD[ligne[0:4]]=int(ligne[5:].replace("\n","")) 
+        
+    json.dump(DICT_QUAD,new_file)
+    new_file.close()  
+
+def logscore(chain):
+    logsum = 0
+    min_freq = 1e-10
+    for i in range(len(chain)-3): 
+        logsum += np.log10(DICT_QUAD.get(chain[i:i+4], min_freq))
+    return -logsum
+    
+
+DICT_QUAD={}
+key=crypt(string.ascii_uppercase)
+print(key)
+
+#print(chiffre('abc',key))
+#c.decrpt("texte_chiffre_substitution.txt")
+dictquad("quadrigramme.txt")
+
+with open("texte_chiffre_substitution.txt", "r",encoding="UTF-8") as txt:
+    lignes = txt.read()
+    oldkey=[]
+    ls=999999999999999.9
+    ret=lignes
+    new=chiffre(lignes,key)
+    
+    for i in range(0,1000000):
+        key=crypt(string.ascii_uppercase)
+        #print(oldkey)
+        if key not in oldkey:
+            oldkey+=key
+            new=chiffre(lignes,key)
+            #print("new",logscore(new))
+            #print("ret",logscore(ret))
+            if logscore(ret)>logscore(new):
+                ls=logscore(new)
+                ret=new
+            #print("best ls",ls)
+    print("ret",ret)
+