Question

I’m coding a scraper like the Facebook one to get info from a given url. Im just one step to finish the basic functionality. The problem so far is to remove useless images.
For example, when i run some random url, i get all this images:

Scraper Object
(
    [url] => http://buzz.money.cnn.com/2012/07/23/spain-italy-short-selling/?iid=HP_Highlight
    [title] => Spain and Italy ban short selling  - The Buzz  - Investment and Stock Market News
    [description] => The Euronext 100 stock index falls sharply on renewed concerns about Spain.    Securities regulators in Spain and Italy both instituted short-selling bans Monday as financial markets tumbled.    The move is designed to limit the downward pressure
    [imageUrls] => Array
        (
            [0] => http://cnnmoneybuzzblog.files.wordpress.com/2012/07/chart_ws_index_euronext100_201272310932-09.png
            [1] => http://i.cdn.turner.com/money/.e1m/img/5.0/data/feargreed/scale.316x95.png
            [2] => http://i2.cdn.turner.com/money/.element/img/5.0/sections/contributors/ben_rooney_130.jpg
            [3] => http://i2.cdn.turner.com/money/.element/img/5.0/sections/contributors/catherine_tymkiw.02.jpg
            [4] => http://i2.cdn.turner.com/money/.element/img/5.0/sections/contributors/paul_lamonica.02.jpg
            [5] => http://i2.cdn.turner.com/money/.element/img/5.0/sections/contributors/hibah_yousuf.02.jpg
            [6] => http://i2.cdn.turner.com/money/.element/img/5.0/sections/contributors/maureen_farrell.02.jpg
            [7] => http://i2.cdn.turner.com/money/.element/img/5.0/sections/contributors/ben_rooney.02.jpg
            [8] => http://i.cdn.turner.com/money/.element/img/4.0/services/button_login.gif
            [9] => http://www.bizographics.com/collect/?fmt=gif&pid=311
            [10] => http://pixel.quantserve.com/pixel/p-5dyPa639IrgIw.gif
            [11] => http://i.cdn.turner.com/money/.element/img/1.0/misc/1.gif
            [12] => http://buzz.money.cnn.com/2012/07/23/spain-italy-short-selling/?iid=HP_Highlight//pixel.quantserve.com/pixel/p-18-mFEk4J448M.gif?labels=%2Clanguage.en%2Ctype.wpcom%2Cposttag.bonds%2Cposttag.dow%2Cposttag.ibex%2Cposttag.italy%2Cposttag.lehman%2Cposttag.milan%2Cposttag.nasdaq%2Cposttag.sp-500%2Cposttag.short-selling%2Cposttag.spain%2Cposttag.stock%2Cposttag.yields%2Cvip.cnnmoneybuzzblog
            [13] => http://stats.wordpress.com/b.gif?v=noscript
        )

)

I just need to find a way to remove all those images ending in .gif or .png and just let the .jpg inside the array, so the user can take a look and select proper one for the article.

I’ve tried some array functions but i think this need some regex magic for it to work in almost any url given.

P.S. I can access all array data using $info->url, $info->description and so. Just need to filter that array and it will be ready.

Answer 1

Without using regex:

foreach ($objects_array as $obj){
  foreach($obj->imageUrls as $key => $img){
    if(substr($img, -4) === '.gif' || substr($img, -4) === '.png'){
      unset($key);
    }
  }
}

and using regex:

$pt = '/gif$|png$/';
foreach ($objects_array as $obj){
  foreach($obj->imageUrls as $key => $img){
    $res = preg_match($pt, $img);
    if($res){
      unset($key);
    }
  }
}