用于在Python中查找模式的正则表达式

Question

I am having a trouble with adjusting regular expression formula to do what I want.

I want to find a pattern:

AxxHxxxAxxHxxxbbbbbAxxHxxxAxxHxxx

But not:

AxxHxxxAxxHxxxAxxHxxxAxxHxxxAxxHxxxAxxHxxxAxxHxxxAxxHxxx

If I use:

"(A\w{2}H\w{3}){2,4}.+(A\w{2}H\w{3}){2,4}"

It will find both. I've tried excluding long stretches of AxxHxxx repeat by:

"(?!(A\w{2}H\w{3}){8})(A\w{2}H\w{3}){2,4}.+(A\w{2}H\w{3}){2,4}"

But it's actually not working. Do any of you have any ideas how to solve this problem? Since I will be operating rather on the big set of data, I would like to preferably avoid loop hell with splicing strings and so on.

Thanks in advance!

Edit:

As there was an interest in another set of examples I will try to explain in more details what I am trying to accomplish.

I want to find in a single string a fragment which will consist of two repetition and a set of any characters in between, such as:

.......A..H...A..H...A..H...............A..H...A..H...A..H.......................

where "." is any character and you have two repetetive modules of "A..H...". The only problem is that I do not want to find continues stretch of repetition such as here:

.......A..H...A..H...A..H...A..H...A..H...A..H......................

Answer 1

This seems to produce the correct result with your test cases, but it is very ugly.

/^(?:A.{2}H.{3})+(?:(?!A.{2}H.{3}).)+(?:A.{2}H.{3})+$/

http://regex101.com/r/sW5vG2

Is this what you want?

The efficiency will be directly proportional to the length of the string.

Edit: Improved solution using PCRE http://regex101.com/r/wX8uK2

Answer 2

I don't know how much I understood your problem.

For solving your specific problem , You can use something like this , I think so.

^(.*)\1((?!\1).)+\1{2}$

Check DEMO