Geekality » Tutorial

PHP: Generating transparent PNG fillers

Torleif — Sun, 13 Jun 2010 13:45:38 +0000

So, I was fooling around the other day with an HTML table and wanted to make the odd rows slightly darker. Figured I could use for example an 80% transparent black PNG to do that (or could have just assigned a darker color, but where’s the fun in that?). Either way, ended up making a little something that I thought I could share just for the fun of it. I learned a few things, so maybe someone else can too

So, here’s how to make a small transparent PNG filler image on-the-fly using PHP

I called the script fill.php, and the first line looks like this:

header('Content-type: image/png');

This starts the php script and sets the content-type to image/png which means the web browser will that we are an image and not an html file which usually is the default.

Next up we get the parameters from the user:

$p = $_GET + array
(
'r' => 0,
'g' => 0,
'b' => 0,
'a' => 0,
);

Here I create an array with default values and merge it with the $_GET array so that values from the user will override the default values. This means we are sure we have all the values we need and that the user only have to specify the ones he want to.

The r, g and b values should be a number between 0 and 255 and the alpha value should be between 0 and 127. So, next up we need to make sure that that is the case so we don’t use some other bogus we might have gotten:

array_walk($p, function (&$item, $key)
{
$item = min(max(0, $item), $key == 'a' ? 127 : 255);
});

This goes through each parameter and makes sure it is a number between 0 and 255 (or 127 for the alpha).

Since we have what we need now, we can start with creating our image:

$img = @imagecreatetruecolor(15, 15)
or die('Cannot Initialize new GD image stream');

We also need to set a flag that we want to use full alpha channel information (as opposed to single-color transparency) in this image:

imagesavealpha($img, true);

We are now ready to fill our image with the color we are asked for:

$color = imagecolorallocatealpha($img,
$p['r'],
$p['g'],
$p['b'],
$p['a']);
imagefill($img, 0, 0, $color);

And finally output the result and destroy the image resource we have been working with:

imagepng($img, null, 9);
imagedestroy($img);

And that’s actually all there is to it!

The complete code looks like this:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39

header('Content-type: image/png');

// Replace default parameters with user input
$p = $_GET + array
(
'r' => 0,
'g' => 0,
'b' => 0,
'a' => 0,
);

// Make sure they are within limits
// (colors: 0-255, alpha: 0-127)
array_walk($p, function (&$val, $key)
{
$val = min(max(0, $val), $key == 'a' ? 127 : 255);
});

// Create image
$img = @imagecreatetruecolor(15, 15)
or die('Cannot Initialize new GD image stream');

// Use full alpha channel information
imagesavealpha($img, true);

// Fill image
$color = imagecolorallocatealpha($img,
$p['r'],
$p['g'],
$p['b'],
$p['a']);
imagefill($img, 0, 0, $color);

// Output result
imagepng($img, null, 9);
imagedestroy($img);

If you want to check out a working demo version of this, you can test out my little testing tool I created while having fun with this at tools.geekality.net/filling-factory

Let me know what you think! And as usual, please let me know if you see something that is wrong or could be improved or anything like that

How to reset the MySql root account

Torleif — Sat, 27 Dec 2008 16:38:24 +0000

I just managed to mess up the MySQL root account. Not a smart idea. After some MySQL manual reading and some serious Google-Fu, I figured out how to fix it.

Stop the mysql server
Add the following to the my.cnf file
[mysqld]
skip-grant-tables
Start the mysql server
Start a mysql client and run the following query
REPLACE INTO mysql.user VALUES ('localhost','root',PASSWORD('blah'),'Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','Y','','','','',0,0,0,0);
Quit the mysql client
Stop the mysql server
Remove what we added to my.cnf in step 2
Restart the mysql server

Tadaa.

How to use assembly embedded resources

Torleif — Sat, 27 Dec 2008 16:13:16 +0000

After some digging around, I found that it actually wasn’t that difficult at all.

Getting it in

Putting something in an assembly as an embedded resource is pretty easy. At least if you are using Visual Studio. Just add the file to your project, click on it, and then under Properties set Build Action to Embedded Resource. And thats it!

Getting it out

Lets say we want an image called hello.png in a folder called Wopdidoo as a Stream.

If we are executing code in the same assembly, we can do as follows:

Assembly assembly = Assembly.GetExecutingAssembly();
Stream imageStream = assembly
.GetManifestResourceStream("DefaultNameSpaceOfAssembly.Wopdidoo.hello.png");

If you are not executing code in the same assembly you just have to get that assembly reference in a different way. I often use Assembly.GetAssembly(typeof(T)), where T is some type you know exists in the same assembly as the file you want. The rest is the same.

As far as I know, you use the stream as any other stream. Not sure if it is writable though? Probably not… let me know if you have some brilliant clues on that matter

Remember to Dispose it when you are done.

Finding it

I sometimes find it a bit difficult to figure out that string which identifies the resource. I then often use the following code to “find” it:

foreach (string s in assembly.GetManifestResourceNames())
System.Diagnostics.Debug.WriteLine(s);

It basically just scans through all the resource names and prints them out to the debug console