[SOLVED] simple string parsing

[marklarah]

So I have the HTML for a remote webpage stored in a variable (gotten through cURL). It's htmlentities'd, so I have the text for it. I need to get a specific link from this webpage. I think there are other links, but there is one in a specific format.

 

The format is

<a href="http://linkhere" style="font-size:15px;">

I have read the things online how to parse it, but I can't seem to find a specific one which will do this.  How would I go about doing it?

 

Thanks...Mark.

[premiso]

<?php
$string = '<a href="http://linkhere" style="font-size:15px;">';

preg_match('~<a href="(.+?)" style="font-size:15px;">~si', $string, $matches);

echo $matches[1];
?>

[marklarah]

Oh ok, thats a great way of doing it thanks.

 

Only thing is, it's not echoing anything, which it should. The string has entities on, if that makes a difference.

[premiso]

Then give me the exact string you are trying to get out, or modify it to fit your needs. The above should give you a good idea how to modify it.

 

If it is not echoing anything, it is not finding any matches. Simple as that.

[marklarah]

Ok, sort of half-fixed it. Turns out the entities does matter, i'm using

preg_match('~<a href="(.+?)" style="font-size:15px;">~si', $result, $matches);

 

Now, I get a lot of stuff, but the last thing is the link. the first the characters of what's on the page is

?">

- I presume this is the preg_match. If this needs editing then, then what comes before the link is

id="link">

- there is a space between the end of that div and the link/

[premiso]

Why not parse it before you run it through htmlentities ??? That would make this about 10 times easier on you.

 

Or if it is already entitied then do a html_entity_decode on it before running it through the preg_match function.

[marklarah]

Well okay, it's off, but I'm still getting lots of stuff before the link.

[marklarah]

Okay, we have this on the page:

<a href="?">

 

So it echoing from

?">

to the end of the link.

[premiso]

but I'm still getting lots of stuff before the link.

 

Hmm, that gives me a ton of information. Post your current code, and the "stuff" before the link.

 

Given that you take the original code I posted above it works fine right? So the issue lies with how the data is stored in the string. Do a print_r on the $matches variable, view the source and paste that array here as well.

 

Just to make sure you are echoing $matches[1] and not $matches[0].

[marklarah]

print_r returns exactly the same. The problem is, there are a few links on the page, so it goes from the first link with <a href="..... to our link we're trying to single out, because it is unique, with the style.

 

Can we edit the preg_match to identify a line of text before the link?

[marklarah]

nevermind, done it, there was a tab before the link, so I've put that into the preg and that seems to work. Thanks a million though for your help! <3 you...

[samshel]

try this..

 

<?php
$string = '<a href="http://linkhere" style="font-size:20px;"<a href="http://linkhere" style="font-size:15px;">';

preg_match('~<a href="([^\"]+)" style="font-size:15px;">~si', $string, $matches);

echo $matches[1];
?>