Blog

I don’t generally have a problem importing an Excel spread sheet into one of my SQL Server tables. But today would end my run of Excel importing perfection.

I experienced an problem where all rows that only contained numbers were ending up as NULL in my table after import, which I thought was strange since the Excel spread sheet did not contain empty cells. It contained a mixture of data formats: text and numbers.

I decided to format all rows in my spread sheet to text and try another re-import. No change.

After much experimentation, the solution was to copy all columns and paste them into Notepad in order to remove all formatting inherited from Excel. I then re-copied all my data from Notepad back into my spread sheet and carried out another import. Lo and behold it worked!

I don’t understand why I had this problem. It could have been due to the fact the spread sheet contained cells of different data formats and causing confusing through the import process.

Back in 2007 I started blogging mainly for one selfish reason - to have an online repository of how I've approached things technically to refer back to when required. When I find things interesting, I like to document them for me to expand on later. If a public user wants to expand or contribute to what I’ve posted, then they are welcome to do it.

Blogging soon flourished into something more beneficial and pushed me to better myself in all aspects of web & application development. It had turned me from being a very introverted cowboy-developer to an extrovert with the confidence to push the boundaries in my day to day job just so I could have a reason to blog about it and publicly display what I know.

I highly recommend blogging to anyone, especially in the technical industry. Reading other blogs has shown me that a solution to a problem is always up for interpretation. For example, I may find the solution to one of my issues on another site that I can expand further on my own blog (with references to the original author, of course).

This year, I decided to take things one step further and joined a well known open community called StackOverflow. So far, it's been a great experience and I recently broke the 1000 points barrier. It took a lot of blood, sweat and tears. In some ways, knowing how people rate your answers in a forum can help show you where your skill set is lacking. I'm sure if I look back on some of my earlier posts I've made some shockingly bad suggestions. Thankfully, there are more experienced posters who set you on the right direction.

Blogging and contributing to StackOverflow can also have an unexpected impact - employment. The web development industry is very competitive and it's up to you to set yourself apart from the rest. Potential employers can have a great insight to what you're capable of and demonstrates you can communicate your technical knowledge.

If I known this earlier in my career, I'm sure things would've been different and would have had the opportunity to find a job in web development sooner. So start early even if you're studying at college or university. When the time comes to getting a job, you can truly show your potential!

I had around 2000 webpage URL’s listed in a text file that needed to be generated into a simple Google sitemap.

I decided to create a quick Google Sitemap generator console application fit for purpose. The program iterates through each line of a text file and parses it to a XmlTextWriter to create the required XML format.

Feel free to copy and make modifications to the code below.

Code:

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.IO;
using System.Xml;

namespace GoogleSitemapGenerator
{
    class Program
    {
        static void Main(string[] args)
        {
            string textFileLocation = String.Empty;

            if (args != null && args.Length > 0)
            {
                textFileLocation = args[0];
            }

            if (!String.IsNullOrEmpty(textFileLocation))
            {
                string fullSitemapPath = String.Format("{0}sitemap.xml", GetCurrentFileDirectory(textFileLocation));

                //Read text file
                StreamReader sr = File.OpenText(textFileLocation);

                using (XmlTextWriter xmlWriter = new XmlTextWriter(fullSitemapPath, Encoding.UTF8))
                {
                    xmlWriter.WriteStartDocument();
                    xmlWriter.WriteStartElement("urlset");
                    xmlWriter.WriteAttributeString("xmlns", "http://www.sitemaps.org/schemas/sitemap/0.9");

                    while (!sr.EndOfStream)
                    {
                        string currentLine = sr.ReadLine();

                        if (!String.IsNullOrEmpty(currentLine))
                        {
                            xmlWriter.WriteStartElement("url");
                            xmlWriter.WriteElementString("loc", currentLine);
                            xmlWriter.WriteElementString("lastmod", DateTime.Now.ToString("yyyy-MM-dd"));
                            //xmlWriter.WriteElementString("changefreq", "weekly");
                            //xmlWriter.WriteElementString("priority", "1.0");

                            xmlWriter.WriteEndElement();
                        }
                    }

                    xmlWriter.WriteEndElement();
                    xmlWriter.WriteEndDocument();
                    xmlWriter.Flush();

                    if (File.Exists(fullSitemapPath))
                        Console.Write("Sitemap successfully created at: {0}", fullSitemapPath);
                    else
                        Console.Write("Sitemap has not been generated. Please check your text file for any problems.");

                }
            }
            else
            {
                Console.Write("Please enter the full path to where the text file is situated.");
            }
        }

        static string GetCurrentFileDirectory(string path)
        {
            string[] pathArr = path.Split('\\');

            string newPath = String.Empty;

            for (int i = 0; i < pathArr.Length - 1; i++)
            {
                newPath += pathArr[i] + "\\";
            }

            return newPath;
        }
    }
}

I will be uploading a the console application project including the executable shortly.

Today I came across this really interesting tweet on my Twitter timeline today:

Read about why we’re deleting our Facebook page: facebook.com/limitedpressin… — Limited Run (@limitedrun) July 30, 2012

Limited Run, posted on their Facebook profile stating that they would be deleting their account due to the amount Facebook is charging for clicks on their advertising. Here’s the interesting part: About 80% of the clicks Facebook charged Limited Run, JavaScript wasn't on. And if the person clicking the ad doesn't have JavaScript, it's very difficult for an analytics service to verify the click. Only 1-2% of people going to their site have JavaScript disabled, not 80% like the clicks coming from Facebook.

Interesting stuff.

Before Limited Run takes down their Facebook profile, I’ve attached a screenshot of their post below:

Reading this post today reminded me on a news article I read on “virtual likes” and how advertising through Facebook doesn’t necessarily mean you’ll be any better off. It all comes down to the level of engagement user’s have with a profile page. If users are just liking the page and not interacting with your posts or general content, those likes are worth nothing. Some companies are wising up to the effectiveness of Facebook’s advertising strategy.

Limited Run isn’t the first to ditch Facebook ad’s, General Motor’s pulled away from Facebook ad’s earlier this year due to the ad’s Facebook produce do not have the visual impact needed to justify the cost.

I think certain aspects of Facebook is a joke filled mostly of people looking for attention, not an effective marketing tool.

If I need to login and authenticate a Facebook user in my ASP.NET website, I either use the Facebook Connect's JavaScript library or SocialAuth.NET. Even though these two methods are sufficient for the purpose, I don't think it's the most ideal or efficient way.

The Facebook Connect JavaScript library is quite basic and doesn't have the flexibility required for full .NET integration through FormsAuthentication. Whereas SocialAuth.NET provides full .NET integration and all authentication is done server-side with minimal development.

I'd say if you are looking for a straight-forward way to integrate social site authentication, SocialAuth.NET is the way to go. It's API can communicate with other social sites such as Twitter, LinkedIn and Gmail.

Recently, I found a better and more efficient way to authenticate Facebook users on my site using Graph API and Hammock.

Hammock is a C# a REST library for .NET that greatly simplifies consuming and wrapping RESTful services. This allows us to embrace the social site’s core technology instead of using varied SDK's or API's. There are many community driven frameworks and API's readily available on the Internet, but they can really cause problems if they evolve too quickly or haven’t been thoroughly tested.

Suddenelfilio, has written a useful blog post on connecting Facebook using Hammock. You will see by his example that you can interact with Facebook anyway you want.

The same principle could also be applied to other website API's that use REST based services, such as Twitter.

I always found writing code to read an RSS feed within my .NET application very time-consuming and long-winded. My RSS code was always a combination of using WebRequest, WebResponse, Stream, XmlDocument, XmlNodeList and XmlNode. That’s a lot of classes just to read an RSS feed.

Yesterday, I stumbled on an interesting piece of code on my favourite programming site StackOverflow.com, where someone asked how to parse an RSS feed in ASP.NET. The answer was surprisingly simple. RSS feeds can now be consumed using the System.ServiceModel.Syndication namespace in .NET 3.5 SP1. All you need is two lines of code:

var reader = XmlReader.Create("http://mysite.com/feeds/serializedFeed.xml");
var feed = SyndicationFeed.Load(reader);

Here’s a full example on how we can iterate through through the SyndicationFeed class:

public static List<BlogPost> Get(string rssFeedUrl)
{
    var reader = XmlReader.Create(rssFeedUrl);
    var feed = SyndicationFeed.Load(reader);

    List<BlogPost> postList = new List<BlogPost>();

    //Loop through all items in the SyndicationFeed
    foreach (var i in feed.Items)
    {
        BlogPost bp = new BlogPost();
        bp.Title = i.Title.Text;
        bp.Body = i.Summary.Text;
        bp.Url = i.Links[0].Uri.OriginalString;
        postList.Add(bp);
    }

    return postList;
}

That’s too simple, especially when compared to the 70 lines of code I normally use to do the exact same thing.

Ever since I decided to expand my online presence, I thought the best step would be to have a better domain name. My current domain name is around twenty-nine characters in length. Ouch! So I was determined to find another name that was shorter and easier to remember.

Ever since “.me” top level domain (TLD) came out, I snapped up “surinder.me”, partly because all other domains with my first name were gone (you know who you are!) and the “.me” extension seemed to fulfil what I wanted my website to focus on. ME! Having said that, I would have loved to get a “.com” domain, but I guess that’s what happens when you enter the online world so late.

I was ready to move over all my content to “surinder.me” until one on my techy friends told me that things are still undecided when it comes to “.me” TLD’s in general. Originally, the “.me” extension was assigned to Montenegro’s locale only. But it’s fast gained traction over the years due to it’s simplicity and wide range of possible domain names. Even companies such as Microsoft, Facebook, Wordpress and Samsung rushed to register their “.me” domains. Hence the reason why I decided to get one.

Companies seem to be using “.me” extensions for either URL shortening services or redirects to partner sites with “.com” extensions. It doesn’t fill me with much confidence when “.me” extensions are used this way. Google’s software engineer, Matt Cutts wrote a reassuring post on his Google+ profile earlier this year by stating:

“…regardless of the top-level domain (TLD). Google will attempt to rank new TLDs appropriately, but I don't expect a new TLD to get any kind of initial preference over .com…If you want to register an entirely new TLD for other reasons, that's your choice, but you shouldn't register a TLD in the mistaken belief that you'll get some sort of boost in search engine rankings.”

This should put all my “.me” fears to rest…right? Well it’s nice to know Google won’t penalise a site based on an extension. In the world of web, a search optimised site is king (as it should be). It’s nice that Google have given “.me” (as a country extension) global status given the nature of how its been used of late. But if you check Google’s Geotargetable Domains article, the text in brackets worries me.

I get the feeling you can’t go wrong with a “.com” domain providing you can find something meaningful to your cause. Steps are being made in the right direction for gccTLD’s. For example, Webmaster Tools gives you the option to geographically target your “.me” site. However, I can’t find anything concrete to alleviate my concerns in the long-run.

So where does this leave me? Well, we’ll just have to find out if my future domain contains a .me extension.

I’ve been looking for a suitable replacement domain name for a while now and even making purchases that has some reference to my name. Since I’m not having much luck with new registrations, I decided to snoop around for domains that are up for sale. Lo and behold I found my ideal second-hand “.com” domain: surinder.com. However, there’s a catch…

Currently “surinder.com” sales price is £5000! Whaaaa!!!!!?????

I know Surinder is a really cool name and damn right popular with the ladies, but seriously £5000. Even I wouldn’t have the audacity to sell my domain for that much (offers will be accepted though :-) ).

When reading numerous articles on how domain names are valued, it seems to evolve around the sum of the domains generic value and the value of its traffic. So its not exactly clear cut. I highly recommend reading this post on “How To Value a Domain Name”, it has some really useful information.

One of my colleagues pointed my to a really useful tool called HTTrack, that has the ability to download a website from the internet to a local directory by simply copying and pasting the site URL of your choice.

I found this tool helpful when working on an existing site where I was unable to 100% recreate the files and directory structure within my local development environment. For example, you could be developing a new micro-site for an existing online site you may not currently have in your local environment.

HTTrack has the ability to build recursively all directories, images and other files from the server to your computer. What’s even better, HTTrack arranges the original site's relative link-structure.

Microsoft’s new command-line tool, PowerShell has been out for quite a few years now and I thought today will be the day I would start using it. I needed to write a script that would move n number of files from one directory to another. This job seemed a perfect fit for PowerShell.

#Get 'n' number of files
$FileLimit = 10 

#Destination for files
$DropDirectory = "C:\Drop\"

$PickupDirectory = Get-ChildItem -Path "C:\Pickup\"

$Counter = 0
foreach ($file in $PickupDirectory)
{
    if ($Counter -ne $FileLimit)
    {
        $Destination = $DropDirectory+$file.Name

        Write-Host $file.FullName #Output file fullname to screen
        Write-Host $Destination   #Output Full Destination path to screen
        
        Move-Item $file.FullName -destination $Destination
        $Counter++
    }  
}

From the get go, I was really impressed with the flexibility of the scripting language. This is where command line fails. It is sufficient for simple tasks but not so much for complex jobs.

As you can see from the code above, I can implement complex operations that supports variables, conditional statements, loops (while, do, for, and foreach), and that’s just the start. I don’t know why I hadn’t used PowerShell sooner. If I didn’t have the option to use PowerShell, I would have probably created a C# service or executable to do the exact same thing. Time saver!

Since PowerShell is built on the .NET Framework, Windows PowerShell helps control and automate the administration of the operating system and applications that run on Windows. So if you are a C# programmer, you should feel comfortable in writing PowerShell scripts. All you need to be aware of is syntax differences when declaring variables and keywords.

To end with, I will quote an amusing forum post I found when researching the difference between good ol’ Command Prompt and PowerShell:

“PowerShell has a default blue background and Command Prompt has a default black background.”