To start off, I have been thinking about a couple of programming languages, more specifically Python and Ruby on Rails and wondering if I should take some time out to learn them. I was checking out a rather interesting blog post by Amy Hoy that is kind of getting me interested and leaning towards Ruby on Rails…

Yet at the same time I have my PHP/MySQL project that I had sort of set aside to turn my attention to other things of a higher priority, but now I am just about ready to pick up where I left off and continue to work on my web app…

At the same time I have started doing a little more SharePoint, MOSS 2007 stuff and that I will be posting shortly some how-tos.

I found this blog post that I think will be very helpful. I will try his suggestion and provide than update shortly.

Updated: By the way, as long as you follow the instructions, you should have no problem with the upgrade.

WSS 3.0 and Microsoft Office SharePoint 2007 (MOSS) gives one the ability to develop and use custom user controls in SharePoint. Creating a basic custom web user controls is relatively easy and can be done in Visual Studio or notepad if you will. The control is given an extension, “.ascx” and can be loaded into an aspx page and run. In a later post, I will show you have to develop and run a simple custom control in a .Net web environment but in this post I want to talk about an issue I had trying to load and run my user control in a SharePoint page.

After I had created my simple ‘do nothing’ user control, I wanted to make sure that I was able to reference this control and load it into my .aspx page. Once I knew I had the correct references paths and the user control loaded successfully, I could then go in and develop the control. However, two issues arose from this attempt. The first problem was figuring out the path to the user control, the second was enabling permissions to allow the user control to actually run on the SharePoint page.

With regards to the first problem, simply loading the control took a bit time more than I had anticipated since I was not too familiar with the correct paths to use in SharePoint. When you reference your user control in an aspx page, you would use the following code for each user control you load

<%@ Register TagPrefix="MyUserControl" TagName="Mytag"
Src="PathToMyUserControl/MyUserControl.ascx" %>

Of course you will have the code that will load the user control into a placeholder control or panel control or used by another control… To find out how to do so, please read my posting on using custom controls in aspx pages.

The tricky part here is where do you place your user control, then what is the path to insert to point the Src attribute to the user control? What I figured out is that you really want to place your user controls in the CONTROLTEMPLATES folder where you’ll find a number of other SharePoint/WSS 3.0 web controls. This complete path is as follows

c:\program files\Common Files\Microsoft shared\web server extensions\12\TEMPLATE\CONTROLTEMPLATES

As long as this control is in the CONTROLTEMPLATES folder, all you will need in the Register directive’s Src attribute is to point the path to the “/_controls/MyUserControl.ascx”. Your register directive will now look like this

<%@ Register TagPrefix="MyUserControl" TagName="Mytag"
Src="/_controls/MyUserControl.ascx" %>

You will know the path is correct because either your ‘do nothing’ control will load successfully and your page will be blank, or you will get an error to the effect that the control you are trying to run does not have permission to run, or you may just get a HTTP 403 Forbidden error.

This brings me to the second issue that I found. I found that SharePoint will not just simply run any user control you drop in there. After working with the web.config file (someone had mistakenly pointed me to the file), I finally figured out that being that MOSS/WSS 3.0 runs on Windows Server 2003 (or up coming Windows Server 2008), I had to go and actually set permissions on the file itself to allow itself to be run by SharePoint/.Net.

What you will do in this case is right click on the file to bring up the properties popup menu and under security, you will need to give the local WSS_WPG group account permissions to modify the user control. While you are at it, you may also consider giving the local WSS_ADMIN_WPG group account the same permissions.

After you have completed the above steps, you should find that your user control will load and run successfully… well in this case, it does nothing, but since you don’t get an error, then you know you have been able to load your custom user control.

The only thing I would add is that you should be able create a folder in the CONTROLTEMPLATES folder where you can store all your user controls and part of that will require that you make some minor edit to the web.config file. I will update this post shortly to include details on how you can do that.

To get MySQL 5.2.5 (which is the latest release version at the time of this posting), you can go to the downloads section of the site and download the “MySQL Community Server” version which is released under the GNU license. There really isn’t too much technical difference between the “MySQL Community Server” version and the “MySQL Enterprise” version other than with the enterprise version you get the added on services such as updates, testing and monitoring, technical support etc. The kinds of services you would need if you were running your enterprise data and applications on MySQL. Since my needs are rather minimal (for now) I will stick to the community server.

Before I continue, you need to note that I am installing MySQL on a Windows XP Pro machine that I use for development in addition to the everyday computer usage. In addition, I am doing a fresh clean install and not upgrading from a previous version. This is important to note because somewhere in the process I had to delete certain data files and so I didn’t have to worry about losing any valuable data. If you are upgrading and your have data that you want to preserve, my advice is to back-up all your data and then make sure you can restore that data before taking down your old MySQL install!

Now back to the installation. For the most part, installing MySQL 5.x is pretty straightforward.

The first step is installing the core engine. All you need to do is double-click on the installer and follow the wizard prompts to install the engine.

Once that is complete, the next step you’ll take is to configure your database instance. At the end of the core engine install, you will be given the opportunity to configure the install. If you choose choose not to at this time, or you want to reconfigure your instance, you can launch the wizard by going to

Start Menu > Programs > MySQL > MySQL Server 5.0 > MySQL Server Instance Config Wizard

Once you have launched the wizard, once again most of the configuration options are straight forward and will depend on your preferences. Once you have all your preferences selected, you will click the ‘Execute’ button so that the instance will be configured. This is where it can get a little tricky. The wizard performs four main actions. The first two actions, 1. ‘Prepare the Configuration’ and 2. ‘Write Configuration File’ are pretty much straightforward.

The third step, ‘Start Service’ tends to have a problem for many, in that case, they will get the following error,

‘The service could not be started. Error: 0.’.

The screenshot below shows what the error message looks like.

To fix this error, you will want to navigate to your ‘data’ folder in your MySQL install directory, i.e.

…\MySQL\MySQL Server 5.0\data

and delete the following three files

1. ibdata1
2. ib_logfile0
3. ib_logfile1

At this point, you should now be able to rerun the wizard and move part the third step that is ‘Start the Service’.

I also had a problem with the forth step which was ‘Apply Security Settings’. The wizard below shows the error that I got, i.e. ‘

The security settings could not be applied
Error Number 2003.
Can’t connect to MySQL server on ‘localhost’ (10061)

This one was even trickier to solve since none of the advice over the internet was able to solve this problem. Eventually I was able to solve this kind of by luck. All I did was to click the ‘Back’ button, then the ‘Execute’ button to rerun the four configurations steps… and it worked. Just like that! Pretty bizarre I must say.

I haven’t had any problems at all though and I now a full working install of MySQL 5.2.5 working on my desktop. I also have local copy of Wordpress running very smoothly so I know my MySQL install is running fine.

There are a couple of things to note. Even though I claim that I was doing a fresh clean install, I actually had a previous install of MySQL 5.1 which I had corrupted when I was playing around with the settings. Since I didn’t actually have any data stored in the database, I wasn’t too worried about having corrupted the database and rather than trying to repair it, I decided to upgrade to 5.2.5. So I uninstalled the older instance and installed 5.2.5. My understanding about MySQL is that when you uninstall it, the uninstall process does not remove the contents of the …\data folder for obvious reasons. If this is the case, then this will explain why one would get the The service could not be started. Error: 0.’ error in step 3 and consequently why one would need to go in and delete the three mentioned files. This is also why you must back-up your database before upgrading!

Hope this posting is helpful if you have having issues installing/upgrading MySQL

What’s even worse is that there is no longer good documentation on installing and configuring Xitami. In a bid to find out why this was so, plus the fact that there did not really seem to have been an updated (or new) release in a while, plus the fact that the Xitami homepage design had not changed in the last 8 years, I did some digging around and found this interesting entry in Wikipedia.  In essence, the entry notes that development on Xitami ended in about 2000 thus confirming some of my suspicions about no new versions having been released in a while…. more like 7 years!

Although I think Xitami web server is still a decent web server for local machines (or laptops), I have finally come to the conclusion that its time to let Xitami rest in peace.  Since I was able to get PHP configured and working great on IIS 6, I finally decided that I really don’t need Xitami and more and it would make sense that I use IIS instead. Another reason for this decision is that since I have a number of C#, ASP.Net projects planned, rather than switching back and forth between two web servers, I should stick to IIS.

Yes, I also though about installing and using the Apache web server, but I don’t really want to spend the time to download and figure it out. I think a better investment of my time would be to make sure that the rest of my scripting engines, PERL, Ruby, Python, all work with IIS.

Currently the templates are straight XHTML and CSS but I will have to incorporate some PHP code in them. However before I can do this, I must first turn my attention back to the PHP classes that I had started developing.

The plan now is to finish the base Web Engine PHP class that will at least call and display the templates, after which I will then derive a child class that will have the needed logic to call the right templates and input the right content. At the same time, I am also developing my PHP Template Handler class that will handle all template information, links and all that.  I could also create a template class that would hold individual template file information, but that may be overkill… this is something to think about later.

But the reason why I am deriving the Web Engine PHP class is to make the class flexible so that with minimal code I can create customized versions for different types of websites if I need to. But this may not be necessary… it will depend on how I decide to approach the site.. and also on the underlying database design and content…

This decision is long over due. In this day and age, I find it amazing that many of these large internet and telecoms companies are still stuck in the mindset that they need to close and protect their web platforms, and develop all the services in-house. Another example are phone companies that refuse to allow third parties access to the customers and instead prefer to provide all content themselves.

This is a bad idea in many aspects. The biggest reason though is that there is absolutely no way the companies can come up with unlimited content and other innovations for their customers. At the same time, customers will continue to demand more and more from these sites, and eventually, if they don’t get what they want, they will simply abandon these sites for those that offer what they want.

Take the example of Firefox browser. The fact that the developers allow user to build and install their own extensions and add-ins makes Firefox a very powerful browser. And this fact alone make Firefox a very compelling browser for me to use. I have downloaded a large number of extensions that I use often and makes my browser experience much more enjoyable and efficient. If I don’t like the way my browser looks, I simply install a more appealing theme. I really don’t like IE7. I think its really ugly and not very extensible. Yet there are some websites that require the use of IE. The funny thing here is that there is a Firefox add-in that allows me to open web pages in IE within the Firefox browser.

In the same way, and even more so, I do believe that many of these tech companies, that target mass customers need to have the same sort of mindset if they want to continue to grow and attract new customers.

They need to tear down the walled gardens they have maintained and open up to third party developers and other companies…