Stripping a string

[Tekime]

What would be the fastest way to decrease the length of a string?

For example, say I have a string $news_contents that is about 500 characters long. What would be the quickest way to strip all but the first 255 characters?

(Even a push in the right direction would help, most of my attempts seem like more work than necessary).

Thanks, all 🙂

[sarah]

I like this bit when it shouldn't be broken off in the middle of the word...

$fullText="This is the full text that I am going to summarize so that it only prints out part of it, broken up by words.";
//
$chars = 30;
$summary = ereg_replace("^(.{1,$chars})[ .,].*","\1", $fullText);
//
print("$summary");

[keith73]

use substr()

http://www.php.net/manual/en/function.substr.php

like this:

$shortstring = substr($longstring,0,255);

the first argument to substr is the original string. the second argument is the starting point and the third arg is the # of chars to return.

• keith

[Tekime]

Thanks for the quick replies 🙂

I like the idea of stripping only between words, but when I run that code sarah, all it prints out is "1", lol... I'll look at it and see if I'm missing something.

[sarah]

You're the second person who has said that and I just figured out what the deal was...There are supposed to be two backslashes before the 1 in "1"...When I put it inside the PHP code tags, it messes with the slashes... Add the slashes and it will work...

$summary = ereg_replace("^{(.{1,$chars})[} .,].*","\1", $fullText);

[Tekime]

It works perfectly now, thanks sarah!

[Tekime]

Hmm, I added this into applicationa nd it seems to take an extremely long time to process. If I cut at about 40-50 characters it works okay, but near 100+ it starts to crawl big time. Is this an inherent problem with ereg_replace or do you think I could be using it inefficiently? I thought my script broke the first parse at 255 chars, but it just took about a minute.

(This IS on a P55C 233MHz server, but things usually run much faster).

Thanks again 🙂

[Tekime]

Well, with a little more research I've found that ereg_replace really is that slow. I always knew it was slow, but wow...

Searching the forums I found this bit of code:

    $content = $full_text;
    $content=substr($content,0,$chars);
    $content=substr($content,0,strrpos($content," "));

Which works very well, and quite a bit faster. I may try rewriting the ereg_replace with a preg_replace or str_replace later on, but this seems good for now.

[sarah]

Perfect...I was just about to submit the following alternative...

$summary=substr($summary,0,strrpos(substr($fullText,0,$chars)," "));

[Tekime]

Ahh, the condensed version 🙂 This appeals to my old Perl-ian ways (not that I was ever much a master Golfer, but I do like condensation (which I'm not sure is even a real word in this context)).

Thanks again for all your help, 🙂

[sarah]

This is exactly the reason I love these forums...There's always a better way...Looks like I'm doin' a little code revision today...🙂