python实现逐行替换超大文件中的字符串

import re
#自定义正则
rex="\'0000-00-00 00:00:00\'"
new_str='now()'
old_file_path=r'C:\Users\Ths\Desktop\test.txt'
new_file_path=r'C:\Users\Ths\Desktop\newtest.txt'
def match_timestamp(repex,eachline):
    p=re.compile(repex)
    return p.findall(eachline)
    
#打开旧文件，将每一行yield后作为迭代器返回。
def old_file_yield(old_file_path):
    with open(old_file_path,'r') as  oldf:
      while True:
         line=oldf.readline()
         yield line
         if not line:
             oldf.close()
             break
#打开新文件开始逐行读取替换。     
def replace_match(old_file_path,new_file_path):
  count=0
  with open(new_file_path,"w") as newf:
     for line in old_file_yield(old_file_path):
                ifmatch=match_timestamp(rex,line)
                if not line:
                    newf.close()
                    return count
                    break
                elif ifmatch !=[]:
                     count+=1
                     print("替换前：%s" % line)
                     line=line.replace(rex,new_str)
                     print("替换后：%s" % line)
                     newf.write(line)
                else:
                     newf.write(line)
print('一共替换了%s行' % replace_match(old_file_path,new_file_path))

 

转载于:https://www.cnblogs.com/lalalaxixixi/p/10861919.html