Split string with regex from end

Question

My question is about function clojure.string/split. One can specify maximum number of splits for the function, and it works like a charm:

user> (clojure.string/split "1{1,2{3,4},5}6" #"\{" 2)
;; => ["1" "1,2{3,4},5}6"]

However, the function traverses the string from left to right. Sometimes I want it to traverse a string form right to left (from end):

user> (clojure.string/split "1{1,2{3,4},5}6" #"\}" 2)
;; => ["1{1,2{3,4" ",5}6"]
;; desired result: ["1{1,2{3,4},5" "6"]

How can I achieve it using regex?

Jerry · Accepted Answer · 2014-09-10 12:28:17Z

1

You could try using a negative lookahead in your particular case to ensure that there is no more } after the one you're splitting at:

user> (clojure.string/split "1{1,2{3,4},5}6" #"\}(?![^\}]*\})" 2)

(?![^\}]*\}) is a negative lookahead and will prevent a match if there is another } after the } matched. I'm using the negated class [^\}]* to make it faster than using something like .* and I'm not entirely sure if you need the escape since I'm not familiar with clojure. Usually, you can safely use \}(?![^}]*\}), but escaping will work whether it is required or not.

answered Sep 10, 2014 at 12:28

Jerry

71.8k14 gold badges106 silver badges148 bronze badges

Sign up to request clarification or add additional context in comments.

7 Comments

Mark Karpov Over a year ago

Wow, it works, but I do not understand how :-( You said you was using 'negated class' to make it faster. Can you post also slower and simpler variant (just to start digging). P.S. One should use escape character because clojure uses java regex and without it clojure throws PatternSyntaxException.

Jerry Over a year ago

@Mark The difference in speed is not much different, but here it is: #"\}(?!.*?\})" The way the negative lookahead works is, if what's inside matches, it fails the match. So if .*?\} matches, the match will fail. .*? will match any character, but it checks one character at a time. This is more suitable for instances where the string is short. #"\}(?!.*\})" (without ?) will do the same thing, except this one works when the last } is near to the end of the string and can be very slow if the two } are at the beginning of a very long string. [^\}]* works quickly in any situation.

Jerry Over a year ago

About the escaping, I said that because most engines, including Java doesn't need escaping when you put a meta character in a character class, so that: [\}] is the same as [}]. But there's so much variety of things out there, one can never be too sure :)

Mark Karpov Over a year ago

Thanks for explaining the 'machinery' :-) I like to understand what's going on in my code.

Jerry Over a year ago

@Mark Should you need some more elaboration, let me know. I ran out of characters in that comment above lol.

|

Collectives™ on Stack Overflow

Split string with regex from end

1 Answer 1

7 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

7 Comments

Your Answer

Sign up or log in

Post as a guest

Related