如何在特定块中保留字符串，但在另一个块中更改字符串

Question

我有一个2个块彼此略有不同的文件。下面是文件的内容

 Other codes in the file
 function void one(int x)
     message_same
     rest of the code
  endfunction
 Other codes in the file
  function void othercheck ::two(int x)
      message_same
      rest of the code
   endfunction

   Different codes in the file 

我在列表中读取了此文件，并进行了一些更改，并希望写入另一个文件。

但是我想如果在函数1下看到“ message_same”，则应照原样编写，但是如果在函数2下看到，则应删除该行或不要将该行写入输出文件。 其他代码行应保持原样

预期产量：

 Other codes in the file
 virtual function void one(int x)
 message_same
 rest of the code
 endfunction
 Other codes in the file

  function void two:: othercheck(int x)
  rest of the code
  endfunction  
  Different codes in the file

我尝试使用以下代码：

for word in words:
     found_one_function=re.search('virtual function',word)
     if found_in_function :
        found_in_end=re.search('endfunction',word)
        if not found_in_end:
            found_in_function=True

   while(found_in_function):
           fw.write(word)
           continue

   if re.search('message_same', word):
        continue

   fw.write(word)

我理解这在逻辑上是不正确的，但是我不确定在找到虚函数之后如何迭代直到获得结束函数。

任何帮助都会很棒。

Answer 1

这样做相对容易-您想要遍历words列表（假设每个元素包含示例数据中的一行），并检查第二个“类型”函数的开头，然后删除包含message_same行，直到您遇到唯一的endfunction ，例如：

# assuming `words` list with each line of your data
# if not it's as easy as: with open("input.txt") as f: words = [line for line in f]
with open("output.txt", "w") as f:  # open output.txt for writing
    in_function = False  # an identifier to tell us we are within a `::` function
    for line in words:  # iterate over words
        if in_function:  # we are inside of a `::` function...
            if line.strip() == "endfunction":  # end of the function
                in_function = False
            elif "message_same" in line:  # skip this line
                continue
        # detect function begin if there is "function" in the line followed with ::
        elif "function" in line and line.find("function") < line.find("::"):
            in_function = True
        f.write(line)  # write the line to the output file
        # f.write("\n")  # uncomment if the lines in your `words` are not terminated

对于其行已作为包含以下内容的words元素加载的文件：

Other codes in the file
 function void one(int x)
     message_same
     rest of the code
  endfunction
 Other codes in the file
  function void othercheck ::two(int x)
      message_same
      rest of the code
   endfunction

   Different codes in the file

它将产生包含以下内容的output.txt ：

Other codes in the file
 function void one(int x)
     message_same
     rest of the code
  endfunction
 Other codes in the file
  function void othercheck ::two(int x)
      rest of the code
   endfunction

   Different codes in the file

您可以根据需要拥有任意数量的函数，而无需订购它们-处理将仅应用于带有::函数。

Answer 2

遍历文件中的每一行； 使用一个标志来跟踪进程是否在 ::函数中； 使用该标志丢弃message_same行； 根据需要修改行； 将行写入新文件。

import re

special = re.compile(r'function.*?::')
in_special_func = False
with open(in_filepath) as in_file, open(out_filepath, 'w') as out_file:
    for line in in_file:
        if special.search(line):
            in_special_func = True
        if 'endfunction' in line:
            in_special_func = False
        if in_special_func and 'message_same' in line:
            #skip
            continue
        # make line modifications here if needed
        # line = modify(line)
        # line = some_variation_of(line)
        # print(line)
        out_file.write(line)

---

先前的错误尝试

构造一个将捕获完整功能的正则表达式

f_re = re.compile(r'function.*?endfunction', flags = re.DOTALL)

构造正则表达式以识别特殊功能

special = re.compile(r'function.*?::')

构造一个正则表达式，使其与需要删除的行匹配

message_same = re.compile(r'^\s*message_same\s*\n', flags = re.MULTILINE)

将文件读取为字符串：

with open(in_filepath) as in_file:
   s = in_file.read()

遍历所有功能； 如果功能特殊，则删除该行； 对功能进行其他修改； 将其写入文件。

with open(out_filepath, 'w') as out_file:
   for f in f_re.findall(s):
      #print(f)
      if special.search(f):
         f = message_same.sub('', f)
      # make other changes here
      # assuming the result is a single string
      out_file.write(f)
      #print(f)

Answer 3

这是一种删除带有签名的每个函数的“消息相同”行的方法，该签名包含“函数”和“两个”。 假设输入文件的结构非常一致。

# read file into list of lists (each inner list is a block)
with open('code_blocks.txt', 'r') as f:
    blocks = [block.split('\n') for block in f.read().split('\n\n')]

# iterate over blocks
for block in blocks:
    # if first line contains 'function' and 'two' and second line contains 'message same'
    if 'function' in block[0] and '::' in block[0] and 'message_same' in block[1]:
        # remove message same
        block.pop(block.index(block[1]))

# combine list of lists back into single string and write it out
with open('code_blocks_out.txt', 'w') as f:
    f.write('\n\n'.join(['\n'.join(block) for block in blocks]))