1. # coding=utf-8
2. import json
3. import io
4. import re
5.  
6.  
7. ## Black -> White -> ❶ -> (1) (see hon)
8.  
9. # words to test: are, suru, naru
10.  
11. def process_kanji (dfn):
12.     # gets string and returns list of ■一■-level dfns processed
13.     #re_obj = re.split(ur"■[一二三四五六七八九十]■", dfn)
14.     re_obj = re.split(r'\n■[一二三四五六七八九]■', dfn, re.UNICODE)
15.     if(len(re_obj) > 2):
16.         common = re_obj.pop(0) + "\n"
17.         re_obj = map (lambda x: common + x, re_obj)
18.  
19.     return re_obj
20.  
21. def process_white (dfn_lst):
22.     # gets lst of dfns and returns (possibly bigger) final lst of dfns
23.     final_lst = []
24.     for s in dfn_lst:
25.         re_obj = re.split(r'\n□[一二三四五六七八九]□', s, re.UNICODE)
26.         if (len(re_obj) > 2):
27.             common = re_obj.pop(0) + "\n"
28.             re_obj = map( lambda x: common + x, re_obj)
29.         final_lst.extend(re_obj)
30.     return final_lst
31.  
32. def process_black_ball (dfn_lst):
33.     # gets lst of dfns and returns (possibly bigger) final lst of dfns
34.     final_lst = []
35.     for s in dfn_lst:
36.         re_obj = re.split(r'\n[❶❷❸❹❺❻❼❽❾❿⓫⓬⓭⓮⓯⓰⓱⓲⓳⓴]', s, re.UNICODE)
37.         if (len(re_obj) > 2):
38.             common = re_obj.pop(0) + "\n"
39.             re_obj = map( lambda x: common + x, re_obj)
40.         final_lst.extend(re_obj)
41.     return final_lst
42.  
43. def process_numbers (dfn_lst):
44.     # gets lst of dfns and returns (possibly bigger) final lst of dfns
45.     final_lst = []
46.     for s in dfn_lst:
47.         re_obj = re.split(r'\n（[１２３４５６７８９０]+）', s, re.UNICODE)
48.         if (len(re_obj) > 2):
49.             common = re_obj.pop(0) + "\n"
50.             re_obj = map( lambda x: common + x, re_obj)
51.         final_lst.extend(re_obj)
52.     return final_lst
53.  
54.  
55.  
56. def process_dfn (dfn):
57.     # receives raw string and returns LIST of strings (one for each definition)
58.     #dfn = dfn.replace('{（１）}','-- (1) --')
59.     #dfn = dfn.replace('〔（１）','〔 (1) ')
60.  
61.     # process "■一■" and similar
62.     dfn = process_kanji (dfn) # returns list of strings
63.  
64.     dfn = process_white (dfn) # returns list of strings
65.  
66.     dfn = process_black_ball(dfn) # returns list of strings
67.    
68.     # process "（１）" and similar
69.     lst = process_numbers (dfn)
70.  
71.     return lst
72.  
73.  
74. number = 1
75.  
76. names = []
77.  
78. for number in range(1,34):
79.         name = 'term_bank_' + str(number) + '.json'
80.         names.append(name)
81.  
82.  
83. for file_name in names:
84.     print(file_name)
85.  
86.     glob = []
87.  
88.     with open(file_name) as json_file:
89.         data = json.load(json_file)
90.  
91.     for elem in data:
92.         lst = elem[5]
93.  
94.         lst_dfn = []
95.  
96.         for raw_dfn in lst:
97.             lst_dfn.extend(process_dfn(raw_dfn))
98.  
99.         for dfn in lst_dfn:
100.             current_elem = []
101.  
102.             current_elem.append( elem[0] )
103.             current_elem.append( elem[1] )
104.             current_elem.append( elem[2] )
105.             current_elem.append( elem[3] )
106.             current_elem.append( elem[4] )
107.             current_elem.append( [dfn] )
108.             current_elem.append( elem[6] )
109.             current_elem.append( elem[7] )
110.  
111.             glob.append( current_elem )
112.  
113.  
114.     with io.open(str(file_name), 'w', encoding='utf-8') as f:
115.         json.dump(glob, f, ensure_ascii=False)
116.