Regexp to find words longer than 'X' in a given string

[themainman]

Does anyone know if it is possible to build a regexp that can search a string $message for words longer than 30 charcters and if so break the word at 30 with a " - "

[trib4lmaniac]

$string = preg_replace('/\w{30}/', "$0-", $string); 

[themainman]

thanks tri4almaniac, but it doesnt seem to work

thanks for trying anyway,
mainman

[themainman]

any1?

[themainman]

bump

[trib4lmaniac]

That worked for me, at least as I understood your question. Why not post some sample input and output?

[bokehman]

It works for me. Post what doesn't work for you.

[themainman]

im using what you gave me:

$message = preg_replace('/\w{30}/', "$0-", $message);

and say a user sends a message to another user, with a word like aaaaaaaaaaaaaaaaaaaaaaaaaaaaaa which would break the table its displayed in, i want to cut that to aaaaaaaaaaaa - aaaaaaaaaaaa

Thanks

[7stud]

Does this:

say a user sends a message to another user, with a word like aaaaaaaaaaaaaaaaaaaaaaaaaaaaaa which would break the table its displayed in, i want to cut that to aaaaaaaaaaaa - aaaaaaaaaaaa

match what you said here:

that can search a string $message for words longer than 30 charcters and if so break the word at 30

You neither have a word longer than 30 characters, nor does it break the word after 30 characters.

[trib4lmaniac]

Well that's cutting the word after 15 characters; not 30 as your first post stated. Just change the 30 to a 15 in the original snippet (now with spaces). Although on reflection I'd actually recommend using:

$message = preg_replace('/\w{15}(?=\w)/', "$0 - ", $message); 

As that causes the expression to refrain from adding a trailing hypen to the tail end of the word.

[7stud]

Well that's cutting the word after 15 characters

Count again. Or, rather don't and wait a week until the poster makes up his mind.

[trib4lmaniac]

Count again. Or, rather don't and wait a week until the poster makes up his mind.

Curses, didn't bother to actually count the individual letters

If you want to split the words a little more intelligently you're going to need more than a single regular expression.
I think the one I just posted will suffice for stopping the tables expanding waistband.

[Icheb]

Alternatively use explode() to put the string into an array and walk through the array to find words that are longer with strlen(). Would probably be easier than with a regular expression.

[trib4lmaniac]

I think the regex is by far the simplest. At least to that degree of accuracy.

[Icheb]

How is that more accurate? Unless strlen() stops working both provide equally accurate results. And for a beginner it's probably much easier to comprehend than regular expressions.

[themainman]

ok thanks for the help guys, and sorry i explained it badly

[trib4lmaniac]

How is that more accurate? Unless strlen() stops working both provide equally accurate results. And for a beginner it's probably much easier to comprehend than regular expressions.

I meant more simplistic for that degree of accuracy. If you write functionally equivalent code with string functions and make it look even simpler I will be very suprised.
Learn regular expressions if you haven't already. They are your key to happiness

[Icheb]

I meant more simplistic for that degree of accuracy. If you write functionally equivalent code with string functions and make it look even simpler I will be very suprised.

I guess you didn't get the phrase "And for a beginner it's probably much easier to comprehend than regular expressions".

[trib4lmaniac]

I guess you didn't get the phrase "And for a beginner it's probably much easier to comprehend than regular expressions".

We'll have to agree to differ I suppose.

[Icheb]

So you really want to tell me that some construction with strlen() etc. is more difficult to understand to a BEGINNER than a regular expression? You are absolutely hilarious.

[exam]

As a side note, wordwrap might be what you're looking for.

[bokehman]

As a side note, wordwrap might be what you're looking for.

Wordwrap doesn't work in this context.

[exam]

Wordwrap doesn't work in this context.

In retrospect, I have to agree with you. What I meant to say was that maybe instead of what the OP was trying to do, he could do with wordwrap, which does allow you to split long words. But after reconsidering, I don't think it'd be the best option.

[trib4lmaniac]

So you really want to tell me that some construction with strlen() etc. is more difficult to understand to a BEGINNER than a regular expression? You are absolutely hilarious.

function word_split($text, $divider, $length) {
    $text = explode(" ", $text);
    foreach($text as $key => $word) {
        $word = chunk_split($word, $length, " - ");
        $text[$key] = substr($word, 0, -strlen($divider));
    }
    $text = join($text, " ");
    return $text;
}

$message = word_split($message, " - ", 15); 

That does it very similarly, except it only splits at a single spaces. I'm not sure how it would play with things like tabs, multiple spaces and punctuation. Probably not nicely.

Compare to:

$message = preg_replace('/\w{15}(?=\w)/', "$0 - ", $message); 

I think the regex is simpler and better