资源

Python Regular Expressions - part #1 - YouTube
Regular Expressions are used to match string patterns. 正则表达式用于匹配字符串模式。
- They are very powerful 它们非常强大
- If you want to pull out a string pattern RE can do it 如果你想拉出一个字符串模式正则可以做到
- They may seem intimidating 他们可能看起来很吓人

课程

Things to note

The first thing I want start off with is the the back slash character
- 我想要开始的第一件事是反斜杠字符
Very confusing to people
- 很让人困惑
Python uses back slash to indicate special characters
- Python 使用反斜杠表示特殊字符
'\n' Backslash followed by n denotes a newline
- 反斜杠后面加 n 表示换行符
'\t' denotes a tab
- '\t' 表示制表符
'r' expression, that voids the Python's special characters
- 'r'表达式，将使 Python 的特殊字符无效
r'\n' means it's a raw string with two characters 'n' and '' as opposed to just one special character'
- r'\n'表示它是一个有两个字符'n'和''的原始字符串而不是只有一个特殊字符'
Let's see some examples of this dont mind the python syntax
- 让我们看一些这样的例子，不要介意 python 语法

re.search(pattern, string, flags =0)

python

import re
re.search('n', '\n')  # first item is pattern, second item is string

python

# two ways to handle this one way is to use \ for every backslash
# 有两种处理方法，一种方法是对每个反斜杠使用\ (另一种是在前面加 r)
re.search('n', '\\n')

<re.Match object; span=(1, 2), match='n'>

python

# not the best way if we have too many \s
# 如果我们有太多的\，这不是最好的方法
re.search('n',  '\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n')

python

# r converts to raw string
# r 转换为原始字符串
re.search('n',  r'\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n')

<re.Match object; span=(1, 2), match='n'>

python

"""
there are some nuances that you should be aware of regular expressions 
has its own special characters as well regex with '\n' and r'\n' both
look for newline.
你应该注意到一些细微的差别，正则表达式也有自己的特殊字符，
带有'\n'和 r'\n'的正则表达式都查找换行符
"""
re.search('\n',  '\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n')

<re.Match object; span=(0, 1), match='\n'>

python

# this works as well because r'\n' also looks for new line
# 同样有效，因为 pattern 中 r'\n' 也会查找新行
re.search(r'\n',  '\n\n')

<re.Match object; span=(0, 1), match='\n'>

python

# doesn't work because string doesn't use newline and r'\n' looks for newline
# 不能工作，因为 string 不使用换行符，而 pattern 中 r'\n'查找换行符
re.search(r'\n',  r'\n\n')

MATCH and SEARCH EXAMPLES

REs common methods - Match and Search

searches anywhere in the sentence 搜索句子中的任何位置
flags: special options 标志

re.search(pattern, string, flags)

only beginning of the string 只搜索字符串的开始

re.match(pattern, string, flags)

python

# returns none because only looks at the start of string
# 返回 none，因为只查看字符串的开头
re.match("c", "abcdef")

python

re.search("c", "abcdef")  # searches anywhere

<re.Match object; span=(2, 3), match='c'>

python

bool(re.match("c", "abcdef"))  # no match returns boolean false

False

python

bool(re.match("a", "abcdef"))  # match returns true

True

python

# tells you where it matched first and only first
# 告诉你它首先匹配的位置
re.search("c", "abcdef")

<re.Match object; span=(2, 3), match='c'>

python

re.search("c", "abcdefc")  # multiple 'c's first instance only 返回多个 c 的第一个实例

<re.Match object; span=(2, 3), match='c'>

python

re.search("c", "abdef\nc")  # multiline works with search 多行与搜索一起工作

<re.Match object; span=(6, 7), match='c'>

python

re.match("c", "\nc")  # match doesn't work with newline 匹配对换行符无效

Printing the output of match and search

python

(re.match("a", "abcdef"))  # match objects

<re.Match object; span=(0, 1), match='a'>

python 正则匹配中 re.match().group(num=0)

python

re.match("a", "abcdef").group()  # string output # defautlt value is 0 默认值为 0

'a'

python

re.match("a", "abcdef").group(0)

'a'

python

re.search("n", "abcdefnc abcd").group()

'n'

python

re.search('n.+', "abcdefnc abcd").group()  # pull out different types of strings 拿出不同类型的字符串
                                           # depending on the wildcards you use 这取决于您使用的通配符

'nc abcd'

python 正则(2)group/start/end/span 方法

python

re.search("c", "abdef\nc").start()

python

re.search("c", "abdef\nc").end()

Literal matching

python

re.search('na',"abcdefnc abcd" )  # doesn't work, because they are ordered 无效，因为它们是有序的

python

re.search('n|a',"abcdefnc abcda" )  # n or a

<re.Match object; span=(0, 1), match='a'>

python

 re.search('n|a',"bcdefnc abcda" )  # replaced the a with b, first match is an n

<re.Match object; span=(5, 6), match='n'>

python

re.search('n|a|b',"bcdefnc abcda" ) # as many OR expressions

<re.Match object; span=(0, 1), match='b'>

re.findall

python

re.findall('n|a',"bcdefnc abcda" ) # find all pulls out all instances 取出所有实例

['n', 'a', 'a']

python

re.search('abcd',"abcdefnc abcd" ) # multiple characters - literal search 多字符-文字搜索

<re.Match object; span=(0, 4), match='abcd'>

python

re.findall('abcd',"abcdefnc abcd" )

['abcd', 'abcd']

CHARACTER SETS

Character sets can match a set of characters
- 简化正则表达式

python

import re
re.search('abcd',"abcdefnc abcd" ) # earlier code 之前的代码

<re.Match object; span=(0, 4), match='abcd'>

python

re.search(r'\w\w\w\w',"abcdefnc abcd" )  # matches characters and numbers 匹配字符和数字
                                         # alpha numeric characters

<re.Match object; span=(0, 4), match='abcd'>

\w matches alpha numeric characters [a-zA-Z0-9_]

python

re.search(r'\w\w\w\w',"ab_cdefnc abcd" )  # matches _ character 匹配_字符

<re.Match object; span=(0, 4), match='ab_c'>

python

re.search(r'\w\w\w', "a3.!-!")  # doesn't match symbols only numbers and characters 不匹配符号，只匹配数字和字符

python

re.search(r'\w\w\w', "a33-_!") .group()

'a33'

\W opposite of \w ;

so nothing included in [a-zA-Z0-9_]

python

re.search(r'\w\w\W', "a3.-_!")  # \W matches non characters and numbers
                                # \W 匹配非字符和数字

<re.Match object; span=(0, 3), match='a3.'>

python

re.search(r'\w\w\W', "a3 .-_!")  # matches empty space as well 
                                 # \W 也可以匹配空格

<re.Match object; span=(0, 3), match='a3 '>

We will go over other character sets later on

Let's go over quantifiers'

quantifiers
- '+' = 1 or more
- '?' = 0 or 1
- '*' = 0 or more
- '{n,m}' = n to m repetitions {,3}, {3,}

python

re.search(r'\w\w',"abcdefnc abcd" )

<re.Match object; span=(0, 2), match='ab'>

python

re.search(r'\w+',"abcdefnc abcd" ).group()  # don't know the numbers of letters 不知道单词的字母个数

'abcdefnc'

python

re.search(r'\w+\W+\w+',"abcdefnc abcd").group()

'abcdefnc abcd'

python

re.search('\w+\W+\w+',"abcdefnc       abcd").group()  # added spaces

'abcdefnc       abcd'

python

re.search(r'\w+\W?\w+',"abcdefnabcd").group()  # ? = 0 or 1 instances

'abcdefnabcd'

python

re.search(r'\w+\W?\w+',"abcde fnabcd").group()

'abcde fnabcd'

python

re.search(r'\w+\W+\w+', "abcdefnabcd")

Pulling out specific amounts
- 取出特定数量

python

re.search(r'\w{3}', 'aaaaaaaaaaa')  # only 3 \w characters

<re.Match object; span=(0, 3), match='aaa'>

python

re.search(r'\w{1,4}', 'aaaaaaaaaaa').group()  #1 is min, 4 is max

'aaaa'

python

re.search(r'\w{1,10}\W{0,4}\w+',"abcdefnc abcd").group()
# 1-10 \w characters,
# 0-4  \W chracters,
# 1+ \w characters

'abcdefnc abcd'

python

re.search(r'\w{1,}\W{0,}\w+',"abcdefnc abcd").group() #at least 1, at least 0, 1+

'abcdefnc abcd'

Other types of characters sets

'\d' = matches digits [0-9]

'\D' = matches This matches any non-digit character; ~\d

python

import re
string = '23abced++'
re.search('\d+', string).group()

'23'

'\s' = matches any whitespace character, new lines, tabs, spaces etc 匹配任何空白字符，新行，制表符，空格等

'\S' = matches any non-whitespace chracter : ~~\s 匹配任何非空格字符:~~\s

python

string = '23abced++'
re.search('\S+', string).group()  # no spaces

'23abced++'

python

string = '''Robots are branching out. A new prototype soft robot takes inspiration from plants by growing to explore its environment.
 
Vines and some fungi extend from their tips to explore their surroundings. 
Elliot Hawkes of the University of California in Santa Barbara 
and his colleagues designed a bot that works 
on similar principles. Its mechanical body 
sits inside a plastic tube reel that extends 
through pressurized inflation, a method that some 
invertebrates like peanut worms (Sipunculus nudus)
also use to extend their appendages. The plastic 
tubing has two compartments, and inflating one 
side or the other changes the extension direction. 
A camera sensor at the tip alerts the bot when it’s 
about to run into something.
 
In the lab, Hawkes and his colleagues 
programmed the robot to form 3-D structures such 
as a radio antenna, turn off a valve, navigate a maze, 
swim through glue, act as a fire extinguisher, squeeze 
through tight gaps, shimmy through fly paper and slither 
across a bed of nails. The soft bot can extend up to 
72meters, and unlike plants, it can grow at a speed of 
10meters per second, the team reports July 19 in Science Robotics. 
The design could serve as a model for building robots 
that can traverse constrained environments
 
This isn’t the first robot to take 
inspiration from plants. One plantlike 
predecessor was a robot modeled on roots.'''

python

(re.findall('\S+', string))  # 返回 string 中所有的单词

['Robots',
 'are',
 'branching',
 'out.',
 'A',
 'new',
 'prototype',
 'soft',
 'robot',
 'takes',
 'inspiration',
 'from',
 'plants',
 'by',
 'growing',
 'to',
 'explore',
 'its',
 'environment.',
 'Vines',
 'and',
 'some',
 'fungi',
 'extend',
 'from',
 'their',
 'tips',
 'to',
 'explore',
 'their',
 'surroundings.',
 'Elliot',
 'Hawkes',
 'of',
 'the',
 'University',
 'of',
 'California',
 'in',
 'Santa',
 'Barbara',
 'and',
 'his',
 'colleagues',
 'designed',
 'a',
 'bot',
 'that',
 'works',
 'on',
 'similar',
 'principles.',
 'Its',
 'mechanical',
 'body',
 'sits',
 'inside',
 'a',
 'plastic',
 'tube',
 'reel',
 'that',
 'extends',
 'through',
 'pressurized',
 'inflation,',
 'a',
 'method',
 'that',
 'some',
 'invertebrates',
 'like',
 'peanut',
 'worms',
 '(Sipunculus',
 'nudus)',
 'also',
 'use',
 'to',
 'extend',
 'their',
 'appendages.',
 'The',
 'plastic',
 'tubing',
 'has',
 'two',
 'compartments,',
 'and',
 'inflating',
 'one',
 'side',
 'or',
 'the',
 'other',
 'changes',
 'the',
 'extension',
 'direction.',
 'A',
 'camera',
 'sensor',
 'at',
 'the',
 'tip',
 'alerts',
 'the',
 'bot',
 'when',
 'it’s',
 'about',
 'to',
 'run',
 'into',
 'something.',
 'In',
 'the',
 'lab,',
 'Hawkes',
 'and',
 'his',
 'colleagues',
 'programmed',
 'the',
 'robot',
 'to',
 'form',
 '3-D',
 'structures',
 'such',
 'as',
 'a',
 'radio',
 'antenna,',
 'turn',
 'off',
 'a',
 'valve,',
 'navigate',
 'a',
 'maze,',
 'swim',
 'through',
 'glue,',
 'act',
 'as',
 'a',
 'fire',
 'extinguisher,',
 'squeeze',
 'through',
 'tight',
 'gaps,',
 'shimmy',
 'through',
 'fly',
 'paper',
 'and',
 'slither',
 'across',
 'a',
 'bed',
 'of',
 'nails.',
 'The',
 'soft',
 'bot',
 'can',
 'extend',
 'up',
 'to',
 '72',
 'meters,',
 'and',
 'unlike',
 'plants,',
 'it',
 'can',
 'grow',
 'at',
 'a',
 'speed',
 'of',
 '10',
 'meters',
 'per',
 'second,',
 'the',
 'team',
 'reports',
 'July',
 '19',
 'in',
 'Science',
 'Robotics.',
 'The',
 'design',
 'could',
 'serve',
 'as',
 'a',
 'model',
 'for',
 'building',
 'robots',
 'that',
 'can',
 'traverse',
 'constrained',
 'environments',
 'This',
 'isn’t',
 'the',
 'first',
 'robot',
 'to',
 'take',
 'inspiration',
 'from',
 'plants.',
 'One',
 'plantlike',
 'predecessor',
 'was',
 'a',
 'robot',
 'modeled',
 'on',
 'roots.']

python

' '.join(re.findall('\S+', string))

'Robots are branching out. A new prototype soft robot takes inspiration from plants by growing to explore its environment. Vines and some fungi extend from their tips to explore their surroundings. Elliot Hawkes of the University of California in Santa Barbara and his colleagues designed a bot that works on similar principles. Its mechanical body sits inside a plastic tube reel that extends through pressurized inflation, a method that some invertebrates like peanut worms (Sipunculus nudus) also use to extend their appendages. The plastic tubing has two compartments, and inflating one side or the other changes the extension direction. A camera sensor at the tip alerts the bot when it’s about to run into something. In the lab, Hawkes and his colleagues programmed the robot to form 3-D structures such as a radio antenna, turn off a valve, navigate a maze, swim through glue, act as a fire extinguisher, squeeze through tight gaps, shimmy through fly paper and slither across a bed of nails. The soft bot can extend up to 72meters, and unlike plants, it can grow at a speed of 10meters per second, the team reports July 19 in Science Robotics. The design could serve as a model for building robots that can traverse constrained environments This isn’t the first robot to take inspiration from plants. One plantlike predecessor was a robot modeled on roots.'

. the dot matches any character excerpt the newline. 点匹配除换行符以外的任何字符。

python

string = '''Robots are branching out. A new prototype soft robot takes inspiration from plants by growing to explore its environment.
 
Vines and some fungi extend from their tips to explore their surroundings. Elliot Hawkes of the University of California in Santa Barbara and his colleagues designed a bot that works on similar principles. Its mechanical body sits inside a plastic tube reel that extends through pressurized inflation, a method that some invertebrates like peanut worms (Sipunculus nudus) also use to extend their appendages. The plastic tubing has two compartments, and inflating one side or the other changes the extension direction. A camera sensor at the tip alerts the bot when it’s about to run into something.
 
In the lab, Hawkes and his colleagues programmed the robot to form 3-D structures such as a radio antenna, turn off a valve, navigate a maze, swim through glue, act as a fire extinguisher, squeeze through tight gaps, shimmy through fly paper and slither across a bed of nails. The soft bot can extend up to 72meters, and unlike plants, it can grow at a speed of 10meters per second, the team reports July 19 in Science Robotics. The design could serve as a model for building robots that can traverse constrained environments
 
This isn’t the first robot to take inspiration from plants. One plantlike predecessor was a robot modeled on roots.'''

python

re.search('.+', string).group()  # no new line

'Robots are branching out. A new prototype soft robot takes inspiration from plants by growing to explore its environment.'

python

re.search('.+', string, flags = re.DOTALL).group()

'Robots are branching out. A new prototype soft robot takes inspiration from plants by growing to explore its environment.\n\nVines and some fungi extend from their tips to explore their surroundings. Elliot Hawkes of the University of California in Santa Barbara and his colleagues designed a bot that works on similar principles. Its mechanical body sits inside a plastic tube reel that extends through pressurized inflation, a method that some invertebrates like peanut worms (Sipunculus nudus) also use to extend their appendages. The plastic tubing has two compartments, and inflating one side or the other changes the extension direction. A camera sensor at the tip alerts the bot when it’s about to run into something.\n\nIn the lab, Hawkes and his colleagues programmed the robot to form 3-D structures such as a radio antenna, turn off a valve, navigate a maze, swim through glue, act as a fire extinguisher, squeeze through tight gaps, shimmy through fly paper and slither across a bed of nails. The soft bot can extend up to 72meters, and unlike plants, it can grow at a speed of 10meters per second, the team reports July 19 in Science Robotics. The design could serve as a model for building robots that can traverse constrained environments\n\nThis isn’t the first robot to take inspiration from plants. One plantlike predecessor was a robot modeled on roots.'

Creating your own character sets

[A-Z] '-' is a metacharacter when used in [] (custom character sets) '-'在[](自定义字符集)中使用时是一个元字符

python

string = 'Hello, There, How, Are, You'

python

re.findall('[A-Z]', string)  # pulls out all capital letters 取出所有大写字母

['H', 'T', 'H', 'A', 'Y']

python

re.findall('[A-Z,]', string)
# here we search for any capital letters or a comma
# 这里我们搜索大写字母或逗号

['H', ',', 'T', ',', 'H', ',', 'A', ',', 'Y']

python

string = 'Hello, There, How, Are, You...'
re.findall('[A-Z,.]', string)  # 在这里 . 只是一个字符集而不是之前所讲的所有非换行符

['H', ',', 'T', ',', 'H', ',', 'A', ',', 'Y', '.', '.', '.']

python

string = 'Hello, There, How, Are, You...'
re.findall('[A-Za-z,\s.]', string)  # 大写字母, 小写字母, 逗号, 非空白, 句点

['H',
 'e',
 'l',
 'l',
 'o',
 ',',
 ' ',
 'T',
 'h',
 'e',
 'r',
 'e',
 ',',
 ' ',
 'H',
 'o',
 'w',
 ',',
 ' ',
 'A',
 'r',
 'e',
 ',',
 ' ',
 'Y',
 'o',
 'u',
 '.',
 '.',
 '.']

Quantifers with custom sets

python

import re

+ 出现一次或更多
? 出现 0 或 1 次
* 出现 0 次或更多
{} 自定义出现次数

python

string = 'HELLO, There, How, Are, You...'

python

re.search('[A-Z]+', string)

<re.Match object; span=(0, 5), match='HELLO'>

python

re.findall('[A-Z]+', string)

['HELLO', 'T', 'H', 'A', 'Y']

python

re.findall('[A-Z]{2,}', string)  # 2 or more

['HELLO']

python

# one or more of 4 types of characters
# 四种字符中的一种或多种
re.search('[A-Za-z\s,]+', string).group()

'HELLO, There, How, Are, You'

python

re.findall('[A-Z]?[a-z\s,]+', string)

['O, ', 'There, ', 'How, ', 'Are, ', 'You']

python

# ^ is a metacharacter within brackets
# ^是括号中的元字符
# 表示相反
re.search('[^A-Za-z\s,]+', string).group()

'...'

python

re.findall('[^A-Z]+', string) # 匹配所有非大写字符

[', ', 'here, ', 'ow, ', 're, ', 'ou...']

GROUPS

groups allow us to pull out sections of a match and store them
- groups 允许我们提取匹配的部分并存储它们

python

# contrived example 举例
import re
string = 'John has 6 cats but I think my friend Susan has 3 dogs and Mike has 8 fishes'

python

re.findall('[A-Za-z]+ \w+ \d+ \w+', string)

['John has 6 cats', 'Susan has 3 dogs', 'Mike has 8 fishes']

the use of brackets denotes a group 使用括号表示一个组
- () = metacharacter 元字符

python

re.findall('([A-Za-z]+) \w+ \d+ \w+', string) # to pull out just the names 只把名字取出来

['John', 'Susan', 'Mike']

python

re.findall('[A-Za-z]+ \w+ \d+ (\w+)', string) # pull out animals 取出所有动物

['cats', 'dogs', 'fishes']

python

re.findall('([A-Za-z]+) \w+ (\d+) (\w+)', string)
# use original string to make sure matching is correct, then use groups to pull out the info you want
# 使用原始字符串确保匹配是正确的，然后使用组拉出你想要的信息

[('John', '6', 'cats'), ('Susan', '3', 'dogs'), ('Mike', '8', 'fishes')]

python

# organize the data by data-types 按数据类型组织数据
info = re.findall('([A-Za-z]+) \w+ (\d+) (\w+)', string)

python

info

[('John', '6', 'cats'), ('Susan', '3', 'dogs'), ('Mike', '8', 'fishes')]

Python3 zip() 函数

zip() 函数用于将可迭代的对象作为参数，将对象中对应的元素打包成一个个元组，然后返回由这些元组组成的对象，这样做的好处是节约了不少的内存。
与 zip 相反，zip(*) 可理解为解压，返回二维矩阵式

python

list(zip(*info))  # organize your data by categories 按类别组织数据

[('John', 'Susan', 'Mike'), ('6', '3', '8'), ('cats', 'dogs', 'fishes')]

python

match = re.search('([A-Za-z]+) \w+ (\d+) (\w+)', string)  # pulls out three groups 抽出三组
match

<re.Match object; span=(0, 15), match='John has 6 cats'>

python

match.group(0)

'John has 6 cats'

python

match.groups()

('John', '6', 'cats')

python

match.group(1)

'John'

python

match.group(2)

'6'

python

match.group(3)

'cats'

python

match.group(1, 3)  # multiple groups 多个组

('John', 'cats')

python

match.group(3, 2, 1, 1)  # change the order 改变顺序

('cats', '6', 'John', 'John')

python

match.span()

(0, 15)

python

match.span(2)

(9, 10)

python

match.span(3)

(11, 15)

python

match.start(3)

python

# find all has no group function
# re.findall 没有 group 函数
re.findall('([A-Za-z]+) \w+ (\d+) (\w+)', string).group(1)

---------------------------------------------------------------------------

AttributeError                            Traceback (most recent call last)

Input In [101], in <cell line: 3>()
      1 # find all has no group function
      2 # re.findall 没有 group 函数
----> 3 re.findall('([A-Za-z]+) \w+ (\d+) (\w+)', string).group(1)


AttributeError: 'list' object has no attribute 'group'

python

re.findall('([A-Za-z]+) \w+ (\d+) (\w+)', string)[0]

('John', '6', 'cats')

python

re.findall('([A-Za-z]+) \w+ (\d+) (\w+)', string)[0].group(1)  # 这也不好使

---------------------------------------------------------------------------

AttributeError                            Traceback (most recent call last)

Input In [39], in <cell line: 1>()
----> 1 re.findall('([A-Za-z]+) \w+ (\d+) (\w+)', string)[0].group(1)


AttributeError: 'tuple' object has no attribute 'group'

python

re.findall('([A-Za-z]+) \w+ (\d+) (\w+)', string)

[('John', '6', 'cats'), ('Susan', '3', 'dogs'), ('Mike', '8', 'fishes')]

python

data = re.findall('(([A-Za-z]+) \w+ (\d+) (\w+))', string)  # 组中组

python

data

[('John has 6 cats', 'John', '6', 'cats'),
 ('Susan has 3 dogs', 'Susan', '3', 'dogs'),
 ('Mike has 8 fishes', 'Mike', '8', 'fishes')]

python

# 你只能这么干
for i in data:
    print(i[3])

cats
dogs
fishes

we can use iteration
- 我们可以使用迭代
Python next() 函数

python

it = re.finditer('([A-Za-z]+) \w+ (\d+) (\w+)', string)
next(it).groups()

('John', '6', 'cats')

python

it = re.finditer('([A-Za-z]+) \w+ (\d+) (\w+)', string)
for element in it:
    print (element.group(1, 3, 2))   # don't forget iterators exhaust

('John', 'cats', '6')
('Susan', 'dogs', '3')
('Mike', 'fishes', '8')

python

it = re.finditer('([A-Za-z]+) \w+ (\d+) (\w+)', string)
for element in it:
    print(element.group())

John has 6 cats
Susan has 3 dogs
Mike has 8 fishes

python

it = re.finditer('([A-Za-z]+) \w+ (\d+) (\w+)', string)
for element in it:
    print(element.groups())

('John', '6', 'cats')
('Susan', '3', 'dogs')
('Mike', '8', 'fishes')

Naming Groups

python

import re

python

string = 'New York, New York 11369'

([A-Za-z\s]+) 寄件地址
([A-Za-z\s]+) 收件地址
(\d+) 邮编

python

match = re.search('([A-Za-z\s]+),([A-Za-z\s]+)(\d+)', string)

python

match.group(1), match.group(2), match.group(3), match.group(0)

('New York', ' New York ', '11369', 'New York, New York 11369')

?P< > to name a group-- group name inside the <>, followed by RE for group

(?P<City>)
(?P<State>)
(?P<ZipCode>)

python

pattern = re.compile('(?P<City>[A-Za-z\s]+),(?P<State>[A-Za-z\s]+)(?P<ZipCode>\d+)')

python

match = re.search(pattern, string)

python

match.group('City'), match.group('State'), match.group('ZipCode')

('New York', ' New York ', '11369')

python

match.group(1)

'New York'

python

match.groups()

('New York', ' New York ', '11369')

python

# Just incase you forget the names of the groups you used
# 以防您忘记了您使用的组的名称
match.groupdict()

{'City': 'New York', 'State': ' New York ', 'ZipCode': '11369'}

Quantifiers on groups

Using quantifiers on groups has some nuances, but very useful
- 在组上使用量词有一些细微差别，但非常有用

python

import re

python

string = 'abababababab'  # ab repeated many times
re.search('(ab)+', string)  #(ab)+ is many instances of one group repeated 同一组的许多实例重复出现

<re.Match object; span=(0, 12), match='abababababab'>

python

string = 'abababababab'  # ab repeated many times 重复了很多次
 
re.search('[ab]+', string)  # this is different

<re.Match object; span=(0, 12), match='abababababab'>

difference explained below
- (ab) 表示 a 和 b
- [ab] 表示 a 或 b

python

string = 'abababbbbbbb'  # only partial fit to our new string 只有部分符合我们的新字符串
re.search('(ab)+', string)

<re.Match object; span=(0, 6), match='ababab'>

python

string = 'abababbbbbbb'  # but this pattern fits perfectly 但这个模式完全吻合
re.search('[ab]+', string)

<re.Match object; span=(0, 12), match='abababbbbbbb'>

python

string = 'abababbbbbbb'  # allows flexibility 允许的灵活性
re.search('(ab)+\w+', string)

<re.Match object; span=(0, 12), match='abababbbbbbb'>

python

string = 'abababsssss'  # allows flexibility
re.search('(ab)+\w+', string)

<re.Match object; span=(0, 11), match='abababsssss'>

Nuances to be wary of

需要注意的细微差别

python

# only one group not multiple groups 只有一个组而不是多个组

python

string = 'abababababab' # original string
match = re.search('(ab)+', string) 
 
match.group(1)
# capturing only one group; value is overwritten each time
# 只捕获一个群体; 值每次都会被覆盖

'ab'

python

match.group(2) # no value 没有值

---------------------------------------------------------------------------

IndexError                                Traceback (most recent call last)

Input In [10], in <cell line: 1>()
----> 1match.group(2)


IndexError: no such group

python

match.groups()  # only one group, group just overwritten 只有一个组，组被覆盖了

('ab',)

python

match.group(0) # the full match, not related to groups 完全匹配，与组无关

'abababababab'

Another simple example with two groups using quantifiers
- 另一个使用量词的两个组的简单例子

python

string = 'ababababab'

python

match = re.search ('(ab)+(ab)+', string)
match

<re.Match object; span=(0, 10), match='ababababab'>

python

match.groups()

('ab', 'ab')

python

match.span(2) # the first group is greedy

(8, 10)

Only one group captured
- 只捕获了一个群体

python

string = '123456789'
 
match = re.search('(\d)+', string)

python

match

<re.Match object; span=(0, 9), match='123456789'>

python

(match.groups())  # only one group, and it uses the last value 只有一个组，它使用最后一个值

('9',)

Quantifiers with groups within findall

在 findall 中包含组的量词

python

string = '123456789'
 
re.findall('(\d)+', string)
# only pulls out group and last instance
# 只取出组和最后一个实例

['9']

python

string = '1234 56789'
re.findall('(\d)+', string)
# Here we have two matches 匹配了两个

['4', '9']

python

re.findall('((\d)+)', string)[1][0] 
# to find full match create a main group engulfing the smaller groups
# 要找到完全匹配，创建一个主组，吞噬较小的组

'56789'

python

# another example
string  = 'abbbbb ababababab'
re.findall('(ab)+', string)  # two instances

['ab', 'ab']

python

string  = 'abbbbb ababababab'
re.findall('((ab)+)', string)   #full match

[('ab', 'ab'), ('ababababab', 'ab')]

Groups for word completion

python

re.search('Happy (Valentines|Birthday|Anniversary)', 'Happy Birthday')

<re.Match object; span=(0, 14), match='Happy Birthday'>

python

re.search('Happy (Valentines|Birthday|Anniversary)', 'Happy Valentines')

<re.Match object; span=(0, 16), match='Happy Valentines'>

python

re.search('Happy Valentines| Happy Birthday | Happy Anniversary', 'Happy Valentines')

<re.Match object; span=(0, 16), match='Happy Valentines'>

Non-capture Groups

python

import re

python

# Here is one such example:
import re
 
string = '1234 56789'
re.findall('(\d)+', string)

['4', '9']

python

re.search('(\d)+', string).groups()  #using search

('4',)

捕获组（capturing group）是把多个字符当作一个单元对待的一种方式。通过把字符括在括号内创建捕获组。例如，正则表达式(dog)创建包含字母“d”、“o”和“g”的一个组。输入字符串和捕获组匹配的那一部分将被保存在内存中，以便以后通过反向引用再次使用。

而非捕获组就是输入字符串和捕获组匹配的那一部分将不被保存在内存中。

non-capture groups syntax
- ?: The symbol above represents non-capture groups and looks slightly similar to the syntax for naming groups
  - 上面的符号表示非捕获组，看起来有点类似于命名组的语法
- ?P don't confuse the two please.
  - 请不要混淆这两者

python

# comparison 比较

python

re.findall('(\d)+', string)

['4', '9']

python

re.findall('(?:\d)+', string)  # with non capture group 非捕获组

['1234', '56789']

So the group is part of the pattern, but we don't output the groups' results
- 所以 group 是模式的一部分，但我们不输出 group 的结果

python

re.findall('\d+', string)
# when RE has no groups in findall, we output entire match
# 当 RE 在 findall 中没有组时，我们输出整个匹配

['1234', '56789']

python

# Another example

python

string  = '123123 = Alex, 123123123 = Danny, 123123123123 = Mike, 456456 = rick, 121212 = John, 132132 = Luis,'

python

# We want to pull out all names whose ID has 123 within in
# 我们要取出所有 ID 包含 123 的名字

python

re.findall('(?:123)+ = (\w+),', string)  # three instances

['Alex', 'Danny', 'Mike']

python

# Another example
string = '1*1*1*1*22222  1*1*3333  2*1*2*1*222  1*2*2*2*333 3*3*3*444'

python

re.findall('(?:1\*){2,}\d+', string)

['1*1*1*1*22222', '1*1*3333']

Now, non-captured groups doesn't just affect the findall method
it also affects the search and match methods
- 现在，未捕获的组不仅影响 findall 方法——它还影响搜索和匹配方法

BE CAREFUL WITH SYNTAX

?: correct!
:? incorrect!

python

string = '1234 56789'
match = re.search('(?:\d)+', string)  # correct syntax
print(match.groups())

()

python

string = '1234 56789'
match = re.search('(:?\d)+', string)  # :? incorrect syntax!!!! 
print(match.groups())

('4',)

Summary:

when we capture groups we are either storing the value or outputting them.
- 当我们捕获组时，我们要么存储值，要么输出值。

Backreferences - Using captured groups inside other operations

反向引用——在其他操作中使用捕获的组

backreferencing is making a refererence to the captured group within the same regular expression
- 反向引用是在同一个正则表达式中引用捕获的组

python

# syntax and example

python

re.search(r'(\w+) \1','Merry Merry Christmas')  # Looking for repeated words 寻找重复的单词

<re.Match object; span=(0, 11), match='Merry Merry'>

python

re.search(r'(\w+) \1','Merry Merry Christmas').groups()

('Merry',)

\1 is just referencing the first group within the regular expression

‘\1’ 匹配的是所获取的第 1 个()匹配的引用。例如，’(\d)\1’ 匹配两个连续数字字符。如 33aa 中的 33

python

# Another example

python

re.findall(r'(\w+)','Happy Happy Holidays. Merry Christmas Christmas')

['Happy', 'Happy', 'Holidays', 'Merry', 'Christmas', 'Christmas']

python

re.findall(r'(\w+) \1','Happy Happy Holidays. Merry Christmas Christmas')   # Want to look for repeated words 想要寻找重复的单词

['Happy', 'Christmas']

python

# another example

python

re.findall(r'(\w+) \1','Merry Merry Christmas Christmas Merry Merry Christmas')

['Merry', 'Christmas', 'Merry']