Python如何按顺序替换指定关键词且每个关键词仅替换首次出现实例

需求说明 #

我知道在句子中替换字符串有多种实现方式，原有示例代码如下：

sentences= 'This is a sentence, even there may be a lot of sentence coming'
word_list = ["sentence", 'be', 'coming']

to_replace = ["changes"]  

res = ' '.join([to_replace if idx in word_list else idx for idx in sentences.split()])
 
print("After replace: " + str(res)) 

# 输出
'This is a changes, even there may changes a lot of sentence changes'

我实际的需求是按word_list的顺序，仅替换对应关键词的首次出现，即使句子中存在重复关键词也不全量替换，比如示例中的sentence重复出现两次，仅替换第一次出现的实例即可。需求对应的参数如下：

sentence = 'This is a sentence, even there may be a lot of sentence coming'
word_list = ["sentence", 'be', 'coming']
to_replace = ['change1', 'change2', 'change3']

期望输出结果为：

output: 'This is a change1, even there may change2 a lot of sentence change3'

补充问题 #

我写了如下测试代码，运行后输出了重复的句子，我期望仅得到一个完成所有替换的句子，请问该如何解决？
原有测试代码：

k = []
dummy = 'Lorem what your doing ? how are you guys doing ? dummy me doing bad '

for p1 in x1:
  for n1, n2 in zip(['Lorem', 'dummy'], ['hello', 'hai']):
     new = p1.replace(n1, n2)
     k.append(new)

# 输出
['hello what your doing ? how are you guys doing ? dummy me doing bad ',
'Lorem what your doing ? how are you guys doing ? hai me doing bad ']

解决方案 #

1. 实现按顺序仅替换关键词首次出现的需求 #

核心逻辑是利用Python字符串replace方法的第三个参数count，指定count=1即可每次仅替换第一个匹配项，同时要在已经完成替换的字符串基础上继续操作，不要每次都用原始字符串替换。
实现代码：

sentence = 'This is a sentence, even there may be a lot of sentence coming'
word_list = ["sentence", 'be', 'coming']
to_replace = ['change1', 'change2', 'change3']

res = sentence
# 按顺序配对关键词和替换词，每次仅替换首个匹配项
for old_word, new_word in zip(word_list, to_replace):
    res = res.replace(old_word, new_word, 1)

print("After replace: " + res)

运行后输出和你期望的结果完全一致。

2. 修正测试代码重复输出的问题 #

原有代码的问题是每次都用原始字符串p1做替换，且每次替换后就往列表里添加一次结果，才会输出两个拆分的句子。正确做法是先初始化结果为原始字符串，依次完成所有替换后，仅添加一次最终结果到列表中即可。
修正后的代码：

k = []
# 此处x1为存储原句子的列表，用你给出的dummy做示例
x1 = ['Lorem what your doing ? how are you guys doing ? dummy me doing bad ']
replace_pairs = zip(['Lorem', 'dummy'], ['hello', 'hai'])

for p1 in x1:
    res = p1
    for n1, n2 in replace_pairs:
        # 在上一次替换后的结果基础上继续操作
        res = res.replace(n1, n2, 1)
    k.append(res)

print(k)

运行后输出为：['hello what your doing ? how are you guys doing ? hai me doing bad ']，符合需求。

内容的提问来源于stack exchange，提问作者Titan