from random import randint

inFile = open("sherlock-holmes.txt", "r")

wLens = {}
lFreq = {}
lCorel = {}

lFreqCheck = {}
def generateLetter(prev = " "):
    d = lCorel[prev]
    #print(d)
    #print("suma: " + str(sum(d.values())))
    r = randint(0, sum(d.values()))
    #print(r)
    for i in d.items():
        r -= i[1]
        if r <= 0:
            return i[0]
        
def generateWordLen():
    r = randint(0, sum(wLens.values()))
    #print(r)
    for i in wLens.items():
        r -= i[1]
        if r <= 0:
            return i[0]

def generateWords(wordCount):
    result = " "
    for i in range(wordCount):
        wordLen = generateWordLen()
        for j in range(wordLen):
            result += generateLetter(result[-1])
        result += " "
    return result  
        

lines = inFile.readlines(10000)
for l in lines:
    tmp = l.strip().replace('"', "").replace("'", "").replace('-', "").replace('.', "").replace('?', "").replace('!', "").replace(';', "").replace(',', "")
    if tmp != "":
        tmp = tmp.lower()
        wList = tmp.split(" ")
        for w in wList:
            if len(w) not in wLens.keys():
                wLens[len(w)] = 1
            else:
                wLens[len(w)] += 1
            for i in range(len(w)):
                if w[i] not in lFreq.keys():
                    lFreq[w[i]] = 1
                else:
                    lFreq[w[i]] += 1
                    
                if i > 0:
                    if w[i-1] not in lCorel.keys():
                        lCorel[w[i-1]] = {}
                    if w[i] not in lCorel[w[i-1]].keys():
                        lCorel[w[i-1]][w[i]] = 1
                    else:
                        lCorel[w[i-1]][w[i]] += 1
                else:  # first letter in a word
                    if " " not in lCorel.keys():
                        lCorel[" "] = {}
                    if w[i] not in lCorel[" "].keys():
                        lCorel[" "][w[i]] = 1
                    else:
                        lCorel[" "][w[i]] += 1
    #print(wList)
print("Word lentghs")
print(wLens)
print("Letter frequency")
#print(lFreq)
print("\t" + str(sorted(lFreq.items(), key=lambda s : -s[1])[:8]) + "...")
print("Letter corelation")
#print(lCorel)
for c in lCorel.keys():
    print("\t" + c + ":\t" + str(sorted(lCorel[c].items(), key=lambda s : -s[1])[:8]) + "...")
print("\n\n")
print(generateWords(40))