# coding=utf-8
import json
import io
import re

def process_kanji (dfn):
    # gets string and returns list of ■一■-level dfns processed
    #re_obj = re.split(ur"■[一二三四五六七八九十]■", dfn)
    re_obj = re.split(r'■[一二三四五六七八九]■', dfn, re.UNICODE)
    if(len(re_obj) > 2):
        common = re_obj.pop(0)
        re_obj = map (lambda x: common + x, re_obj)

    return re_obj

def process_numbers (dfn_lst):
    # gets lst of dfns and returns (possibly bigger) final lst of dfns
    final_lst = []
    for s in dfn_lst:
        re_obj = re.split(r'（[１２３４５６７８９]）', s, re.UNICODE)
        if (len(re_obj) > 2):
            common = re_obj.pop(0)
            re_obj = map( lambda x: common + x, re_obj)
        final_lst.extend(re_obj)
    return final_lst



def process_dfn (dfn):
    # receives raw string and returns LIST of strings (one for each definition)

    # process "■一■" and similar
    kanji_lst = process_kanji (dfn) # returns list of strings

    # process "（１）" and similar
    lst = process_numbers (kanji_lst)

    return lst


number = 1

names = []

for number in range(1,34):
        name = 'term_bank_' + str(number) + '.json'
        names.append(name)


for file_name in names:
    print(file_name)

    glob = []

    with open(file_name) as json_file:
        data = json.load(json_file)

    for elem in data:
        lst = elem[5]

        lst_dfn = []

        for raw_dfn in lst:
            lst_dfn.extend(process_dfn(raw_dfn))

        for dfn in lst_dfn:
            current_elem = []

            current_elem.append( elem[0] )
            current_elem.append( elem[1] )
            current_elem.append( elem[2] )
            current_elem.append( elem[3] )
            current_elem.append( elem[4] )
            current_elem.append( [dfn] )
            current_elem.append( elem[6] )
            current_elem.append( elem[7] )

            glob.append( current_elem )


    with io.open(str(file_name), 'w', encoding='utf-8') as f:
        json.dump(glob, f, ensure_ascii=False)