Grabbing the first paragraph

[karinne]

Ok ... a while back a friend of mine gave me this script that would grab the text in <p>...</p> from an article and display it. For some reason ... I just noticed that it display the 2 paragraph instead of the first one like I want it to.

I have no idea how to fix this ... I suck at regular expressions and things like that in php. Anyways ... here's the code:

PHP: Select all

function paraSummary($subject) {
$matches = array();
preg_match("/(<p>[^<]+<\/p>)/", $subject, $matches);
return $matches[1];
} 

I tried changing the return $matches[1]; to return $matches[0]; but that didn't work.

Normally I would call the functions like this

PHP: Select all

<?=paraSummary($latestblogs)?>

where $latestblog is taken from my db.

Any help would be great!

[karinne]

Just a little bump

I really need help with this

[grahame]

I tried

Code: Select all

<?php

$latestblogs = <<<HERE
<p>The price of cheese</p>
<p>The Prince of Cheshire</p>
<p>The Polish Child</p>
HERE;

function paraSummary($subject) {
$matches = array();
preg_match("/(<p>[^<]+<\/p>)/", $subject, $matches);
return $matches[1];
}
?>

And the Winner is ... <?=paraSummary($latestblogs)?>

And it printed out

Code: Select all

And the Winner is ... <p>The price of cheese</p>

which is what I would have expected - the first and not the second paragraph. Have you checked your data? Are you sure that your first paragraph is the same as the second one - or does it have a div= class= or id= attribute, or something else that makes it different and causes it to fail to match? Spurious spaces? Capital Ps?

[karinne]

Thank you so much Graham for looking into this.

Here's what I have in my db ...

Code: Select all

<p>Many people tend to forget that the body \"tag\" is also an element and can be used to style your site. In <a href=\"http://www.karinne.net/archive/?bid=5\">How to center your site horizontally</a> I showed you how to take a wrap your site in that div to center it. Now ... I\'m going to use the body element to do the same. Of course, you use this option when your whole site is of a certain width.</p>

<p>Again, the code is pretty simple: margin: 0 auto; and a width: 780px (which should be changed to accommodate the width you have set-up for you site) but this time ... in the body.</p>


<pre>&lt;!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.0 Strict//EN\" 
\"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd\"&gt;
&lt;html xmlns=\"http://www.w3.org/1999/xhtml\" lang=\"en-CA\" xml:lang=\"EN-CA\"&gt;

&lt;head&gt;
    &lt;title&gt;Some title&lt;/title&gt;
    &lt;meta http-equiv=\"Content-Type\" content=\"text/html; charset=iso-8859-1\" /&gt;
    
    &lt;style type=\"text/css\"&gt;
    html {
        background-color: #fff;
    }
    
    body {
        color: #fff;
        margin: 0 auto;
        width: 780px;
        background-color: #900;
    }
    &lt;/style&gt;    
&lt;/head&gt;

&lt;body&gt;

this site is centered

&lt;/body&gt;
&lt;/html&gt;</pre>

<p><strong>Tested in:</strong> Firefox 2.0, Opera 9, IE6</p>

Then I do a $latestblogs = stripslashes($row_blog_for_category['blog']); to get rid of the slashes and then <?=paraSummary($latestblogs)?> .

So ... there's not extra stuff in there that I can see?

The first paragraph that should be spitting out is

Code: Select all

<p>Many people tend to forget that the body \"tag\" is also an element and can be used to style your site. In <a href=\"http://www.karinne.net/archive/?bid=5\">How to center your site horizontally</a> I showed you how to take a wrap your site in that div to center it. Now ... I\'m going to use the body element to do the same. Of course, you use this option when your whole site is of a certain width.</p>

but it's showing

Code: Select all

<p>Again, the code is pretty simple: margin: 0 auto; and a width: 780px (which should be changed to accommodate the width you have set-up for you site) but this time ... in the body.</p>

help

[grahame]

But there's *something* that's triggering a failure to match on the first line. Have you tried to print out $latestblogs and see what you have in there? Putting it through htmlspecialchars() will let you see all the tags.

[grahame]

GOT IT ....

It's the extra "href" tag within the first paragraph.

You're looking for <p> then characters which are NOT < then </p> .... but in the case of the first paragraph you have an extra <a href= .... in there, so it fails to match.

[karinne]

Ok ... so ... how can I make this work regardless of what's between the <p> ... </p>?

[grahame]

If you want to match *regardless* of intermediate tags, try

preg_match("/(<p>.*?<\/p>)/s", $subject, $matches);

(any number of any character, but as few as possible - a sparse match. And note the extra "s" after the second slash. Forces the "." to match against new line characters too in case the <p> and </p> are on different lines.

[karinne]

AH!!!! thank you thank you thank you thank you thank you!!!!!!!!!!!!!!!!!

*smooches*

[dannyboy26]

On a slight tangent, however if you want a generic solution that avoids issues with low-level changes to code and so on, you could do worse than learn to use Perl and the LWP. Tokenise the HTML with something like html tokeparser or even split it into a tree and you can develop some very robust solutions with minimal code to do all sorts of data extraction tasks.

Cheers

Dan

[karinne]

What?!?! Was that whole paragraph even written in English?! I didn't understand a word of what you just wrote in that post!

The code works ... and works just the way I want it.

[dannyboy26]

Lol

It was a bit of a tangent. If your script works fine then obviously no reason at all to change it - just added this for anyone else who is looking to do more intensive parsing of HTML at any stage and comes across this thread

Dan

[Ryan Fait]

That sounds very interesting. Any links to resources on the subject?