Php - Remove repeated words in a string

Question

Input: ball ball code

Output should be: ball code

Input: awycodeawy

Output should be: awycode

I tried these, but didn't work:

$q = preg_replace("/\s(\w+\s)\1/i", "$1", $q);
$q = preg_replace("/s(w+s)1/i", "$1", $q);

regex does not retain state, right? so why would this be a pure regex solution? somebody? — Kristian
– Kristian, Commented May 18, 2012 at 15:20
@Kristian: They can. To some extend. In particular anything that just repeats or follows a pattern is well suited for pattern matching. regular-expressions.info/brackets.html — mario
– mario, Commented May 18, 2012 at 15:23
@Mark Byers "aabaaabaabaaa" should be "aababaaba". 2 or more chars is considered. — user1055645
– user1055645, Commented May 18, 2012 at 15:33

anubhava · Accepted Answer · 2012-05-18 15:59:20Z

5

Here is positive lookahead base attempt on regex based solution to OP's problem.

$arr = array('ball ball code', 'abcabc bde bde', 'awycodeawy');
foreach($arr as $str)
   echo "'$str' => '" . preg_replace('/(\w{2,})(?=.*?\\1)\W*/', '', $str) ."'\n";

OUTPUT

'ball ball code' => 'ball code'
'abcabc bde bde' => 'abc bde'
'awycodeawy' => 'codeawy'

As you can for the input 'awycodeawy' it makes it to 'codeawy' instead of 'awycode'. The reason is that it is possible to find a variable length lookahead something which is not possible for lookbehind.

answered May 18, 2012 at 15:59

anubhava

790k67 gold badges603 silver badges671 bronze badges

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

5 Comments

user1055645 Over a year ago

Thank you. May I ask how much system resources does this code use? Is it make sense it to use for a website that run this code 3 time per minute. (Forgive me for being a noob :))

anubhava Over a year ago

@noarm: In the present form this code is not resource intensive however I think resource usage depends on the length of the input string. If your input string is huge then it can take a while to scan through all the string to find duplicate.

Junior Mayhe Over a year ago

it is not recommended to use it for testing against user sentences. This could harm the original string.

anubhava Over a year ago

@JuniorMayhé: Original string is just echoed here not being updated.

Junior Mayhe Over a year ago

My mistake I meant if someone try to update the original string, which is a sentence, it would mess with the original. I see it is being echoed ;-)

Amber · Accepted Answer · 2012-05-18 15:19:27Z

4

$q = preg_replace("/\b(\w+)\s+\\1\b/i", "$1", $q);

answered May 18, 2012 at 15:19

Amber

531k89 gold badges643 silver badges558 bronze badges

6 Comments

user1055645 Over a year ago

"abcabc bde bde" becomes "abcabc bde". Can't it be "abc bde"?

Amber Over a year ago

@noarm You said you wanted repeated words, not repeated parts of words. If you want it to apply to parts, however, change the \s+ in the middle to a \s*.

Amber Over a year ago

@anubhava It was never intended to. Trying to match occurrences later on like that is nearly impossible to do well with regex (without breaking other things, like making book into bok, or mississippi into missippi).

anubhava Over a year ago

@Amber: Yes, that's the whole point. The way problem was stated by OP cannot be solved by a pure regex solution.

Mark Byers Over a year ago

The OP gave some examples of what he wants. This answer doesn't work for the second example he gave: awycodeawy -> awycode.

|

Collectives™ on Stack Overflow

Php - Remove repeated words in a string

2 Answers 2

OUTPUT

5 Comments

6 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

OUTPUT

5 Comments

6 Comments

Your Answer

Sign up or log in

Post as a guest

Related