Trying to find tag and get attributes using regex in php

Question

I have made a regex in php to find the tag and the attributes in an html line. It works but only works on the first attribute instead of repeating. The follow code gets me the first attribute and value.

'@<barcode(\s([a-z]+)="([^"]+)").*/>@m'

So then I added the plus to make it repeat, but it won't work.

'@<barcode(\s([a-z]+)="([^"]+)")+.*/>@m'

What happens after adding the plus is that it only selects the last attribute and value.

I just need all the attributes and values in an array, so I am wondering what I am doing wrong. Here is the possible html that I am searching through. Sometimes attribute are not always needed so I have to take that into account.

<barcode type="C128B" height="10" fontsize="0.4" code="testcode" align="L"/>
<barcode type="Hello"/>
<barcode type="Hello" code="balls"/>
<barcode type="C128B" height="10" fontsize="0.7" code="test" align="L"/>

I have an example on regex101 to see the problem https://regex101.com/r/jMdA6S/1

Our current application works, but only by repeating the following lines

'@<barcode ([a-z]+)="(.*)" ([a-z]+)="(.*)" ([a-z]+)="(.*)" ([a-z]+)="(.*)" ([a-z]+)="(.*)".*/>@m'

Which means everytime I add a new attribute I have to add another block of code in the regex. I am trying to avoid this as we sometimes have to add a new attribute to add different features.

Answer 1

You need to put /g at the end of your regex like so:

<barcode(\s([a-z]+)="([^"]+)").*/g>

Answer 2

A good practice is to parse HTML content with a relevant manipulation tool. For your question, you can parse while reading the file (SAX approach), or load the file at one time and then accessing its content (DOM approach).

Here is a way to perform what you need. I like using the SAX way if I don't need to keep the whole content (widely based on XML Element Structure Example of PHP official website):

<?php
$file = "data.html"; // your file
$depth = array();

function startElement($parser, $tagname, $attrs)
{
    // For each tag encountered
    //   - $tagname contains the name
    //   - $attrs is an associative array name -> value of the attributes

    // Add the code below the code to deal with it:
    echo "<pre>\n";
    echo "Tags : $tagname\n";
    echo "Attributes:\n";
    print_r($attrs);
    echo "</pre>\n";
}

// Create the parser
$xml_parser = xml_parser_create();

// Set element handles for the parser (we just need start element handler, 
// so the end element is set as FALSE
xml_set_element_handler($xml_parser, "startElement", FALSE);

// Open your file
if (!($fp = fopen($file, "r"))) {
    die("Oops.");
}

// Loop reading and parsing the file
while ($data = fread($fp, 4096)) {
    if (!xml_parse($xml_parser, $data, feof($fp))) {
        die("Oops.");
    }
}

// Done. Free your parser.
xml_parser_free($xml_parser);
?>

Answer 3

Well even though there were some good answers nobody was able to tell me if there was a way to do this in one regex which is what my question was. However I have had to succumb and do it in two regex's.I was trying to avoid 2 regexs, as I thought the plus was supposed to repeat the middle part.

The first regex finds the tags, and I have a getAttributes function which gets the attributes. The getAttributes function then puts each into a flat array for me to process. I am giving an answer, but even this answer does not really answer my question on how to do this in one regex. However I will post what I got working in case it helps anyone else.

Both Amessihel and Maciej Król gave good advice, and I would probably take that advice if this was a new project being built. However I have gone with the following code.

<?php
$str = '<barcode type="C128B" height="10" fontsize="0.4" code="pdfbarcode_content" align="L"/>
<barcode href="Hello"/>
<barcode href="Hello" type="balls"/>
<barcode type="C128B" height="10" fontsize="0.4"/>
<barcode type="C128B" height="10" fontsize="0.4" code="test" align="L"/>';

function getAttributes($attr){  
    preg_match_all('@(?:([a-z]+)="([^"]+)")+@m', $attr, $matches,PREG_SET_ORDER);
    $rArray=[];
    foreach($matches as $line):
        array_push($rArray,$line[1]);
        array_push($rArray,$line[2]);
    endforeach;
    return $rArray;
}
function barcode($file){
    return preg_replace_callback(
        '@<barcode(.*)/>@m',
        function($matches) {
            echo '<pre>'.print_r($matches[1],1).'</pre>';
            echo '<pre>'.print_r(getAttributes($matches[1]),1).'</pre>';
            echo "-----------------------";
            //Here is where I process the array
            return '';
    },
    $file);
}
barcode($str);

Answer 4

You probably need to write parser for this if u want to match unlimited amount of xml objects and access key values pairs from them (using regex).

I've prepared for you working example.

   $offset = 0;

   $lines = '
       <barcode type="C128B" height="10" fontsize="0.4" code="testcode" align="L"/>
       <barcode type="Hello"/>
       <barcode type="Hello" code="balls"/>
       <barcode type="C128B" height="10" fontsize="0.7" code="test" align="L"/>
   ';

   while (preg_match('/<(\S*)[\s]*(.*)[\s]*\/>/', $lines, $line_matches, PREG_OFFSET_CAPTURE, $offset))
   {
       // Set offset to the next line
       $offset = $line_matches[0][1] + strlen($line_matches[0][0]);

       // Get the line name
       $name = $line_matches[1][0];

       // Get the line content
       $line_content = $line_matches[2][0];

       if(preg_match_all('/([a-z]+)="([^"]+)"/', $line_content, $key_values_matches))
       {
           // Access all matched keys
           $keys = $key_values_matches[1];

           // Access all matches values
           $values = $key_values_matches[2];

           foreach ($keys as $index => $key) {
               // Access matched value for key
               $value = $values[$index];

               // Do something with your match
               echo "Found match in \"{$name}\" for key \"{$key}\" with value \"{$value}\"\n";
           }
       }

   };