Patterns

[DeadFly]

I'm not very good with the pattern stuff, so I'm hoping someon can help me.

I'm looking for a pattern that will search an html document, and find all the "<img src=*.jpg>".
I know how to do a pattern to validate an email address, and to valid a URL, but this one im looking for is a bit trickeier than I though.

Side note:
This is used to search a gallery page with about 16 thumbnails, among other graphics. I want it to pic out the <img src> tags with the series of thumbnails only, not the rest of the images on the page, even if they are jpg format too.

[DeadFly]

What I have for the patter right now is :

$pat="^<(img)?([^[:space:]]+)?(src)?([^[:space:]]+)([[:alnum:]\.,-_?/&=])?([^[:space:]]+)?(.jpg)>$";

[DeadFly]

This is what I have so far, and all it does it grab the html file and then repeated print the whole html file to screen.

if ($f=@fopen($url,"r")) {
		echo "<select name=ImgFound>";			
		$Content=fread($f,10*1024);
		while (!end($Content)) {
			$pat="(img)?([^[:space:]]+)?(src)?([^[:space:]]+)([[:alnum:]\.,-_?/&=])?([^[:space:]]+)?(.jpg)";
			$rep = "\\1\\2\\3\\4\\5\\6\\7";
			$SelTxt=eregi_replace($pat,$rep,$Content);
			echo "<option>$SelTxt</option>"; 
		}   

} else {
    	echo "<font color=FF0000>Error Counting Images on Specified URL</font> &laquo;<br>";
}

[douceur]

I'm not entirely sure of what you want, but I think this might do the trick:

$Content = preg_replace('/<img src="([^"]*)\.jpg">/i', '$1', $Content);

To me, that says replace any text that starts with '<img src="', is followed by any number of any character except a " (because that would end the src attribute), followed by '.jpg">', and replace it with just the name of the file (minus the extension).

[DeadFly]

I think that got me started on the right track . . . but I'm not quite there.

if ($f=@fopen($url,"r")) {
    echo "<select>";
    $Content=fread($f,100*1024);
    while (!feof($f)) {		
        $SelTxt = preg_replace('/<img src="([^"]*).jpg">/i', '$1', $Content);
        echo "<option>$SelTxt</option>";		
    }
    echo "</select>";
}

Basically I want it to find all of the images on the page, and then print their info into a list box, for viewing.

[douceur]

By 'their info', do you simply mean the filename? If so, this should work:

$SelTxt = preg_replace('/<img src="([^"]*\\.jpg)">/i', '$1', $Content);
echo "<option>$SelTxt</option>";

Hope it helps!

[DeadFly]

Its not filling the list box . . .
Is there another error somewhere that anyone can see ?

Is while (!feof($f)) correct ?
I want "while not the end of the file". Basically to read through the whole file.

[douceur]

if ($f = fopen($url,'r')) {
  echo "<select>";
  $Content = fread($f, filesize($url));
  preg_match('/<img src="([^"]*\\.jpg)"[^>]*>/i', $Content, $imgs);
  foreach ($imgs as $img) {
    echo "<option>$img</option";
  }
  echo "</select>";
}

That should work as you want. The extra [^>]* in there is in case thare are extra attributes within the tag... Things such as height, width, alt, etc. You could even throw that right in front of the 'src=' part if you wanted to be safe.

[DeadFly]

Well it looks like it should work, but its not filling the select case at all. Only filling it full of empty spots.
This is really f*cked up ! LOL ! I'm sooo going insane. I've been trying to figure this damn code out for almost 6 hours !

BTW, thanks for all the help.
Gimme More ! LOL
🙂

[DeadFly]

if ($f = fopen($url,'r')) {
		echo "<select>";
		echo "<option>hello</option>";
		$Content = fread($f, filesize($url));
		echo "<option>hello</option>";
		preg_match('/<img src="([^"]*\.jpg)"[^>]*>/i', $Content, $imgs);
		echo "<option>hello</option>";
		foreach ($imgs as $img) {
			echo "<option>$img</option>";
		}
		echo "<option>hello</option>";
		echo "</select>";
	}

It will fill the list box with all the hellos, but it wont add the $img parts.

[douceur]

if ($f = fopen($url, 'r')) {
	echo "<select>";
	$Content = fread($f, filesize($url));
	preg_match_all('/<img src="([^"]*\\.jpg)"[^>]*>/i', $Content, $imgs);
	foreach ($imgs[1] as $test => $img) {
		echo "<option>$img</option>";
	}
	echo "</select>";
}

Stupid me... I should have been using the preg_match_all function instead of the preg_match function. preg_match() stops after it finds the first match..... Anyway, this has been tested on my server and works for me.

[DeadFly]

sigh
It wont work for me.

The only time it prints anything is when I remove the [1] and $test=> from : $imgs[1] as $test => $img
And what ti prints then is "Array".

I guess I'm too n00b.

[DeadFly]

Still no success.
Here is the code with all the variables manually placed into the statement:

$url="http://www.eatmygirlfriend.com/2003/May/08/index02.html";
	if ($f = fopen($url, 'r')) {
	    echo "<select>";
	    $Content = fread($f, filesize($url));
	    preg_match_all('/<img src="([^"]*\.jpg)"[^>]*>/i', $Content, $imgs);
	    foreach ($imgs as $test => $img) {
	        echo "<option>$img</option>";
	    }
	    echo "</select><br>";
	}

All it does is fill the list box with two lines of "Array".

[douceur]

To make it work for that page, here is the code I use:

$url = 'http://www.eatmygirlfriend.com/2003/May/08/index02.html';
if ($f = fopen($url, 'r')) {
	echo "<select>";
	$Content = fread($f, 10 * 1024);
	preg_match_all('/<img src="([^"]*\\.jpg)"[^>]*>/i', $Content, $imgs);
	foreach ($imgs[1] as $img) {
		echo "<option>$img</option>";
	}
	echo "</select>";
}

The only thing I'm not sure about is the 10 * 1024. I used that because that is what you used earlier, and filesize() does not work on remote files. Again, this has been tested on my server, and works fine.

[DeadFly]

HOLY SH*T IT WORKS !

Dude you're awesome !
Thanks for all the help !
😃 😃 😃

[DeadFly]

Now . . .
how can I make it recognize all the similar files ?
For that page it was everything in the "images/*.jpg".
I would like it to only count them . . .

Is that a whole new ball game ?

[douceur]

Give me a more specific example of what you need, and I'll see what I can do. 🙂

[DeadFly]

okay !

Well if you run that script against the URL I was using, youll see that it brings up "../../Recips/thumbnailseries.jpg" and then 16 of "images/#.jpg" where # is the number 1 thru 16.

I want it to just pick out the pattern of the "images/#.jpg" and not any other "*.jpg" files, but I don't what it to be specific only to this page. Forexample if it we to go to the next page and it had all the files in "pics/#.jpg", then it could still pick the pattern up.

[DeadFly]

There will always be alteast 12-20 jpgs to compare the patter against, but they won't always be named "/#.jpg", they could potentially be named "/a1.jpg", "/b2.jpg", etc.

Is this a huge task ?

[douceur]

Hmmm.... Well, if you know that the pages will always be in either 'images/' or 'pics/', you could do this:

if ($f = fopen($url, 'r')) {
    echo "<select>";
    $Content = fread($f, 10 * 1024);
    preg_match_all('#<img src="((pics|images)[^"]*\\.jpg)"[^>]*>#i', $Content, $imgs);
    foreach ($imgs[1] as $img) {
        echo "<option>$img</option>";
    }
    echo "</select>";
}

For every directory that the images might be in, just add it to the '(pics|images)' part with a | in the middle.