查找出现在字符串中单词之前的子字符串直到一个数字

Question

I have a string :

"abc mysql 23 rufos kanso engineer"

I want the regex to output the string before the word "engineer" till it sees a number.

That is the regex should output :

23 rufos kanso

Another example:

String:

def grusol defno 1635 minos kalopo, ruso engineer okas puno"

I want the regex to output the string before the word "engineer" till it sees a number.

That is the regex should output :

1635 minos kalopo, ruso

I am able to achieve this by a series of regex .

Can I do this in one shot?

Thanks

Answer 1

The pattern I'd use: ((\\d+)(?!.*\\d).*)engineer -- it looks for the latest digit and goes from there.

Something similar to (\\d.*)engineer would also work but only if there's only one digit in the string.

---

>>> import re
>>> string = '123 abc mysql 23 rufos kanso engineer'
>>> pattern = r'((\d+)(?!.*\d).*)engineer'
>>> re.search(pattern, string).group(1)
'23 rufos kanso '
>>>

---

Edit

In case there are digits after the 'engineer' part, the pattern mentioned above does not work, as you have pointed out in the comment. I tried to solve it, but honestly I couldn't come up with a new pattern (sorry).

The workaround I could suggest is, assuming 'engineer' is still the 'key' word, splitting your initial string by said word.

Here is the illustration of what I mean:

>>> string = '123 abc mysql 23 rufos kanso engineer 1234 b65 de'
>>> string.split('engineer')
['123 abc mysql 23 rufos kanso ', ' 1234 b65 de']
>>> string.split('engineer')[0] 
'123 abc mysql 23 rufos kanso '

# hence, there would be no unexpected digits

>>> s = string.split('engineer')[0]
>>> pattern = r'((\d+)(?!.*\d).*)'
>>> re.search(pattern, s).group(1)
'23 rufos kanso '

Answer 2

Use positive look-ahead to match until the word engineer preceded by a digit.

The regex - (?=\\d)(.+)(?=engineer)

Just to get an idea:

import re
pattern = r"(?=\d)(.+)(?=engineer)"
input = [ "\"def grusol defno 1635 minos kalopo, ruso engineer okas puno\"", "\"abc mysql 23 rufos kanso engineer\"" ]

matches = []

for item in input:
    matches.append(re.findall(pattern, item))

Outputting:

[['1635 minos kalopo, ruso '], ['23 rufos kanso ']]

Answer 3

Have a look at this site . It is great to play around with regex and it explains every steps.
Here is a solution to your problem: link