from pathlib import Path
import os
import re


class Markov:
    def __init__(self,filenames):
        self.__sources = filenames
        self.tree = dict()



    # Section from assignment 5
    def clean_line(self, line):
        # YOUR CODE HERE
        # Returning line without punctuation and numeric Symbols
        line = re.sub("'", '', line)
        line = re.sub("-", '', line)
        line = re.sub("/", '', line)
        # line = re.sub("\", '', line) ger något fel
        line = re.sub('[^A-Za-z.,]+', ' ', line)
        line = line.split()

        #print(line)
        #return []
        return line


    def text_gen(self):
        for fname in self.__sources:
            with open(fname, encoding='utf8', errors='ignore') as f:
                for line in f:

                    yield self.clean_line(line)



    def build_tree(self):
        """
        Build vocabulary of words from the provided text files
        """
        print("Du lyckas köra build_vocabulary")
        # YOUR CODE HERE

        # Go through all words and if they don't exist in self.__vocab = set() (are unique)
        # Add to self.__vocab = set()
        word_prev=""
        count=0
        factor=1
        bool=False
        for line in self.text_gen():
            for a in line:
                a=a.lower()
                if bool:
                    #print("a:   ",word_prev,"   b:      ",word )
                    #count+=1
                    if b not in self.tree:
                        count+=1
                        print(a," count: ", count)
                        self.tree[b] = dict()
                    #self.tree[a][b] = factor if b not in self.tree[a] else self.tree[a][b] + self.tree[a][b] * factor
                    if a not in self.tree[b]:
                        self.tree[b][a]= factor
                    else:
                        self.tree[b][a] = self.tree[b][a] + self.tree[b][a]*factor
                else:
                    bool=True
                    print("a:   ",a)
                    #self.tree[b] = dict()
                b=a




            '''for word in line:
                if bool:
                    #print("a:   ",word_prev,"   b:      ",word )
                    #count+=1
                    if word not in self.tree:
                        self.tree[word] = dict()

                    self.tree[word][word_prev] = factor if word_prev not in self.tree[word] else self.tree[word][word_prev] + self.tree[word][word_prev] * factor

                else:
                    bool=True
                word_prev=word'''
        #print(self.tree)
        #print(count)



                #if w not in self.__vocab:
                    #self.__vocab.add(w)
                    #print("Updated vocab with: ", w)

        #self.write_vocabulary()