datestex/generate_testcases.py

import random
#from flask import
#
class Koeln():
    # Konfig
    KOELN = {
        "00" : "AI,AY,AU,EI,EY,EU,OI,Ä,Ö,Ü", # vokale
        "40" : "CA,CO,CU",
        "80" : "CE,CI,CY",
        "88" : "SCH,ZC,DC,DS,DZ,TC,TS,TZ",
        "48" : "X",
        "0" : "AA,A,EE,IE,E,I,J,OU,OO,O,U,Y", # vokale
        "1" : "B,P",             # labiale
        "2" : "D,T",
        "3" : "PH,F,V,W",        # labiale
        "4" : "G,K,Q",
        "5" : "L",
        "6" : "M,N",
        "7" : "R",
        "8" : "S,Z"
    }
    def __init__(self):
        self.init = "ja"
    def getPhonWertMitVokal(self, wort):
        out = ""
        txt = wort.upper()
        while (len(txt) > 0):
            print("while "+txt)
            a = self.getFirstPhonem(txt)
            if (a[0] == "99"):
                txt = txt[1:]
            else:
                out = out+a[0]
                txt = txt[len(a[1]):]
            print("while - " + txt + " "+out)
        print("-------- RETURN - " + txt + " " + out)
        return out

    def getPhonWert(self, wort):
        out = self.getPhonWertMitVokal(wort)
        for x in ["11", "22", "33", "44", "55", "66", "77", "88"]:
            out = out.replace(x, x[0:1])
        out = out.replace("0", "")
        print("-------- RETURN - " + wort + " " + out)
        return out

    def getAehnlicherPhon(self, wort):
        out = [wort]
        gen = [""]
        txt = self.getPhonWertMitVokal(wort)
        while (len(txt) > 0):
            for k in ["00", "40", "80", "88"]:
                print ("+++ Umlaute +++")
                if (txt[0:2]==k):
                    neu = []
                    chc = []
                    for y in gen:
                        c = "XXX"
                        for i in range(0, len(self.KOELN[k])/2):
                            c = random.choice(self.KOELN[k])
                            while not c in chc:
                                print (" choice "+k+" = "+c+" :: "+self.KOELN[k])
                                neu.append(y+c)
                                c = random.choice(self.KOELN[k])
                                break

            for x in ["11", "22", "33", "44", "55", "66", "77"]:
                if (txt[0:2] == x):
                    pass
    def getAehnlicherPhonem(self, txt):
        gen = [""]
        phonems = {
            "umlaut" : ["00", "40", "80", "88"],
            "doppelk" : ["11", "22", "33", "44", "55", "66", "77"],
            "literal" : ["0","1","2","3","4","5","6","7","8"]
        }
        while (len(txt) > 0):
            for p in phonems:
                for k in phonems[p]:
                    # print ("+++ Umlaute +++"+txt+"--"+k)
                    if (txt[0:len(k)]==k):
                        print("+++ Umlaute +++" + txt + "--" + k)
                        neu = []
                        chc = []
                        #print("+++ Umlaute +++" + txt + "--" + k)
                        c = "XXX"
                        for i in range(0, int(len(self.KOELN[k].split(","))/3 +1)):
                            status = 1
                            c = random.choice(self.KOELN[k].split(","))
                            while ((status==1) and (not c in chc)):
                                #print (str(i)+" choice "+k+" = "+c)
                                for y in gen:
                                    neu.append(y+c)
                                chc.append(c)
                                c = random.choice(self.KOELN[k].split(","))
                                status=2
                        gen = neu
                        print(neu)
                        txt=txt[len(k):]
            # txt=""
        print(gen)

    def getFirstPhonem(self, txt):
        for k in self.KOELN:
            print(k + " wert " + self.KOELN[k])
            vv = self.KOELN[k].split(",")
            for x in vv:
                if (txt.find(x) == 0):
                    return (k, x);
        return ("99", "H")

# Funktionen
def generiereAehnlicheWerte(feld, basis):
    print ("generiereAehnlicheWerte("+feld+", "+basis)
    if (k == "FAMNAM"):
        koeln.getPhonWert(basis)
# Parameter
koeln = Koeln()
job = {}
job["option"] = {}
job["option"]["regression"] = "g" # kreuz | orthogonal
job["option"]["clusteranz"] = 2 #
content = {}
content["Tabelle"] = {}
content["Tabelle"]["Person"] =  {}
content["Tabelle"]["Person"]["FAMNAM"] = {}
content["Tabelle"]["Person"]["FAMNAM"]["option"] = "aehnlich"
content["Tabelle"]["Person"]["FAMNAM"]["gruppe"] = 1
content["Tabelle"]["Person"]["FAMNAM"]["basis"] = ["Müller", "Schmidt"]
content["Tabelle"]["Person"]["GESCHLECHT"] = {}
content["Tabelle"]["Person"]["GESCHLECHT"]["option"] = "diff"
content["Tabelle"]["Person"]["GESCHLECHT"]["gruppe"] = 1
content["Tabelle"]["Person"]["VORNAM"] = {}
content["Tabelle"]["Person"]["VORNAM"]["option"] = "beliebig"
content["Tabelle"]["Person"]["GEBDAT"] = {}
content["Tabelle"]["Person"]["GEBDAT"]["option"] = "identisch"
content["Tabelle"]["Person"]["GEBDAT"]["basis"] = ["19920123", "20210908"]

gruppen = []
for i in range(0, job["option"]["clusteranz"]):
    print("index "+str(i))
    for k in content["Tabelle"]["Person"]:
        print("key "+ k)
        if content["Tabelle"]["Person"][k]["option"] == "aehnlich":
            if (content["Tabelle"]["Person"][k]["basis"]):
                werte = generiereAehnlicheWerte(k, content["Tabelle"]["Person"][k]["basis"][i])

koeln.getAehnlicherPhonem(("00"))
koeln.getAehnlicherPhonem(("3"))
koeln.getAehnlicherPhonem(("8"))
koeln.getAehnlicherPhonem(("886022"))