Following are the functions to extract img tag's src from html code.

I have taken care of almost all the scenarios I could think of, I hope it helps you

-Nitin

<?php

/*********************************************
 * BEGIN parser related functions
 *********************************************/

function getSrcPos(&$html, $pos)
{
  $posSrc = stripos($html,"src ", $pos);
  $posSrc2 = stripos($html,"src=", $pos);
  $posSrc3 = stripos($html,"src\t", $pos);
   
  if($posSrc === false || ($posSrc2 !== false && $posSrc > $posSrc2))
  {
    $posSrc = $posSrc2;
  }

  if($posSrc === false || ($posSrc3 !== false && $posSrc > $posSrc3))
  {
    $posSrc = $posSrc3;
  }
  return $posSrc;
}


function &extractString(&$html, &$pos, $char='')
{
  $posTemp = $pos;
  if($char == '')
  {
    while($html[$pos] != " " && $html[$pos] != "\t" && $html[$pos] != ">")
    {
      ++$pos;
    }
  }
  else
  {
    while($html[$pos] != $char && $html[$pos] != ">")
    {
      ++$pos;
    }
  }
  $str = substr($html, $posTemp, ($pos-$posTemp));
  $str = trim(str_replace(array("\n","\r"),"", $str));
  return $str;
}

function parseNextImage(&$html, &$offset, &$srcStr)
{
 
  $pos = stripos($html,"<img", $offset);

  if($pos !== false)
  {
    $posImg = stripos($html,">",$pos);
   
    $posSrc = getSrcPos($html, $pos);
   
    //echo " == $posImg $posSrc == ";
   
    if($posSrc !== false  && ($posImg === false || $posImg > $posSrc))
    {
      //we have position of src now, lets extract the string
      $i = $posSrc + 3;
      while($html[$i] == "=" || $html[$i] == " " || $html[$i] == "\t" || $html[$i] == "\r" || $html[$i] == "\n")   
      {
        ++$i;
      }
     
      //echo "**".ord($html[$i])."**".substr($html, $i)."***\n";
     
      if($html[$i] == "'" || $html[$i] == '"')
      {
        $j = $i+1;
        $srcStr = extractString($html, $j, $html[$i]);
      }
      else
      {
        $j = $i;
        $srcStr = extractString($html, $j);
      }
      $offset = $j;
      return 1;    
    }
    else if($posImg < $posSrc)
    {
      $offset = $posImg;
      return -1;
    }
    else
    {
      return false;
    }
  }
  else
  {
    return false;
  }
}


/*********************************************
 * END parser related functions
 *********************************************/
 


//this function demonstrates how to use the parse function
function testImageExtraction()
{
  $html = "<img src=abc.com /> <img class=hello src=\"abc1.com\" /> <img class=\"mno\" src='abc2.com ' /> <img src= 'abc3.com' />
          <img src= abc4.com /> <img src = \"abc5.com\" /> <img src   = ' abc6.com ' /> <img src = abc7.com\\n/new/img.jpg />
<img src =
\"
abc8.
com\"/>
          ";

  /*****
   * Always call the function parseNextImage() in this way
   * first make two variables $offset = 0 and $str = ""
   * and pass them along with $html
   *****/ 
  $offset = 0;
  $str = "";

  while(parseNextImage($html, $offset, $str))
  {
    // $str contains the src value extracted, use it the way you want
    echo "### $str ###\n";
  }

}

testImageExtraction();

?>

Rate this blog

   Report Abuse