正则表达式捕获最多2位数字和逗号（如果后面跟另一个单词和数字）

Question

I need a regular expression that matches and return 2 numbers from a string when conditions are met

1. only numbers with a maximum of 2 digits and not greater than 29 (might include a decimal case - so up to 2 digits plus 1 decimal case)

2. they must have in between either one of the characters y or + and after the second number the word 'houses'

And then capture both numbers

for the string below:

8 y 13 houses, 13 y 8 houses, 13 y 13 houses, 8 y 8 houses, 120 y 8 houses, 8 y 120 houses, 13,5 y 8 houses, 13,5 y 120 houses

I would get

8 and 13 / 13 and 8 / 13 and 13 / 8,8 / 13,5 and 5

I was trying with this

\b([0-9][0-9]?)\s[y|\+]\s([0-9]{1,2})\shouses\b

but can't manage to get the ',' as well.

Answer 1

You If you want to match the optional decimal value with an optional group:

re.compile(r"\b([1-2]?\d(?:,\d)?)\s[y+]\s([1-2]?\d(?:,\d)?)\shouses\b")

where (?:,[0-9])? will match a comma followed by a digit if present. Note that I limit the digit matching to values between 0 and 29; matching an optional 1 or 2 first, followed by 0-9 .

Demo:

>>> import re
>>> demo = '8 y 13 houses, 13 y 8 houses, 13 y 13 houses, 8 y 8 houses, 120 y 8 houses, 8 y 120 houses, 13,5 y 8 houses, 13,5 y 120 houses'
>>> pattern = re.compile(r"\b([1-2]?\d(?:,\d)?)\s[y+]\s([1-2]?\d(?:,\d)?)\shouses\b")
>>> pattern.findall(demo)
[('8', '13'), ('13', '8'), ('13', '13'), ('8', '8'), ('13,5', '8')]

Answer 2

Here's a try:

#! /usr/bin/env python

import re

str = '8 y 13 houses, 13 y 8 houses, 13 y 13 houses, 8 y 8 houses, 120 y 8 houses, 8 y 120 houses, 13,5 y 8 houses, 13,5 y 120 houses'

regex = r'''
\b (
    [012]?     # number may go up to 29, so could have a leading 0, 1, or 2
    [0-9]      # but there must be at least one digit 0-9 here
    (,[0-9])?  # and the digits might be followed by one decimal point
)
\s* [y+] \s*   # must be a 'y' or '+' in between
(
    [012]?     # followed by another 0-29
    [0-9]
    (,[0-9])?  # and an optional decimal point
)
\s* houses \b  # followed by the word "houses"
'''

for match in re.finditer(regex, str, re.VERBOSE):
    print "found: %s and %s" % (match.group(1), match.group(3))

Demonstration:

$ python pyregex.py 
found: 8 and 13
found: 13 and 8
found: 13 and 13
found: 8 and 8
found: 13,5 and 8

When that regex matches a string in your input, the first number will be in match.group(1) and the second number will be in match.group(3) .