如何用Python正确高亮显示两个字符串的差异？

需求说明 #

我想用Python代码高亮显示两个字符串之间的差异，具体示例如下：

示例1 #

原字符串：

sentence1 = "I'm enjoying the summer breeze on the beach while I do some pilates."
sentence2 = "I am enjoying the summer breeze on the beach while I am doing some pilates."

预期结果（星号标注部分需显示为红色）：

I *am* enjoying the summer breeze on the beach while I *am doing* some pilates.

示例2 #

原字符串：

sentence1 = "My favourite season is Autumn while my sister's favourite season is Winter."
sentence2 = "My favourite season is Autumn, while my sister's favourite season is Winter."

预期结果（逗号为差异部分）：

"My favourite season is Autumn*,* while my sister's favourite season is Winter." 

尝试的代码及问题 #

我写了以下代码：

sentence1 = "I'm enjoying the summer breeze on the beach while I do some pilates."
sentence2 = "I'm enjoying the summer breeze on the beach while I am doing some pilates."

# Split the sentences into words
words1 = sentence1.split()
words2 = sentence2.split()

# Find the index where the sentences differ
index_of_difference = next((i for i, (word1, word2) in enumerate(zip(words1, words2)) if word1 != word2), None)

# Highlight differing part "am doing" in red
highlighted_words = []
for i, (word1, word2) in enumerate(zip(words1, words2)):
    if i == index_of_difference:
        highlighted_words.append('\033[91m' + word2 + '\033[0m')
    else:
        highlighted_words.append(word2)

highlighted_sentence = ' '.join(highlighted_words)
print(highlighted_sentence)

但运行后得到的结果是：

I'm enjoying the summer breeze on the beach while I *am* doing some

和预期的不符，预期应该是：

I'm enjoying the summer breeze on the beach while I *am doing* some pilates.

解决方案 #

你的问题出在两个地方：一是仅处理了第一个差异单词，没识别到连续的差异片段；二是基于空格拆分单词的方式无法覆盖字符级的差异（比如示例2的逗号）。可以用Python内置的difflib库精准识别连续差异，再统一高亮：

完整代码 #

import difflib

def highlight_diff(source_str, target_str):
    # 初始化字符串匹配器
    matcher = difflib.SequenceMatcher(None, source_str, target_str)
    result_parts = []
    
    # 遍历所有操作码，处理不同类型的片段
    for tag, src_start, src_end, target_start, target_end in matcher.get_opcodes():
        if tag in ('replace', 'insert'):
            # 对差异片段添加ANSI红色高亮
            result_parts.append(f'\033[91m{target_str[target_start:target_end]}\033[0m')
        elif tag == 'equal':
            # 相同片段直接添加
            result_parts.append(target_str[target_start:target_end])
        # 忽略delete操作，因为我们以目标字符串为展示基准
    
    return ''.join(result_parts)

# 测试示例1
sentence1 = "I'm enjoying the summer breeze on the beach while I do some pilates."
sentence2 = "I am enjoying the summer breeze on the beach while I am doing some pilates."
print(highlight_diff(sentence1, sentence2))

# 测试示例2
sentence3 = "My favourite season is Autumn while my sister's favourite season is Winter."
sentence4 = "My favourite season is Autumn, while my sister's favourite season is Winter."
print(highlight_diff(sentence3, sentence4))

代码说明 #

1. difflib.SequenceMatcher：专门用于对比两个序列（这里是字符串），能精准找出连续的相同或差异片段。
2. get_opcodes()：返回的每个元组包含操作类型和对应片段的索引，操作类型包括：
   • equal：两段字符串相同
   • replace：目标字符串替换了原字符串的片段
   • insert：目标字符串新增了片段
   • delete：原字符串有但目标字符串没有的片段（这里忽略，因为我们展示目标字符串）
3. 高亮逻辑：对replace和insert类型的片段添加ANSI红色控制码，在终端中会显示为红色。
4. 适配场景：既支持单词级的连续差异（示例1的am doing），也支持单个字符的差异（示例2的逗号）。

内容的提问来源于stack exchange，提问作者Oliver