您如何在python中编写正则表达式，以查找仅包含字母，数字和下划线的所有单词？

Question

This is the best I was able to come up with:

b = re.findall(r'\b[a-zA-Z0-9_]\b', 'ahz2gb_ $f heyght78_')

But that doesn't work. Also, not that I'm only interested in regexes at the moment. I can solve the problem the long way.

The expected result is a list containing [ahz2gb_, heyght78_]

Answer 1

There is \\w for capturing those characters, and you need to allow more than one character with + :

b = re.findall(r'\b\w+\b', 'ahz2gb_ $f heyght78_')

As + is greedy, you don't really need the \\b either:

b = re.findall(r'\w+', 'ahz2gb_ $f heyght78_')

If you need words to be split by white space only (not \\b ), then you can use look-around:

b = re.findall(r'(?<!\S)\w+(?!\S)', 'ahz2gb_ $f heyght78_')

The (?<! sequence means: look back to see you don't have the pattern that follows (?<! preceding the current matching position in the target string. So in this case (?<!\\S) means: there should not be a preceding non-white-space character.

Then (?! is similar, but looking forward (without matching).

Answer 2

Simple to understand will be regex ..

• ^[0-9a-zA-Z_]+$ : strictly numbers, alphabets and underscore
• ^[0-9a-zA-Z_ ]+$ : strictly numbers, alphabets, underscore and spaces

If you need words from the matched lines, then spilt using space as delimiter.

You can try python regex online on http://pythex.org/

Sample Run on IDLE

>>> import re
>>> re.findall(r'^[a-zA-Z0-9_ ]+$', 'ahz2gb_ f heyght78_')[0].split(' ')
['ahz2gb_', 'f', 'heyght78_']

---

EDIT : Given new requirement of only having words, here is how you can achieve the same.

import re
mylist =  'ahz2gb_ $f heyght78_'.split(' ')
r = re.compile("^[0-9a-zA-Z_]+$")
newlist = list(filter(r.match, mylist))
print(newlist)

Wish, I could shorten it!!

Sample Run

========= RESTART: C:/regex.py =========
['ahz2gb_', 'heyght78_']