Best way to get certain information from a website using PHP

Question

I want to get a certain information of a website. The problem I'm facing is that this certain information is changing maybe a few times a day. This is because the content is dynamic.

The goal of my PHP script is to get the content (dynamic content from a database) in a PHP variable.

I've set up a codepen to show you what I mean: https://codepen.io/anon/pen/XEVpBo
The HTML from the codepen:

<div class="wrapper">
  <div class="some_useless_div">
    <p>Some useless text paragraph.</p>
    <div id="another_useless_div">
      <p>The actual important part is: SOME_DYNAMIC_TEXT what I want to put into a variable. The text around that dynamic text is static text and will not change.</p>
    </div>
  </div>
</div>

Currently, what I do to capture the information is to explode around the dynamic information:

$content = file_get_contents('https://codepen.io/anon/pen/XEVpBo');
$parts = explode('The actual important part is: ', $content); // some text that is left of the information.
$parts2 = explode(' what I want to put into a variable.', $parts[1]); // some text that is right of the information.
$information = $parts2[0]; // AHA! Now we have the information!

However, this really feels like spaghetti code. Isn't there a function that maybe searches for a string and returns that value such as:
$information = search_string('The actual important part is: %s what I want to put into a variable.'); where %s would be the information put into the $information variable.

Again, the code I use (above) works but it really feels like bad code. I'm looking for a clean function of PHP.

you could also look into using [`DOMDocument`](https://stackoverflow.com/questions/2571232/parse-html-with-phps-html-domdocument) — Kevin, Mar 28 '18 at 01:40
does the site have an API? are your breaking any terms scraping it? — , Mar 28 '18 at 01:53

hanshenrik · Accepted Answer · 2018-03-28T10:06:42.547

maybe you're looking for preg_match ?

test code seems to work fine: https://3v4l.org/6YeSh ,

<?php
$html=<<<'HTML'
<div class="wrapper">
  <div class="some_useless_div">
    <p>Some useless text paragraph.</p>
    <div id="another_useless_div">
      <p>The actual important part is: SOME_DYNAMIC_TEXT what I want to put into a variable. The text around that dynamic text is static text and will not change.</p>
    </div>
  </div>
</div>
HTML;
preg_match('/The actual important part is\: (.*?) what I want to put into a variable\./',$html,$matches);
$str=$matches[1];
var_dump($str);

also, when you're talking about the "best" way, it's definitely not file_get_contents, for at least 2 reasons:

file_get_contents keep reading until the socket is closed by the target server, but should stop reading once content-length bytes has been read, which, depending on the server, might have executed much faster

file_get_contents does not support compressed transfers.

curl reads until content-length bytes have been read, then returns, it also supports compressed transfers, thus curl should run significantly faster than file_get_contents.

(and i disagree, your code is not spaghetti code. i don't think it's good code, because you should have been using preg_match instead of explode(), it's probably faster, use less memory, and easier to write and maintain than your explode code, but your explode code is not spaghetti.)

Best way to get certain information from a website using PHP

1 Answers1