Revisited: Tcpdf – Variable Height Table Rows With MultiCell

A few weeks ago I wrote an article about creating variable height table rows using tcpdf. It was a neat solution and I liked it, which was the reason I blogged about it in the first place. However, it turns out that ’sharing the love’ is not the only reason to blog about these things. It can also be a learning experience thanks to the commenting from you out there in the interwebs.

The particular comment in question, from mike, made me realise that there was a bit of a flaw in my neat idea – page breaks. Because I was first drawing the text, the automatic page breaks played havoc with the positioning when it came to drawing the borders. If the text was so long that it wrapped across enough lines to cause a page break, when I tried to reposition to draw the borders I was in completely the wrong place and it all went a little Pete Tong.

Still, not deterred by this little setback, I re-looked at the problem and have come up with a different approach that solves this problem and adds the ability to do a few more things.

When I relooked at the documentation, I discovered the useful little function: getNumLines(). This actually allows us to determine how many lines a string of text will take up, given a particular width. In effect, it allows us to do what I was using MultiCell to return, without actually drawing anything. This lets us to determined the maximum cell height with one line of code:

$linecount = max($pdf->getNumLines($row['cell1data'], 80),$pdf->getNumLines($row['cell2data'], 80),$pdf->getNumLines($row['cell3data'], 80));

Now that we know the cell height before we have drawn anything, we can use MultiCell() to draw the text and the borders at the same time.

This was not the end, however. My particular implementation was a grid of data that had borders between the column and not between the rows. All very pretty until that automatic page break happened and gave us a grid of data without a border at the bottom. To add insult to injury, the new page ended up with a grid of data with no border at the top!

To solve this, I needed to know if the row was going to be the first of the page, so I could draw the bottom border on the previous row and give the new row a top border. With the use of the page dimensions I could work this out:

$dimensions = $pdf->getPageDimensions();
$hasBorder = false; //flag for fringe case
 
foreach($data as $row) {
	$rowcount = 0;
 
	//work out the number of lines required
	$rowcount = max($pdf->getNumLines($row['cell1data'], 80),$pdf->getNumLines($row['cell2data'], 80),$pdf->getNumLines($row['cell3data'], 80));
 
	$startY = $pdf->GetY();
 
	if (($startY + $rowcount * 6) + $dimensions['bm'] > ($dimensions['hk'])) {
		//this row will cause a page break, draw the bottom border on previous row and give this a top border
		//we could force a page break and rewrite grid headings here
		if ($hasborder) {
			$hasborder = false;
		} else {
			$pdf->Cell(240,0,'','T'); //draw bottom border on previous row
			$pdf->Ln();
		}
		$borders = 'LTR';
	} elseif ((ceil($startY) + $rowcount * 6) + $dimensions['bm'] == floor($dimensions['hk'])) {
		//fringe case where this cell will just reach the page break
		//draw the cell with a bottom border as we cannot draw it otherwise
		$borders = 'LRB';	
		$hasborder = true; //stops the attempt to draw the bottom border on the next row
	} else {
		//normal cell
		$borders = 'LR';
	}
 
	//now draw it
	$pdf->MultiCell(80,$rowcount * 6,$row['cell1data'],$borders,'L',0,0);
	$pdf->MultiCell(80,$rowcount * 6,$row['cell2data'],$borders,'L',0,0);
	$pdf->MultiCell(80,$rowcount * 6,$row['cell3data'],$borders,'L',0,0);
 
	$pdf->Ln();
}
 
$pdf->Cell(240,0,'','T');  //last bottom border

Note that there is a fringe case that I came across. This was when the row would not cause a page break as it was exactly the height of the remaining space. Any attempt to draw the bottom border with a separate call would cause a page break. I solved this by ensuring in that particular case, the cell was drawn with the bottom border already.

What gives this method more power (apart from actually working) is that we know when there is going to be a new page and we could choose to force a new page and if we wanted. We could then re-plot the grid headings before creating the row. That way we can have the grid headings on every page.

I am determined to get a water-tight solution to this problem and I think this gets us closer. If you do find any problems with it, let me know and I will see if we can solve them. Thanks for your feedback.

Update: Bretton Eveleigh has written a class that encapsulates this and other helper methods into a format that makes creating tables with Tcpdf much easier. See the comments below for the link. Thanks Bretton.

Cross Browser Testing – Its All Wrong

It was not long ago that I got all excited when I came across the XenoCode virtualized browsers. This seemed a neat idea. At the time one could download an executable (they seem to only be available in-browser now) and run one of the major browsers without interfering with the operating system or confguration you had.

I was all for throwing away my virtual machines there and then. However, I soon realised that they were as lacking as all the other cross browser testing systems out there. The most recent of which is a product from Microsoft themselves. Called SuperPreview, it is actually closely linked with the Expression product and it allows you to display the same page in different version of Internet Explorer. This is a similar approach to IETester, which allows you to open tabs containing the different rendering engines. This is fine if you just want to see that IE6 renders things in a completely different manner to everything else out there but it does not help you one bit if you are actually trying to solve those rendering problems.

Why do none of the IE cross browser rendering tools contain the Developers Toolbar? I mean, it is not a fraction as useful as the wonderful Firebug but it is the nearest we have to an IE tool that actually assists the developer in ironing out those css problems. I have even resorted to the IECollection which claims to contain all the versions of Internet Explorer that have ever existed and to contain the developers toolbar but only manages to throw up error messages and still no sign of a working developer’s toolbar.

So here it is, a request to the producers of the current cross-browser products or to anyone who fancies the challenge. Create a tool that is actually useful. Make it contain all current browsers and some developers tools that allow real-time changing of CSS and whilst your there, make it highlight the padding and margins in that oh-so useful way that Firebug does. That way we can actually see what is causing the problem and try a few fixes without having to change the css file and refresh the page every time. Also, do a favour for the Mac users out there and make it cross-platform.

If it really was a useful tool, developers would buy it and most be willing to pay a substantial license fee. I spend an awful lot of the time waiting for virtual machines to boot and suffering the performance consequences of running them whilst developing, so I would be happy to fork out some cash to solve this problem.

Accurate Ems

Whilst supporting browsers that have no full-page zoom and keeping the accessibility at a maximum, the use of ems in your style sheet is a must. This is not a difficult thing to do and you can still think in terms of pixels, thanks to great tools like the Em Calculator.

A word of warning however. If you don’t specify your font-sizes with enough accuracy, it can lead to some strange problems. I came across this when creating a multi column list. I used the method where each column is created using a left and a negative top margin. The problem was that because I only used two decimal places for the sizes, my line-heights did not translate a whole number of pixels.

This meant that for each item in the list, the fraction compounded and the total height of each column was not what what I thought – I ended up with a slanted set of columns. This may have not been an issue except each of the browsers interpret the fraction of pixels in a different manner – some round up and some round down.

It turns out that using three decimal places for all Em sizes will solve this and ensure that fractional pixel sizes are eliminated. This way you get no size differences across the different browsers and more predictable layouts. You can adjust the number of decimal places with the Em calculator by clicking on the settings tab and adjusting the value there.

The Importance of a Good jQuery Selector

As a technical lead and working in a company that has had rapid expansion in the last six months or so, I have had the job of being a mentor to some of the new employees. I don’t think any of them had used jQuery before and so for every person that arrived, I needed to give a crash course on the subject. The jQuery selector is always one of the first things I describe and it is usually met with nods of understanding, as it is not a difficult concept to grasp when most developers are familiar with css and the DOM.

There have been two incidents recently however, that have made me realise that my crash course on jQuery selectors has been missing something vital. The first was when I found some time to explore the options of profiling some of our javascript code. When I first started this, the only option I could see was John Resig’s deep profiling script that injected the statistics at the bottom of the page. This had some limitations however, and I was very happy to open my news feeds soon after and find that he had added a couple of extra methods to FireUnit that enabled you to profile javascript function calls.

What I noticed straight away was that there were a lot of seemingly simple calls in our code that took a long time to execute. The reason for this was that the selectors were vague:

    $('.myclass').somefunction();

In order to find all the matches, even if there is only one, jQuery needs to traverse the entire DOM looking for any element that might have the matching class, and that takes time. By narrowing down the selector using, for instance, the type, jQuery can very quickly eliminate all other types. This speeds things up a lot.

The second incident was due to another ‘benefit’ of being a tech lead. I tend to get passed the most peculiar and hard to diagnose bugs. The bug in question was on a page that contained two autocomplete text boxes, one of which was part of a compound control. The bug manifested itself by not being able to select an item from the second autocomplete text box properly. I was handed the bug with the information that it seemed to happen only when you selected from the first autocomplete control first. Needless to say, there was a lot of poking around in the code of both the autocomplete and the compound control, which is particularly complex given the nature of the control.

Thanks to firebug’s great console logging, I discovered that the variable that contained all the autocomplete list items seemed to be a combination of both controls. It was not too long before i discovered the code that caused the problem:

    listitems = $('.auto').children('li');

There was nothing mysterious going on at all, jQuery was asked to select all of list items in both controls, so it did.

What had happened in both of these incidents is that I had failed to mention the importance of an accurate selector in jQuery during my crash course. The new developers were blissfully unaware of the consequences of not specifying exactly which elements you want jQuery to select. What I had also failed to emphasise was another way of narrowing down and speeding up the jQuery selection – by use of the elements that you already have. In the case of the autocomplete, we already had a reference to the list itself, so it was simple:

   listitems = thelist.children('li');

This not only eliminated the cross-contamination of the two controls but seeded things up. So the rules to keep things fast and to save me headaches are simple:

• Use a selector that very accurately matches the elements you require
• Utilise references to elements you already have

Captcha Audio with ASP.NET

The subject of captcha images has been well covered. There are plenty of available resources for creating those warped letters. I created such a control that is a combination of quite a few different ideas and was quite proud of it until not thirty seconds after demonstrating it, someone asked whether it had a ‘play audio’ button for blind people. All of a sudden my fancy captcha control was not so fancy.

I tried to suggest using reCaptcha as a solution that had all the bells and whistles, but the customisation of my control trumped the somewhat fixed design of reCaptcha. So the problem of the audio captcha brewed a little in the back of my mind and a month or two later I turned back to it to see if I could find a solution.

A speech synthesis engine was not on the cards, so I figured that because the captcha image was a random collection of letters and numbers, the only way I could generate the appropriate audio was to have audio files of all the letters and numbers. I would then need to join them together, on demand, and play them from the web page.

Creating all the letters and numbers is straight forward enough (cue microphone and best-est speaking voice) and playing audio files from a web page has also become pretty easy thanks to the great SoundManager 2 javascript plugin.

The trickiest part was definitely joining mp3 files that SoundManager requires. MP3 audio files are particularly tricky as they can contain ID1 or ID3 tag information, so just joining them back to back would not create a correct MP3 file. What I needed to do was determine if a particular file had the tag information in it and strip it out if necessary. This took a lot of Googling and studying of the MP3 specification but I eventually managed to ensure the files had no ID2/3 tags in them before joining the files together:

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
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79

Imports System.IO
 
Public Class MP3Concatenator
 
    Public Shared Function Join(ByVal MP3sToJoin As Generic.List(Of String)) As MemoryStream
        Dim ms As New MemoryStream()
        Dim bw As New BinaryWriter(ms)
 
        'loop around each file and remove the tags and then concatenate the files
        For Each mp3File As String In MP3sToJoin
            Dim bytes() As Byte
            Dim fs As New FileStream(mp3File, FileMode.Open, FileAccess.Read, FileShare.ReadWrite)
            Dim br As New BinaryReader(fs)
            Dim audioStart As Integer = 0
 
            'Check for ID3 Tags
            fs.Position = 0
            If (System.Text.Encoding.ASCII.GetString(br.ReadBytes(3)).ToUpper = "ID3") Then
                'position of the header size bytes
                fs.Position = 6
                'MSB of size is Set to 0 and ignored so we need to convert the value
                audioStart = BitsToLow(br.ReadBytes(9))
                'add the header end position to this
                audioStart += 9
            End If
 
            'Check ID1 Tag
            fs.Seek(-128, SeekOrigin.End)
            If (System.Text.Encoding.ASCII.GetString(br.ReadBytes(3)).ToUpper = "TAG") Then
                'there is a ID3v1 tag on the end which needs removing
                fs.Position = audioStart
                bytes = br.ReadBytes(CInt(fs.Length - 128))
            Else
                fs.Position = audioStart
                bytes = br.ReadBytes(CInt(fs.Length))
            End If
 
            bw.Write(bytes)
            br.Close()
            fs.Close()
        Next
 
        ms.Position = 0
        Return ms
    End Function
 
    Private Shared Function BitsToLow(ByVal Size() As Byte) As Integer
        Dim Ret As Integer
        Ret = Size(3)
        If Size(2) <> 0 Then
            If CBool(Size(2) And 1) Then Ret += 128
            If CBool(Size(2) And 2) Then Ret += 256
            If CBool(Size(2) And 4) Then Ret += 512
            If CBool(Size(2) And 8) Then Ret += 1024
            If CBool(Size(2) And 16) Then Ret += 2048
            If CBool(Size(2) And 32) Then Ret += 4096
            If CBool(Size(2) And 64) Then Ret += 8192
        End If
        If Size(1) <> 0 Then
            If CBool(Size(1) And 1) Then Ret += 16384
            If CBool(Size(1) And 2) Then Ret += 32768
            If CBool(Size(1) And 4) Then Ret += 65536
            If CBool(Size(1) And 8) Then Ret += 131072
            If CBool(Size(1) And 16) Then Ret += 262144
            If CBool(Size(1) And 32) Then Ret += 524288
            If CBool(Size(1) And 64) Then Ret += 1048576
        End If
        If Size(0) <> 0 Then
            If CBool(Size(0) And 1) Then Ret += 2097152
            If CBool(Size(0) And 2) Then Ret += 4194304
            If CBool(Size(0) And 4) Then Ret += 8388608
            If CBool(Size(0) And 8) Then Ret += 16777216
            If CBool(Size(0) And 16) Then Ret += 33554432
            If CBool(Size(0) And 32) Then Ret += 67108864
            If CBool(Size(0) And 64) Then Ret += 134217728
        End If
        BitsToLow = Ret
    End Function
End Class

The trickiest part was handling the ID3 tag, as the specification states:

The ID3v2 tag size is encoded with four bytes where the most significant bit (bit 7) is set to zero in every byte, making a total of 28 bits. The zeroed bits are ignored..

Which is why I have the BitsToLow function in the code above.

The resulting concatenation is returned as a memory stream because I knew I could output this directly to the Response without writing the concatenated file to disk:

1
2
3
4
5
6
7
8
9

            ....
            Dim ms As IO.MemoryStream = Nothing
            ms = MP3Concatenator.Join(MP3FileList)
            Response.ContentType = "audio/mpeg"
            Response.ExpiresAbsolute = Date.MinValue
            If ms IsNot Nothing Then Response.OutputStream.Write(ms.GetBuffer, 0, CInt(ms.Length))
            ms.Close()
            Response.End()
            ....

By wiring up the CaptchaAudio.aspx page that returned the concatenated audio to the SoundManager plugin, I could create a link next to the captcha image, that played the letters in the image. Now my captcha control really was fancy.

Tcpdf – Variable Height Table Rows With MultiCell

This post as been superseded. Please see my new post on the subject.

During my use of the excellent tcpdf library for creating pdf documents with php, I came across an interesting problem when creating a pdf featuring a grid of data. Although I could have use an html grid to present the data, I was trying to avoid the use of html and keep to the Cell(), MultiCell(), Text() etc. methods to render the document.

The content of some of the cells in this particular grid were quite long and so some wrapping of text would (and needed to) occur. The MultiCell method handles this and will wrap the text, expanding accordingly. Now I was trying to make a grid row, so needed all of the cells to be the same height – and there lay the problem – I did not know how high to make the other cells. It was a chicken and egg situation: I needed to know the height of the cell before drawing them but did not know how high they needed to be until after I had drawn them.

I studied the documentation hard but realised that the answer was not to try and generate the cells, borders and all, in one pass. I could achieve what I needed by drawing each row twice – once for the content and once for the borders.

The MultiCell function kindly returns the height that the cell was drawn, so by remembering the maximum value it was when drawing a row of cells, I could then go back and draw the borders to that maximum height:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

foreach($data as $row) {
	$maxnocells = 0;
	$cellcount = 0;
	//write text first
	$startX = $pdf->GetX();
	$startY = $pdf->GetY();
	//draw cells and record maximum cellcount
	//cell height is 6 and width is 80
	$cellcount = $pdf->MultiCell(80,6,$row['cell1data'],0,'L',0,0);
	if ($cellcount > $maxnocells ) {$maxnocells = $cellcount;}
	$cellcount = $pdf->MultiCell(80,6,$row['cell2data'],0,'L',0,0);
	if ($cellcount > $maxnocells ) {$maxnocells = $cellcount;}
	$cellcount = $pdf->MultiCell(80,6,$row['cell3data'],0,'L',0,0);
	if ($cellcount > $maxnocells ) {$maxnocells = $cellcount;}
	$pdf->SetXY($startX,$startY);
 
	//now do borders and fill
	//cell height is 6 times the max number of cells
	$pdf->MultiCell(80,$maxnocells * 6,'','LR','L',0,0);
	$pdf->MultiCell(80,$maxnocells * 6,'','LR','L',0,0);
	$pdf->MultiCell(80,$maxnocells * 6,'','LR','L',0,0);
 
	$pdf->Ln();
}

Note that the last parameter I set in MultiCell is set to 0 which means that no new line occurs once a cell is drawn. This means that the cells will appear side-by-side and also means that I need to manually create the new line when I finish drawing the row.

CodeIgniter Error CSS Classes

When version 1.7 of CodeIgniter was released, they introduced a new form validation class that vastly simplified things. I particularly liked the new way in which any errors that occurred during form validation were displayed on screen.
Where as before, an error message for a field was displayed as follows:

<?=$this->validation->myfield_error?>

It seemed much neater with the new validation class:

<?=form_error('myfield');?>

What I found however, was there was no simple way to style a particular field if it it had an error. I wanted a similar method to displaying the error message but which would output the error css class name, if there was an error with the field.
After poking around the new validation code, I discovered that creating such a function would require the extension of two of the base files – the helper form_helper.php would be the place to put the function that I could call from the view and the library form_validation.php would be the place where I could put a function that checks whether a particular field is valid or not.
CodeIgniter provides a way to extend these files easily by placing a file of the same name with the prefix ‘MY_’ inside the appropriate folder in the application directory. So, to create the functionality I wanted, I created a MY_form_helper.php and put the following code into it:

1
2
3
4
5
6
7
8
9
10
11
12
13
14

<?php if ( ! defined('BASEPATH')) exit('No direct script access allowed');
if ( ! function_exists('error_class'))
{
	function error_class($field = '')
	{
		if (FALSE === ($OBJ =& _get_validation_object()))
		{
			return '';
		}
 
		return $OBJ->errorclass($field);
	}
}
?>

Then I created a MY_Form_Validation.php file and placed the following in it:

1
2
3
4
5
6
7
8
9
10
11
12
13
14

<?php  if ( ! defined('BASEPATH')) exit('No direct script access allowed');
 
class MY_Form_validation extends CI_Form_validation {
 
	function errorclass($field = '') {
		if ( !isset($this->_field_data[$field]['error']) OR $this->_field_data[$field]['error'] == '')
		{
			return '';
		} else {
			return ' invalid';
		}
	}
}
?>

With these two changes, I was then able to set the classes of my field as follows:

<input type="text" name="myfield"  value="<?=set_value('myfield');?>" class="<?=error_class('myfield');?>" />

This kept the method in line with the set_value and the form_error functions and set the class to ‘invalid’ when an error occurred.

Windows UAC. Get Over It.

The slating of Microsofts UAC has recently been reborn now that the early adopters of Windows 7 have found security flaws in the new version. It reminded me of the outrage that occurred when it was first introduced in Vista and at this late stage, prompts me to point out something that was conveniently overlooked at the time.
With all those people that seem to stand on their soap box and claim it was the most rediculous thing ever to have come out of Redmond, no one mentioned that UAC was just Microsoft catching up with the other major operating systems. I recall all those smug-faced OSX and Linux owners saying ‘Vista? Don’t bother. Try a decent operating system’. I even recall an Apple advert poking fun at the fact that UAC was introduced. Yet, when I boot up my Hackintosh or Linux VM and do anything that is outside the realms of everyday use, low and behold I get prompted to elevate my security level just like UAC in Vista.
I admit when I first started using Vista it confused me. Whereas before, I could put files here and there, install things with very little thought, all of a sudden I was prompted to stop and think for a second before I do. After some use, the dust has settled and I realise it makes sense. What’s more, the times I am asked to confirm my actions are relatively few, as it only occurs when I install something new, try to copy something to a system area or do anything that may affect the system configuration.
I realise that I am a bit late with this comment but I never got the chance to talk about it at the time. So please, anyone who wants to moan and complain about UAC please accept that in the long run it’s a good thing, appears in some form in every major operating system and is here to stay. So please, get over it.

PuTTY Auto Login at Media Temple (gs)

For one of my projects, I use the Media Temple Grid Service for hosting. They offer ssh access to the hosting, which is superb and coupled with svn repositories, enables me to keep all the source there too.
Using a nice, secure complex password to log on with the ssh client, my IDE and my svn client is very tedious – I have to dig out the password every time. It was therefore imperative that I configured the ssh private-public keys so that I could login automatically.
It took me a while to figure it out first time and recently something happened with their servers that caused my ssh login root directory to change (of course they denied this), causing my ssh keys and a few scripts I had to ‘disappear’. This meant that I needed to setup the keys again and although not as bad as the first time, it still took longer than it should. I therefore log the process here to help others that may struggle with setting them up and as a reminder to myself, should another ‘incident’ occur at Media Temple. Here is the process:

1. Ensure you have putty, puttygen and psftp installed/downloaded
2. Login to your MediaTemple account using putty
3. Type: ssh-keygen -t rsa
4. Follow the instructions to generate the keys. Do not use a pass phrase as we want autologin.
5. You should now have id_rsa and id_rsa.pub in the .ssh directory
6. Change directory into .ssh and type: cat id_rsa.pub >> authorized_keys2
7. FTP into you MediaTemple account using psftp and download the id_rsa file. For Vista users, watch out for UAC. Its best to ensure you download to a safe directory like your desktop
8. Open up puttygen and select from the menu: conversions->Import Key
9. Select the id_rsa key file you downloaded and puttygen should then import it
10. Click Save Private Key and save the .ppk file in a place of your choice
11. Open up putty and fill in Connection->Data ‘login username’. Then go to Connection->SSH->Auth and select your newly created .ppk file when you click on ‘private key file for authentication’
12. Save the putty session. This session should now login automatically

The important point here is that the keys were created on the server and then imported to putty. Many of the guides I found create the keys with putty in the first place and that just did not work for me.

jQuery Plugin Callbacks and Events

I have written a fair few plugins for jQuery for both work and side projects. As any developer should, I am always trying to improve my techniques for creating them so that they are as efficient and maintainable as possible.
Up until recently, my usual technique for callbacks and events would be to include the function in the options that you pass in. For instance, you would setup you plugin as follows:

1
2
3
4
5
6
7
8
9
10
11

(function($) {
    $.fn.myPlugin= function(options) {
        var settings = {
            setting1: 0,
            setting2: ''
        };
        //overload default settings
        if (options) { jQuery.extend(settings, options); }
 
        return this.each(function() {
        .....

This would allow you to pass in any number of settings, including callbacks to utilise at runtime:

1
2
3
4
5

  $('#myselector').myPlugin({ 
        setting1:1234, 
        setting2:'somesetting',
        callback1: function() {}
  });

Yesterday it occurred to me that jQuery had a neat feature that is much better than doing this – custom event binding. jQuery not only allows you to bind the DOM events to elements such as click, focus, keydown etc but because of the way it stores all the bindings, you can bind any number of custom events as well. This means that you can set your events for the plugin as follows:

1
2
3
4

$('#myselector').myPlugin({ 
        setting1:1234, 
        setting2:'somesetting'
}).bind("mycallbackevent", function() {...});

Within your plugin you would need to trigger that event like so:

$(this).trigger("mycallbackevent", [somedata]);

This technique also provides a neater alternative to public functions on your plugin. Say you wanted to initialise the plugin when a link is clicked. I would have previously setup a public function in the plugin to do this:

this.init = function() { .. }

and then run it as follows:

$('a').click(function() { $('pluginselector').get(0).init(); }

With custom events you would setup the event inside the plugin as follows:

$(this).bind('init',function() { .. });

and run as follows:

$('a').click(function() { $('pluginselector').trigger('init'); }

Perhaps I have just been writing plugins incorrectly all this time but that seems much better to me.