[PHP] file_get_contents and regex

Albino · 06-20-2008, 11:25 AM

Hey,

I have this terrible feeling i'm missing something really simple here.

I'm trying to get my head around file_get_contents, combined with regex, to get specific bits of data from web pages.

Lets say I want the about my part on my plurk page: http://www.plurk.com/user/Toddish

the source code is:

Code:

 <p id="about_me">
                    I'm a web designer from the UK who loves games, music, and films :)
<br>If you like any of my Plurks, feel free to add me!
                </p>

so, I tried this:
????: NamePros.com http://www.namepros.com/programming/483965-php-file_get_contents-and-regex.html

PHP Code:

  <?php

$data = file_get_contents('http://www.plurk.com/user/Toddish');

$regex = '/<p id="about_me">[.*]<\/p>/';

preg_match($regex,$data,$match);

var_dump($match);

?>

but I get nothing.

any idea what I'm missing?

cheers, rep etc as usual

Bruce_KD · 06-20-2008, 12:01 PM

You're not escaping any characters in the regex match.
The white space is also screws it up. If you really need that, there are workarounds, but (.*) excludes white space.
????: NamePros.com http://www.namepros.com/showthread.php?t=483965

Code:

$data = str_replace("\n", '', file_get_contents('http://www.plurk.com/user/Toddish'));
echo preg_match("/\<p id\=\"about_me\"\>(.*)<\/p\>/",$data,$match);
print_r($match);

Bruce

P.S. I think this is the forum for code snippets you're sharing with others; it's parent forum (just "Programming") is for code help. Though I may be wrong, I'm fairly new

Albino · 06-20-2008, 12:29 PM

cheers, didn't realise you had to escape all of them as well.

Now I am getting all the text after it aswell. I tried to limit it via:

PHP Code:

  /\<p id\=\"about_me\"\>(.*){1,400}<\/p\>/ 

????: NamePros.com http://www.namepros.com/showthread.php?t=483965

but I now get NULL.

cheers so far, rep added

oh, and it does look as though I mis clicked on the wrong forum, if a mod cold move it please

Bruce_KD · 06-20-2008, 01:25 PM

You missed escaping the first tag on both p's and equals sign.
Try

Code:

/\<p id\=\"about_me\"\>(.{1,400})<\/p\>/

Bruce

Albino · 06-21-2008, 03:22 AM

I can't even copy and paste properly, oh dear!

my actual code is:

PHP Code:

  preg_match("/\<p id\=\"about_me\"\>(.*){1,1000}<\/p\>/",$data,$match); 

????: NamePros.com http://www.namepros.com/showthread.php?t=483965

Barrucadu · 06-21-2008, 04:06 AM

This will just match up until the first , even across multiple lines:

PHP Code:

  preg_match("/\<p id\=\"about_me\"\>(.*?)<\/p\>/s",$data,$match); 

????: NamePros.com http://www.namepros.com/showthread.php?t=483965

Albino · 06-21-2008, 06:49 AM

perfect, thanks

Albino · 06-21-2008, 12:10 PM

sorry, another few small things

im not sure on the (.*?). the ? means 0 or 1, so surely that would mean 0 or 1 of any character? but i'm guessing it references the instead, just not sure how.

Also, i need a .htaccess file to change some urls. So I have this:

Code:

RewriteEngine on
RewriteRule ^plurk/([a-z]+)/$ /plurk.php?user=$1

to turn /todd/plurk.php?user=Toddish to /todd/plurk/Toddish
????: NamePros.com http://www.namepros.com/showthread.php?t=483965

but this doesn't seem to work, I get a 404

any ideas?

Bruce_KD · 06-22-2008, 03:23 PM

I got this working as follows:

Code:

RewriteEngine on
RewriteRule plurk/(.*) plurk.php?user=$1

I also tested Mikor's regex, and it does match across multiple lines.
If you want to throw a number of characters limit in, you can use this:

Code:

$data = file_get_contents('http://www.plurk.com/user/Toddish');
preg_match("/\<p id\=\"about_me\"\>(.*?)<\/p\>/s",$data,$match); 
print_r($match);

????: NamePros.com http://www.namepros.com/showthread.php?t=483965
Bruce

qbert220 · 06-23-2008, 05:45 AM

Originally Posted by Albino

im not sure on the (.*?). the ? means 0 or 1, so surely that would mean 0 or 1 of any character?

The ? has 2 different meanings, depending on context.
????: NamePros.com http://www.namepros.com/showthread.php?t=483965

After a normal character or expression is means one or zero occurrences of.

After a normally "greedy" operator it makes it non-greedy.

Greedy means that the operator will match as much as possible. * is normally greedy, so the if you just use .* it will match everything until the last on the page. By making it non-greedy, it matches as little as it can. So it matches up to the next on the page, i.e. up to the at the end of the paragraph.

Albino · 06-24-2008, 05:40 AM

cheers, bruce,works fine

and thanks for the explanation qbert, that helped a lot!

06-20-2008, 11:25 AM	THREAD STARTER #1 (permalink)
Albino Munky Designs Join Date: May 2005 Posts: 996	[PHP] file_get_contents and regex Hey, I have this terrible feeling i'm missing something really simple here. I'm trying to get my head around file_get_contents, combined with regex, to get specific bits of data from web pages. Lets say I want the about my part on my plurk page: http://www.plurk.com/user/Toddish the source code is: Code: <p id="about_me"> I'm a web designer from the UK who loves games, music, and films :) <br>If you like any of my Plurks, feel free to add me! </p> so, I tried this: ????: NamePros.com http://www.namepros.com/programming/483965-php-file_get_contents-and-regex.html PHP Code: `<?php $data = file_get_contents('http://www.plurk.com/user/Toddish'); $regex = '/<p id="about_me">[.*]<\/p>/'; preg_match($regex,$data,$match); var_dump($match); ?>` but I get nothing. any idea what I'm missing? cheers, rep etc as usual __________________ Toddish.co.uk - Portfolio/Blog Powcomics.com - Webcomic Hosting/Directory Erant.co.uk - vent your rage!

06-20-2008, 12:01 PM	#2 (permalink)
Bruce_KD NamePros Member Join Date: Sep 2006 Posts: 99	You're not escaping any characters in the regex match. The white space is also screws it up. If you really need that, there are workarounds, but (.) excludes white space. ????: NamePros.com http://www.namepros.com/showthread.php?t=483965 Code: $data = str_replace("\n", '', file_get_contents('http://www.plurk.com/user/Toddish')); echo preg_match("/\<p id\=\"about_me\"\>(.)<\/p\>/",$data,$match); print_r($match); Bruce P.S. I think this is the forum for code snippets you're sharing with others; it's parent forum (just "Programming") is for code help. Though I may be wrong, I'm fairly new __________________ Need a CAPTCHA? www.KingdomsDivided.com

06-20-2008, 12:29 PM	THREAD STARTER #3 (permalink)
Albino Munky Designs Join Date: May 2005 Posts: 996	cheers, didn't realise you had to escape all of them as well. Now I am getting all the text after it aswell. I tried to limit it via: PHP Code: `/\<p id\=\"about_me\"\>(.*){1,400}<\/p\>/ ????: NamePros.com http://www.namepros.com/showthread.php?t=483965` but I now get NULL. cheers so far, rep added oh, and it does look as though I mis clicked on the wrong forum, if a mod cold move it please __________________ Toddish.co.uk - Portfolio/Blog Powcomics.com - Webcomic Hosting/Directory Erant.co.uk - vent your rage!

06-20-2008, 01:25 PM	#4 (permalink)
Bruce_KD NamePros Member Join Date: Sep 2006 Posts: 99	You missed escaping the first tag on both p's and equals sign. Try Code: /\<p id\=\"about_me\"\>(.{1,400})<\/p\>/ Bruce __________________ Need a CAPTCHA? www.KingdomsDivided.com

06-21-2008, 03:22 AM	THREAD STARTER #5 (permalink)
Albino Munky Designs Join Date: May 2005 Posts: 996	I can't even copy and paste properly, oh dear! my actual code is: PHP Code: `preg_match("/\<p id\=\"about_me\"\>(.*){1,1000}<\/p\>/",$data,$match); ????: NamePros.com http://www.namepros.com/showthread.php?t=483965` __________________ Toddish.co.uk - Portfolio/Blog Powcomics.com - Webcomic Hosting/Directory Erant.co.uk - vent your rage!

06-21-2008, 04:06 AM	#6 (permalink)
Barrucadu Senior Member Join Date: Aug 2005 Location: East Yorkshire, England Posts: 2,689 Follow @barrucadu	This will just match up until the first </p>, even across multiple lines: PHP Code: `preg_match("/\<p id\=\"about_me\"\>(.?)<\/p\>/s",$data,$match); ????: NamePros.com http://www.namepros.com/showthread.php?t=483965` __________________ Website \| Blog* Arch Hurd developer; GNU Webmaster; FSF member #8385

06-21-2008, 06:49 AM	THREAD STARTER #7 (permalink)
Albino Munky Designs Join Date: May 2005 Posts: 996	perfect, thanks __________________ Toddish.co.uk - Portfolio/Blog Powcomics.com - Webcomic Hosting/Directory Erant.co.uk - vent your rage!

06-21-2008, 12:10 PM	THREAD STARTER #8 (permalink)
Albino Munky Designs Join Date: May 2005 Posts: 996	sorry, another few small things im not sure on the (.?). the ? means 0 or 1, so surely that would mean 0 or 1 of any character? but i'm guessing it references the </p> instead, just not sure how. Also, i need a .htaccess file to change some urls. So I have this: Code: RewriteEngine on RewriteRule ^plurk/([a-z]+)/$ /plurk.php?user=$1 to turn /todd/plurk.php?user=Toddish to /todd/plurk/Toddish ????: NamePros.com http://www.namepros.com/showthread.php?t=483965 but this doesn't seem to work, I get a 404 any ideas? __________________ Toddish.co.uk - Portfolio/Blog Powcomics.com - Webcomic Hosting/Directory Erant.co.uk - vent your rage! Last edited by Albino; 06-21-2008 at 01:11 PM.*

06-22-2008, 03:23 PM	#9 (permalink)
Bruce_KD NamePros Member Join Date: Sep 2006 Posts: 99	I got this working as follows: Code: RewriteEngine on RewriteRule plurk/(.) plurk.php?user=$1 I also tested Mikor's regex, and it does match across multiple lines. If you want to throw a number of characters limit in, you can use this: Code: $data = file_get_contents('http://www.plurk.com/user/Toddish'); preg_match("/\<p id\=\"about_me\"\>(.?)<\/p\>/s",$data,$match); print_r($match); ????: NamePros.com http://www.namepros.com/showthread.php?t=483965 Bruce __________________ Need a CAPTCHA? www.KingdomsDivided.com

06-24-2008, 05:40 AM	THREAD STARTER #11 (permalink)
Albino Munky Designs Join Date: May 2005 Posts: 996	cheers, bruce,works fine and thanks for the explanation qbert, that helped a lot! __________________ Toddish.co.uk - Portfolio/Blog Powcomics.com - Webcomic Hosting/Directory Erant.co.uk - vent your rage!

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)