我如何结合这两个正则表达式？

Question

I would like to combine the following two regexes into one:

• ^([AZ]{1,2}) ?([0-9]{1,4})$ eg AB 1234
• ^([0-9]{1,4}) ?([AZ]{1,2})$ eg 1234 AB

I thought this would be a simple as this:

^([AZ]{1,2}) ?([0-9]{1,4})|([0-9]{1,4}) ?([AZ]{1,2})$

However, the above always returns 4 groups, eg

1. ''
2. ''
3. '1234'
4. 'AB'

How can I combine these two regexes in a way that will always returns two groups? eg

1. '1234'
2. 'AB'

or

1. 'AB'
2. '1234'

Answer 1

Firstly your combination is not quite right, because the ^ only applies to the first alternative and the $ only applies to the second one. So you need to group the alternation:

^(?:([A-Z]{1,2}) ?([0-9]{1,4})|([0-9]{1,4}) ?([A-Z]{1,2}))$

Now what you want to achieve can not be done with all regex engines, but some (eg PCRE) support a special alternation construct where capturing groups are counted individually for all alternations. This is the syntax:

^(?|([A-Z]{1,2}) ?([0-9]{1,4})|([0-9]{1,4}) ?([A-Z]{1,2}))$

EDIT:

Unfortunately, this is in particular not supported by Python. Neither is the alternative of reusing a named capturing group. Hence you will probably have to filter empty strings out of match.groups() or stick to two regular expressions.