Other WYSIWYG editors I discovered and tested during my hunt where the jQuery based jwysiwyg project and the more complex wymeditor. While these were too complex for my needs they do work quite well and I would recommend you give them a go before making your mind up.

The one problem I do I have with TinyEditor is the unfortunate lack of a paste from word feature, I am hoping that this does become included soon because it is the only additional feature that is a must with WYSIWYG HTML editors due to paste from word screwing your nicely formatted code up upon submit.

Up until I completed the following quick fix, wordpress was using almost 30MB per instance, this means that I could only at peak have 66 or so visitors at the same time before my server became overrun and hid under the bed. What I needed to do was supercharge php and the easiest way to do that is by caching. Thanks to this interesting article on php caching by itst.net I chose APC (Alternative PHP Cache) which is supported by the php foundation and showed the most promising speed improvements.

Installing APC really couldn’t have been made any easier on debian; it does require some dependencies, obviously php 5.x plus the pear package installed but apart from that not much else is needed to be installed and it you use apt-get they are installed as part of the installation process. The instillation process is as simple as follows:

apt-get install php-apc


/etc/init.d/apache2 restart

Once APC has been installed and apache restarted you can issue the following command to ensure that it has installed correctly:

php -r 'phpinfo();' | grep apc

Once that has been done and you get a positive response to the above command you should see that wordpress is using considerably less memory. My install with its 16 or so plugins now uses 3.16MB rather than almost 30MB! That is what you call a good improvement.

One draw back with apc as with many caching processes is that it does cause an initial increased draw on the processor while it compiles the php into an performance enhanced executable file and stores it in a ram disk however once this intial process it done you can stand to see upto 200% increase in performance and a severely reduced memory footprint for each of your php apps.

I had been using the following code to do this:

if($post->post_parent=="11"{ ... }

While this simple switch works if you are one layer deep for example Fish (ID 11) > Tropical Fish (ID 12) it wont work if your more than two layers deep such as Fish (ID 11) > Tropical Fish (ID 12) > Amazon Fish (ID 13) because $post_post_parent for Amazon Fish becomes 12 not 11. In terms of programming this is all very logical, however the solution to finding out the original post_parent otherwise known as the grandparent or great grandparent is not very well documented and it was only luck that helped me stumble upon it. The very simple function get_post_ancestors() solves this problem by effectively returning an array with every post id related to the one in which it was called as the following example shows:

$ancestors = get_post_ancestors();

$ancestors now becomes an array which we can search through the php in_array command by the following code:

in_array(11,$ancestors);

In this example if one of the $post_id’s of the current posts ancestors is 11 then the in_array function will return with true, otherwise it will return false. This then makes showing information on all sub pages (child pages) of Fish(ID 11) a simple matter of using the following code:

$ancestors = get_post_ancestors($post);
if(in_array(11,$ancestors)){
	...
}

I use this on my projects page’s to provide a link back to the projects page from the sidebar of all its children, it could also be used to show a custom header for that section of your blog or certain links from a link category relevant to that section. The usefulness of this function is never ending and should be one of many you use when writing your own templates.

sudo apt-get install curl libcurl3 libcurl3-dev php5-curl

If you dont have sudo installed then simply enter it without then enter the following to reboot your apache server for the changes to take effect.

sudo /etc/init.d/apache2 restart

Again if you dont have sudo installed then remove it from the line before you press return and cURL will be fully working – enjoy.

Fatal error: Allowed memory size of 8388608 bytes exhausted (tried to allocate 79 bytes)

Upon doing the usual google, I discovered a few people have been having the same problem as myself. It is however a very easy problem to fix and although I found no actual answers on the net (obviously I didn’t look hard enough) I figured it out in the end. The problem is caused by scripts requiring more memory to execute than php has available to it. By default php limits itself to 8MB and usually that is enough with which to run most scripts, however either by some error of php or the new heavier than usual requirements of todays scripts sometimes 8MB is just not enough and as you can see in my example only 79 bytes extra where needed.

The fix is simple, find your php.ini file open it and find the Resource Limits section. Under that heading there is a line which by default says:

memory_limit = 8M ;Maximum amount of memory a script may consume (8MB)

All you need to do is change the 8M to 16M, I know that the problem is caused by scripts only needing a few extra bytes but its always nice to have a multiple of the original therefore I have doubled it, you needn’t do this changing it to 9M will fix the problem. Once you have made the changes save the file and exit then restart your server process which in my case is apache and the problem will be fixed and all your scripts will be working

Feel free to comment if you have any other questions about this topic.

function time_since($original) {

// array of time period chunks

$chunks = array(

array(60 * 60 * 24 * 365 , 'year' ),

array(60 * 60 * 24 * 30 , 'month' ),

array(60 * 60 * 24 * 7, 'week' ),

array(60 * 60 * 24 , 'day' ),

array(60 * 60 , 'hour' ),

array(60 , 'minute' ),

);$today = time(); /* Current unix time */

$since = $today - $original;

// $j saves performing the count function each time around the loop

for ($i = 0, $j = count($chunks); $i < $j; $i++) {

    $seconds = $chunks[$i][0];

$name = $chunks[$i][1];

    // finding the biggest chunk (if the chunk fits, break)

if (($count = floor($since / $seconds)) != 0) {

// DEBUG print "<!-- It's $name -->n";

break;

}

}

$print = ($count == 1) ? '1 '.$name : "$count {$name}s";

if ($i + 1 < $j) {

// now getting the second item

$seconds2 = $chunks[$i + 1][0];

$name2 = $chunks[$i + 1][1];

    // add second item if it's greater than 0

if (($count2 = floor(($since - ($seconds * $count)) / $seconds2)) != 0) {

$print .= ($count2 == 1) ? ', 1 '.$name2 : ", $count2 {$name2}s";

}

}

return $print;

}

I also found a second more compact version from byteinsider their version (below) takes the input in the form of the MySQL timestamp format (Y-m-d H:i:s) both functions output the same type of thing tho, so its up to you which you use.

function time_since($mysql_timestamp)

{

$names =  array('year','month','day','hour','minute','second');$r = time()-strtotime($mysql_timestamp) - date('Z');

$posted = array(date('Y',$r)-1970,date('n',$r)-1,date('d',$r)-1,

 date('G',$r)-0,date('i',$r)-0,date('s',$r)-0);

for($n = 0; $n < 5 && $posted[$n] == 0; $n++);

$output = $posted[$n].' '.$names[$n];

if($posted[$n] != 1) $output .='s';

if($n < 5 && $posted[$n+1] != 0)

{

 $output .= ' '.$posted[$n+1].' '.$names[$n+1];

 if($posted[$n+1] != 1) $output .='s';

}

return $output.' ago';

}