没有找到一个好的正则表达式模式来以正确的顺序替换字符串（python）

Question

我有一个字符串格式的列名列表，如下所示：

lst = ["plug", "[plug+wallet]", "(wallet-phone)"]

现在我想使用regex将df[]和" ' "添加到每个列名中，我这样做了，当列表中有(wallet-phone)这种字符串时，它会给出这样的输出df[('wallet']-df['phone')]。我怎么会这样(df['wallet']-df['phone']),我的模式错了。请参阅下文：

import re
lst = ["plug", "[plug+wallet]", "(wallet-phone)"]
x=[]
y=[]
for l in lst: 
    x.append(re.sub(r"([^+\-*\/'\d]+)", r"'\1'", l))
    for f in x:    
        y.append(re.sub(r"('[^+\-*\/'\d]+')", r'df[\1]',f))

print(x)
print(y)

给出：

x:["'plug'", "'[plug'+'wallet]'", "'(wallet'-'phone)'"]
y:["df['plug']", "df['[plug']+df['wallet]']", "df['(wallet']-df['phone)']"]

模式不对吗？ 预期产出：

x:["'plug'", "['plug'+'wallet']", "('wallet'-'phone')"]
y:["df['plug']", "[df['plug']+df['wallet']]", "(df['wallet']-df['phone'])"]

我也尝试了([^+\-*\/()[]'\d]+)这种模式，但它并没有避免() or []

Answer 1

查找单词并将其包含在字典参考中可能更容易：

import re
lst = ["plug", "[plug+wallet]", "(wallet-phone)"]

z = [re.sub(r"(\w+)",r"df['\1']",w) for w in lst]

print(z)
["df['plug']", "[df['plug']+df['wallet']]", "(df['wallet']-df['phone'])"]