1. 正則

正則就是用一些具有特殊意義的符號組合到一起（正則表達式）來描述字符或者字符串的方法，在python中正則匹配時通過re模塊來實現的

2. re模塊

單個字符匹配

# \w與\W
# s2 = "df當你 的_眼  睛瞇|著/笑?sh29 sedn"
# print(re.findall("\w", s2))   # [‘d‘, ‘f‘, ‘當‘, ‘你‘, ‘的‘, ‘_‘, ‘眼‘, ‘睛‘, ‘瞇‘, ‘著‘, ‘笑‘, ‘s‘, ‘h‘, ‘2‘, ‘9‘, ‘s‘, ‘e‘, ‘d‘, ‘n‘]
# print(re.findall("\W", s2))   # [‘ ‘, ‘ ‘, ‘ ‘, ‘|‘, ‘/‘, ‘?‘, ‘ ‘]
 


# \s與\S
# s3 = "df當\b  2\t瞇|著/笑\r?sh\n29 d\nn"
# print(re.findall("\s", s3))    # [‘ ‘, ‘ ‘, ‘\t‘, ‘\r‘, ‘\n‘, ‘ ‘, ‘\n‘]
# print(re.findall("\S", s3))    # [‘d‘, ‘f‘, ‘當‘, ‘\x08‘, ‘2‘, ‘瞇‘, ‘|‘, ‘著‘, ‘/‘, ‘笑‘, ‘?‘, ‘s‘, ‘h‘, ‘2‘, ‘9‘, ‘d‘, ‘n‘]

# \d與\D
# print(re.findall("\d", s3))   # [‘2‘, ‘2‘, ‘9‘]
 

# print(re.findall("\D", s3))   # [‘d‘, ‘f‘, ‘當‘, ‘\x08‘, ‘ ‘, ‘ ‘, ‘\t‘, ‘瞇‘, ‘|‘, ‘著‘, ‘/‘, ‘笑‘, ‘\r‘, ‘?‘, ‘s‘, ‘h‘, ‘\n‘, ‘ ‘, ‘d‘, ‘\n‘, ‘n‘]

# \A與^
# print(re.findall("\Adf", s3))  # [‘df‘]
# print(re.findall("\Ad", s3))   # [‘d‘]
# print(re.findall("\A當", s3))  # []
# print(re.findall("^df", s3))    # [‘df‘]
 

# print(re.findall("^當", s3))    # []

# s4 = "df當\b 你的眼 睛瞇|著/?sh\n29 dn笑"
# \Z  $   \z不能用
# print(re.findall("笑\Z", s4))   # [‘笑‘]
# print(re.findall("笑$", s4))    # [‘笑‘]

# s5 = "s_\t\t\nhe\t哈哈\n\n 愛好 \ru\n"
# \n與\t
# print(re.findall("\n", s5))   # [‘\n‘, ‘\n‘, ‘\n‘, ‘\n‘]
# print(re.findall("\t", s5))   # [‘\t‘, ‘\t‘, ‘\t‘]

重復匹配

# .  ?  *  +  {m,n}  .*  .*?

# . 匹配任意字符，除了換行符(加上re.DOTALL這二個參數可以匹配\n)
# s1 = "aa bbb aabb acb agb bba babbcb"
# print(re.findall("a.b", s1))    # [‘a b‘, ‘aab‘, ‘acb‘, ‘agb‘, ‘a b‘, ‘abb‘]
# print(re.findall("aa.b", s1))
"""
匹配邏輯
1. 讀取三個字符
2. 進行匹配
3. 成功則返回這三個字符，並從最後一個字符下一個字符開始匹配
4. 失敗則從第一個字符的下一個字符開始匹配
"""

# s2 = "aa babb aabb aaab aaaab bab ba"

# ？ 匹配0個或多個左邊(單個)字符表達式，滿足貪婪規則
# print(re.findall("a?b", s2))   # [‘b‘, ‘ab‘, ‘b‘, ‘ab‘, ‘b‘, ‘b‘, ‘ab‘, ‘b‘, ‘b‘, ‘b‘]

# * 匹配0個或多個左邊(單個)字符的表達式  滿足貪婪規則
# s3 = "aa babb aabb aacb aaab bba ba"
# print(re.findall("aa*b", s3))   # [‘ab‘, ‘aab‘, ‘aaab‘]
# print(re.findall("a*b", s3))   # [‘b‘, ‘ab‘, ‘b‘, ‘aab‘, ‘b‘, ‘b‘, ‘aaab‘, ‘b‘, ‘b‘, ‘b‘]

# + 匹配一個或多個左邊字符的表達式，滿足貪婪規則
# print(re.findall("a+b", s3))   # [‘ab‘, ‘aab‘, ‘aaab‘]
# print(re.findall("ab+b", s3))   # [‘abb‘, ‘abb‘]

# {m,n} 匹配m個至n個左邊表達式，滿足貪婪規則
# s4 = ‘ab aab aaab aaaaabb‘
# print(re.findall("a{2,4}b", s4))  # [‘aab‘, ‘aaab‘, ‘aaaab‘]

# .* 貪婪匹配(盡可能地多），從頭到尾
s5 = "ab aa_b a*()b"
# print(re.findall("a.*b", s5))  # [‘ab aa_b a*()b‘] 匹配以a開頭以b結尾的任意長度的字符串
# 上式匹配邏輯：從a開始，找到最後一個b,停止
# print(re.findall("a.*_", s5))  # [‘ab aa_‘]

# .*?從頭到尾匹配，非貪婪
# print(re.findall("a.*?b", s5))  # [‘ab‘, ‘aa_b‘, ‘aa*()b‘]
# 上式匹配邏輯：從a開始，找到第一個b,停止，繼續下一輪匹配

# []
# [] 括號中可以放任意一個字符
# - 在括號中表示範圍，如果你要匹配上-,那麽這個不能放在中間
# s1 = ‘a1b a3b abb a*b acb a_b‘
# print(re.findall("a[abc]b", s1))   # [‘abb‘, ‘acb‘]
# [abc]表示abc中的任意一個字符
# print(re.findall("a[1-9]b", s1))   # [‘a1b‘, ‘a3b‘]

# s2 = ‘aAb aWb aeb a*b arb a_b‘
# print(re.findall("a[A-Z]b", s2))   # [‘aAb‘, ‘aWb‘]
# print(re.findall("a[a-z]b", s2))   # [‘aeb‘, ‘arb‘]
# print(re.findall("a[A-Za-z]b", s2))  # [‘aAb‘, ‘aWb‘, ‘aeb‘, ‘arb‘]

分組

# 分組
# ()制定一個規則，將滿足規則的結果匹配出來
# 練習1:找到s4裏面的hang juan min
# s4 = "hang_1 hang_gr juan_1 min_1"
# print(re.findall("(.*?)_1", s4))  # [‘hang‘, ‘ hang_gr juan‘, ‘ min‘]
# print(re.findall("([a-z]+)_1", s4))  # [‘hang‘, ‘juan‘, ‘min‘]
# 分析：都是以字母開頭，以_1結尾，字母可以有多個

# 練習2：找到一個標簽裏的網址
# s5 = ‘<a href="http://www.baidu.com">點擊</a>‘
# print(re.findall(‘href="([a-z].*?)"‘, s5))  # [‘http://www.baidu.com‘]

# | 匹配左邊或右邊
# s6 = "hanser:149 yousa:148 mandy:160"
# print(re.findall("hanser|yousa|mandy", s6))  # [‘hanser‘, ‘yousa‘, ‘mandy‘]

# s7 = ‘Too many companies have gone bankrupt, and the next one is my company‘
# print(re.findall("compan(?:y|ies)", s7))   # [‘companies‘, ‘company‘]
# ?:表示將整體匹配出來而不只是()你裏面的內容

3. 常用方法

findall

全部找到並返回一個列表

# 找到下面標簽裏面的網址
import re
s1 = ‘<img src="https://pic3.zhimg.com/80/v2-1d1a5e4f422a77372514a57f38503f3e_hd.jpg" data-rawwidth="564" data-rawheight="699" data-size="normal" data-default-watermark-src="https://pic1.zhimg.com/v2-22b99e59d8efc7e7dec3faba8fbf2a24_b.jpg" class="origin_image zh-lightbox-thumb lazy" width="564" data-original="https://pic3.zhimg.com/v2-1d1a5e4f422a77372514a57f38503f3e_r.jpg" data-actualsrc="https://pic3.zhimg.com/v2-1d1a5e4f422a77372514a57f38503f3e_b.jpg">‘
print(re.findall(‘src="([a-z].*?)"‘, s1)) 

# 結果[‘https://pic3.zhimg.com/80/v2-1d1a5e4f422a77372514a57f38503f3e_hd.jpg‘, ‘https://pic1.zhimg.com/v2-22b99e59d8efc7e7dec3faba8fbf2a24_b.jpg‘, ‘https://pic3.zhimg.com/v2-1d1a5e4f422a77372514a57f38503f3e_b.jpg‘]

search

找到第一個並返回包含匹配信息的對象，該對象可以通過group()方法得到匹配的字符串，沒找到返回None

s = "Hanser is a little girl in kindergarten"
ret = re.search("[A-Z][a-z]*", "Hanser is a little girl in kindergarten")
print(ret)   # <_sre.SRE_Match object; span=(0, 6), match=‘Hanser‘>
print(ret.group())   # Hanser

match

同search，區別在於從字符串開始處進行匹配，可以用search+^代替

s = "Hanser is a little girl in kindergarten"
print(re.match("Hanser", s).group())   # Hanser

split

按照指定的分割符分割

s = "Hanser is a little girl in kindergarten"
print(re.split(" ", s))  # [‘Hanser‘, ‘is‘, ‘a‘, ‘little‘, ‘girl‘, ‘in‘, ‘kindergarten‘]

s1 = "花褪殘紅青杏小，燕子飛時，綠水人家繞。枝上柳綿吹又少，天涯何處無芳草。"
# 方法一：
lst = re.split("[，。]", s1)  # [‘花褪殘紅青杏小‘, ‘燕子飛時‘, ‘綠水人家繞‘, ‘枝上柳綿吹又少‘, ‘天涯何處無芳草‘, ‘‘]
for i in lst:
    if i:  # 過濾空字符
        print(i[0])  # 花 燕 綠 枝 天

# 方法二：
lst1 = re.findall(r"[，。]([^，。])", s1)
print(lst1)  # [‘燕‘, ‘綠‘, ‘枝‘, ‘天‘] 只能找到除去開頭的短句首字

sub

替換

s2 = "大家好，我是常山趙子龍"
print(re.sub("常山", "石家莊", s2))   # 大家好，我是石家莊趙子龍

compile

# compile 制定一個匹配規則
obj = re.compile("\d{2}")
print(obj.search("sdfs14523sdf").group())  # 14
print(obj.findall("sdfs14523sdf"))    # [‘14‘, ‘52‘]

finditer

返回一個存放匹配結果的叠代器

ret = re.finditer("\d", "sd283sef8w3o7sh")
print(ret)   # 叠代器 <callable_iterator object at 0x000001C37F3A9C50>
print(next(ret))  # match對象
print(next(ret).group())  # 8
print(next(ret).group())  # 3
print(next(ret).group())  # 8
print([i.group() for i in ret])  # 查看剩余結果

命名分組

# ret = re.search(r"<(?P<tag_name>\w+)>\w+</(?P=tag_name)>", "<h1>hello</h1>")
# 在分組中利用?P<name>給分組起名字
# 獲取的匹配結果可以直接用group("名字")拿到對應的值
# print(ret.group())   # <h1>hello</h1>
# print(ret.group("tag_name"))  # h1

# 如果不給組起名字，也可以用\序號來找到對應的組，獲取的結果可以直接用group(序號)拿到對應的值
# ret = re.search(r"<(\w+)>\w+</\1>", "<h1>hello</h1>")
# print(ret.group())  # <h1>hello</h1>
# print(ret.group(1))  # h1

# ret = re.findall(r"<(?P<tag_name>\w+)>\w+</(?P=tag_name)>", "<h1>hello</h1>")
# print(ret)   # [‘h1‘]

python模塊之re模塊