chat-dataset / utils /07_parse.py
kirisame's picture
Add 07th VN parsing
8549fd4
raw
history blame
1.92 kB
from re import T
import parse
import argparse
from clean import clean
parser = argparse.ArgumentParser(description='Process some 07th stuff idk')
parser.add_argument('-f', '--file', type=str, help='file to process', required=True)
args = parser.parse_args()
with open(args.file, 'r') as f:
lines = f.readlines()
jap_characters = ['。', '「', '」', '?', '!', '、', '…', 'ー']
author = parse.parse('\ufeff== {0} ==\n', lines[0])[0]
with open('output.txt', 'w') as f:
for line in lines:
if line.startswith('\ufeff'):
continue
# check if line has jap_characters
if any(substr in line for substr in jap_characters):
continue
if line.startswith('\n==') or line.startswith('=='):
try:
author = parse.parse('== {0} ==\n', line)[0]
except:
author = 'Narrator'
continue
if line == '\n':
continue
line = line.rstrip()
if author == 'Narrator':
f.write(clean('[' + line.replace('\"', '').lstrip().rstrip() + ']').replace('\"', '')+'\n')
else:
txt = line.replace('\"', '')
if '<Meta End>' in clean(txt) or '<Meta Start>' in clean(txt):
continue
f.write(clean(author + ': ' + line.replace('\"', '')).replace('\"', '')+'\n')
with open('output.txt', 'r') as f:
lines = f.read()
def clean_07(txt):
txt = txt.replace(']\n[', ' ')
txt = txt.replace('[ ', '[')
txt = txt.replace(' ]', ']')
txt = txt.replace('\n\n', '\n')
txt = txt.replace('~ib~', '**')
txt = txt.replace('#fefefe', '')
txt = txt.replace('<white>', '')
txt = txt.replace('<red>', '')
txt = txt.replace('<blue>', '')
txt = txt.replace('<Meta End>', '')
txt = txt.replace('<Meta Start>', '')
return txt
with open('output.txt', 'w') as f:
f.write(clean_07(lines))