sometimes this works other times no

[modtheplanet]

if($_POST["url"]!=""){
$k=1;
$hdl=fopen($_POST["url"], "rb");
do {
    $data = fread($hdl, 100);
    if (strlen($data) == 0) {
        break;
    }
    $cont .= $data;
} while(true);
fclose ($hdl);
$pattern = '/src\\=[\\\'\\"]?([^\\\'\\"]*)[\\"\\\']?/i';
$img = array();
preg_match_all($pattern, $cont, $img);
$realurl=parse_url($_POST["url"]);
$realurl=parse_url($_POST["url"]);
for($i=0; $i<count($img[1]); $i++){
if(ereg(".jpg", $img[1][$i]) || ereg(".jpeg", $img[1][$i])){
if(!ereg("http://", $img[1][$i])){
        $arr=explode("/", $_POST["url"]);
        array_pop($arr);
        $img[1][$i] = implode("/", $arr)."/".$img[1][$i];
}
$filename=basename($img[1][$i]);
$id = ImageCreateFromJPEG($img[1][$i]);
$prefix=uniqid($tom);
$filename=$prefix.$filename;
@ImageJPEG($id,'../xttempthumbs/'.$filename);
$thumbsti='../xttempthumbs/'.$filename;
$sizes = getimagesize($thumbsti);
$sizex = $sizes[0];
$sizey = $sizes[1];
if($sizex>$sizey){
$dif=$sizex/$sizey;
}
if($sizey>$sizex){
$dif=$sizey/$sizex;
}
if($dif<2.5){
        if($k==1){
                echo '<tr><td width="25%">Add: <input class="check" type="checkbox" name="pic" value="'.$thumbsti.'">
                <br><img src="'.$thumbsti.'"></td>';
                $k++;
        }elseif($k<4 && $k>1){
                echo '<td width="25%">Add: <input class="check" type="checkbox" name="pic" value="'.$thumbsti.'">
                <br><img src="'.$thumbsti.'"></td>';
                $k++;
        }elseif($k==4){
                echo '<td width="25%">Add: <input class="check" type="checkbox" name="pic" value="'.$thumbsti.'">
                <br><img src="'.$thumbsti.'"></td></tr>';
                $k=1;
        }
}
}
}
echo '</table></form>';
exit;
}

ok what this script does is take a url and looks at the html source code on the page, then filters out the jpgs to save.

For some jpgs this works great and for some i get an error of

Warning: imagecreatefromjpeg(): 'http://www.somesite.com/pic001.jpg' is not a valid JPEG file

I dont know what is going on, I tried this on various sites, some that worked and some that didnt. As far as i could can tell all the jpegs are RGB. The server that I am running the script on has GB 2.0 or higher.

I can give you two examples of sites that I tried "BEWARE THESE SITES HAVE ADULT PICTURES ON THEM!"

works on this site:
http://galleries.googirls.com/gals/?s=Nasty_Swapping_Bitches&rid=pp11072

dosen't work on this site:
http://www.wannawatch.com/hosted/index.php?ws/gross312/pumpthatass36

[Ceril]

Umm... without looking at your code to much but just the question:

If you filter out the jpegs as I assume which is you get the src for the jpeg file... why not just save it to your server without passing it along to GD first and then saving it?

If you want your page to display:
"This was the jpegs found in the last page you probed" then save all the jpegs you find to a /tmpjpegs directory and have your script build the display from what it finds in the tmpjpegs directory, check the ones you want to save(or discard) and hit continue upon which all the images marked for save is moved to /savedjpegs and the rest is simply removed from /tmpjpegs which now should again be an empty directory...

But I might miss exactly what your intent with using GD for each picture is.

[modtheplanet]

but it is given me a imagecreatefromjpeg() error right now, and you need that function to save it to a folder right?

[Ceril]

No, Imagecreatefromjpeg you do not need to save it to your own discspace. Just fopen the file and write it down.

Imagecreatefromjpeg you would want to use if you would like to modify the image, know the number of colors in it and such.

Think about the jpeg image as a file not an image you want to resize or know things about you just want to save it. You dont need imagecreatefromjpeg to save a file to your own disc, right? (Stating the obvious I know)