从基础到进阶详解Python如何使用正则替换字符串

import re

re.sub(pattern, repl, string, count=0, flags=0)

import re

text = "我的电话是123456789，QQ是987654321"
new_text = re.sub(r'\d+', '***', text)  # \d+ 匹配1个或多个数字
print(new_text)
# 输出: 我的电话是***，QQ是***

text = "Hello! 123 @World#Python"
new_text = re.sub(r'[^a-zA-Z]', ' ', text)  # [^a-zA-Z] 匹配非字母字符
print(new_text)
# 输出: Hello      World Python

text = "Python is FUN, python is powerful"
new_text = re.sub(r'python', 'Java', text, flags=re.IGNORECASE)  # 忽略大小写
print(new_text)
# 输出: Java is FUN, Java is powerful

text = "2023-01-01, 2024-12-31"
new_text = re.sub(r'(\d{4})-(\d{2})-(\d{2})', r'\3/\2/\1', text)  # \3=日, \2=月, \1=年
print(new_text)
# 输出: 01/01/2023, 31/12/2024

def double_num(match):
    num = int(match.group())  # 获取匹配的数字
    return str(num * 2)

text = "1 apple, 2 bananas, 3 oranges"
new_text = re.sub(r'\d+', double_num, text)  # 调用函数替换
print(new_text)
# 输出: 2 apple, 4 bananas, 6 oranges

def hide_email(match):
    email = match.group()
    return email[0] + "***" + email[-4:]  # 保留首字母和后4位

text = "联系我：test@example.com 或 admin@site.org"
new_text = re.sub(r'[\w.-]+@[\w.-]+', hide_email, text)
print(new_text)
# 输出: 联系我：t***mple.com 或 a***site.org

text = "111-222-333-444"
new_text = re.sub(r'\d+', 'X', text, count=2)  # 只替换前2个数字
print(new_text)
# 输出: X-X-333-444

import re

html = "<p>Hello, <b>World</b>!</p>"
clean_text = re.sub(r'<[^>]+>', '', html)  # 匹配所有HTML标签并删除
print(clean_text)
# 输出: Hello, World!

text = "电话：010-12345678，手机：138-1234-5678"
new_text = re.sub(r'(\d{3})-(\d{4})-(\d{4})', r'\1\2\3', text)  # 尝试直接替换（可能不匹配）
# 更通用的方法：
def normalize_phone(match):
    full_num = re.sub(r'[^\d]', '', match.group())  # 先删除所有非数字
    if len(full_num) == 11:  # 手机号码
        return f"手机：{full_num[:3]}-{full_num[3:7]}-{full_num[7:]}"
    elif len(full_num) == 8:  # 座机号码（简化版）
        return f"电话：{full_num[:3]}-{full_num[3:]}"
    else:
        return match.group()

# 先提取所有电话号码（简化版，实际需更复杂正则）
text_with_phones = re.sub(r'(\d{3}[-]?\d{4}[-]?\d{4})|(\d{3}[-]?\d{4})', normalize_phone, text)
# 更准确的方式是分两步：
# 1. 提取所有电话号码
# 2. 替换
print(text_with_phones)  # 输出可能不符合预期，需更精细的正则

# 更合理的实现（分两步）：
phones = re.findall(r'\d{3}[-]?\d{4}[-]?\d{4}|\d{3}[-]?\d{4}', text)
for phone in phones:
    cleaned = re.sub(r'[^\d]', '', phone)
    if len(cleaned) == 11:
        new_phone = f"{cleaned[:3]}-{cleaned[3:7]}-{cleaned[7:]}"
    elif len(cleaned) == 8:
        new_phone = f"{cleaned[:3]}-{cleaned[3:]}"
    else:
        new_phone = phone
    text = text.replace(phone, new_phone)
print(text)
# 输出: 电话：010-12345678，手机：138-1234-5678

text = "电话：010-12345678，手机：138-1234-5678"

def normalize_phone(match):
    num = re.sub(r'[^\d]', '', match.group())
    if len(num) == 11:
        return f"手机：{num[:3]}-{num[3:7]}-{num[7:]}"
    elif len(num) == 8:
        return f"电话：{num[:3]}-{num[3:]}"
    else:
        return match.group()

# 匹配手机号（11位）或座机号（8位，可能带-）
new_text = re.sub(
    r'(手机：)?(\d{3}[-]?\d{4}[-]?\d{4})|(电话：)?(\d{3}[-]?\d{4})',
    lambda m: normalize_phone(m),
    text
)
# 上述正则不够完美，更准确的方式：
# 先提取所有电话号码，再替换
print(new_text)  # 可能需要调整正则

# 更准确的实现（分两步）：
import re

text = "电话：010-12345678，手机：138-1234-5678"

# 匹配手机号或座机号
pattern = r'(?:电话：|手机：)?(\d{3}[-]?\d{4}[-]?\d{4}|\d{3}[-]?\d{4})'

def replace_phone(match):
    raw_phone = match.group(1)
    cleaned = re.sub(r'[^\d]', '', raw_phone)
    if len(cleaned) == 11:
        return f"手机：{cleaned[:3]}-{cleaned[3:7]}-{cleaned[7:]}"
    elif len(cleaned) == 8:
        return f"电话：{cleaned[:3]}-{cleaned[3:]}"
    else:
        return raw_phone

new_text = re.sub(pattern, replace_phone, text)
print(new_text)
# 输出: 电话：010-12345678，手机：138-1234-5678

text = "电话：010-12345678，手机：138-1234-5678"
new_text = re.sub(
    r'(\d{3})-(\d{4})-(\d{4})',  # 匹配手机号格式
    r'手机：\1-\2-\3',
    re.sub(
        r'(\d{3})-(\d{4})',  # 匹配座机号格式
        r'电话：\1-\2',
        text
    )
)
print(new_text)
# 输出: 电话：010-12345678，手机：138-1234-5678

text = "访问 http://example.com 或 https://test.org"
new_text = re.sub(r'http://', 'https://', text)  # 简单替换
print(new_text)
# 输出: 访问 https://example.com 或 https://test.org