import csv
import re
pattern = r'(?<=[\u4e00-\u9fa5])\s+(?=[^\u4e00-\u9fa5])|(?<=[^\u4e00-\u9fa5])\s+(?=[\u4e00-\u9fa5])'
results = []
with open('vl6_20.csv', 'r', encoding="utf-8") as file:
reader = csv.reader(file)
for row in reader:
tm = ''
tm = re.sub(pattern, '', row[2])
tm = tm.replace('。 ', '。')
tm = tm.replace(', ',',')
tm = tm.replace('; ',';');
tm = tm.replace(', ',',');
tm = tm.replace('. ','.');
tm = tm.replace('; ',';');
row[2] = tm
results.append(row)
with open('vl6_20-ok.csv', 'w', newline='', encoding="utf-8") as file2:
writer = csv.writer(file2)
writer.writerows(results)