CodeFrenzy

A couple of weekends ago, I was lucky enough to be able to go along to HackCamp at the Google offices in London together with the mflow team. It’s the first ‘hack’ day/camp I’ve been to – and although I didn’t stay for the whole weekend nor did I actively participate in the hacking, I got to see what a hack camp is all about. For those of you who have never heard of this term before, a hack day/camp is basically a congregation of hackers, developers, designers (geeks in other words) who get together and spend 24~48 hours building an application using whatever tools they have at their disposal, in small groups or alone.

There were about 10 or so different speakers presenting their APIs at HackCamp, and although you didn’t have to use those APIs in particular, it was certainly very interesting hearing about what others have to offer. One of the APIs that I thought was a great idea was this one – 0870.me. 0870.me is an API that allows you to plug in a standard 08* number (which mobile networks charge a fortune for here in the UK) and it will return to you a matching non-08* number if it can find one in it’s database. How neat is that??

There were speakers from Twitter, Google AppEngine and Buzz to name a few big ones, but also lots of smaller companies. Tum and Demis from mflow gave a talk about the mflow API.

Once all the hacked-up apps had been submitted, I had a quick look and found this little gem: tron. It is very cool to see that someone has used the mflow API in their app!

Anyway, having had a small taste of these ‘hack’ camps, I may just have to keep my eyes out for another that I can go to coz they sure do look like they could be lots of geeky-fun :)

It’s been ages since I last updated my blog and I’m feeling super guilty.

So I thought I’d get stuck into it again and start with a whinge – a whinge about something that had me stumped for way too long this morning.

I have to say that I hardly ever use the out-of-the-box ASP.NET server controls (like asp:DropDownList, asp:TextBox, asp:DataGrid, etc etc) because I prefer to use standard HTML controls, stick runat=”server” tags on them and handle them manually. Call me old-fashioned, tell me I’m doing it the “hard” way… I don’t care. It works and I know how to do it quickly. I always find that when I use the more complete and complex server controls, I get 95% of the work done quickly and then spend hours trying to get the last 5% to work. I don’t like that very much.

Anyway, sometimes I end up having to use a variety of these server controls and today was one such day. Everything was going swimmingly until I tried to insert / update a DateTime property on the type that I had bound my ListView control to, using an ObjectDataSource. I’m in the UK and here, we do not use US-formatted dates. For some reason, no matter where I put localization code to set the locale to “en-GB”, when I hit my “Save” button, my code would throw an exception because it could not convert the string representation of the date into a DateTime object. That’s because it was trying to convert it to an American date. Grr!!!

After a lot of Googling, I came across this article. And there was the answer I was looking for.

Turns out that this is a known bug since ASP.NET 2.0 that Microsoft have decided they “won’t fix”. WTF. They’ve decided not to fix it because ”... the potential fix is relatively dangerous and the workaround is relatively simple…”.

The bug occurs because behind the scenes, .NET is calling ConvertFromInvariantString on all the new values – which is the “en-US” culture. This is clearly a massive assumption on their part but alas, it seems it is “too risky” to fix this.

The “simple workaround” they suggest is to handle the OnItemInserting and OnItemUpdating or OnRowUpdating events and in those handlers, extract the date values from the EventArgs, parse them from strings into DateTime objects and save them back. The following is an example of how one might do this:

1
2
3
4
5
6
7
8
9
10
11

private static void MyList_ItemInserting(object sender, ListViewInsertEventArgs e)
{
	var myDate = e.Values["MyDate"].ToString();
	e.Values["MyDate"] = DateTime.Parse(myDate);
}
 
private static void MyList_ItemUpdating(object sender, ListViewUpdateEventArgs e)
{
	var myDate = e.Values["MyDate"].ToString();
	e.Values["MyDate"] = DateTime.Parse(myDate);
}

I realize this is not a hard fix – but the annoying part was figuring out why it was happening in the first place!!

Anyway, I hope this whinge helps someone else solve this problem… :)

It’s been so long since I posted anything on this blog, I thought it was about time for an update.

First off, I’ve moved to London. Jason and I decided on a whim to come over and see if we could find work – which, as it turns out, wasn’t all that hard. Can’t believe we’ve been here for almost 5 months already (wow, time really does fly by when you’re having fun!!).

Even though the economy was going down the drain and everyone was stressing about the lack of work, Jason and I managed to find work within only 2 weeks of starting to look. Jason scored himself a job at Virgin Games as a Senior .NET Software Engineer and I landed a contract also as a Senior .NET Software Engineer at a small start-up called mFlow. mFlow is actually a desktop application that mixes the music side of iTunes with social networking. The social networking side of it is quite simple really – as a member, you can follow other members and see which tracks and albums they’re buying, as well as which tracks and albums they’re recommending. And the best bit? If someone who is following you buys a song that you recommended, you get paid for it! Great huh? ;)

The atmosphere at mFlow is really cool – we work in an industrial loft style office where everyone sits together. That means everyone interacts with everyone and nobody feels excluded. They’re super well equipped in terms of technology – lots of really grunty servers and well spec’ed dev machines. I personally love my twin 27” monitor set-up drooooool :D.

All the developers at mFlow are incredibly smart and there has been a heap of stuff for me to learn. I work with a couple of other guys on the back-end services whilst a team of other developers work on the front-end. Overall it’s a really cool place to work and it was awesome to be at mFlow when they first launched the application. It’s currently in beta invite-only mode but I hear the ‘real’ launch will be soon. It’s all very exciting because it’s a really neat idea, it’s been well designed and looks reeeeally sexy, and it has the potential to be really big. Fingers crossed eh? :)

In my chosen career, I solve problems everyday. Sometimes the problems I’m given are small and simple to solve, and sometimes they are most definitely not. That, I can deal with. What annoys me is when a problem seems like it should only take five or so minutes to solve and yet somehow, it takes you all day. It’s not that you’re bad at what you do, it’s just that the thing you’re trying to do wasn’t quite as simple as you or anyone else had thought it would be. Or perhaps the framework you’re working with has some nasty little quirks that you weren’t aware of. Or perhaps the thing you’re trying to do just hasn’t been done before, or at least hasn’t been well documented. Whatever the reason, spending all day on a “simple problem” isn’t good for your sanity and makes you feel like you ought to be doing COMP101 all over again.

One such problem that I’ve come across more than once in the last few weeks is how to make a text string fit inside a label or other similar control. If the string is too long to fit, the end should be truncated and replaced by an ellipsis. Sounds pretty straightforward eh? Nope, think again.

After coming up with a few complicated solutions including but not limited to iterating through every character in the string and measuring the length of the string (yes, highly inefficient, I know!), I finally found a solution which seems to do the trick. I feel this amazing solution needs to be shared with everyone so here it is:

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

private string textToDraw = "Hello, how are you? Why does this string not fit in the label that I have drawn? Maybe it is because it is too long?";
 
public Form1()
{
    InitializeComponent();
    label1.Text = ShortenString(textToDraw, this.Width - 10, label1.Font);
}
 
public string ShortenString(string myString, int width, Font font)
{
    string result = string.Copy(myString);
    TextRenderer.MeasureText(result, font, new Size(width, 0), TextFormatFlags.EndEllipsis | TextFormatFlags.ModifyString);
 
    return result;
}
 
private void Form1_ResizeEnd(object sender, EventArgs e)
{
    // Set the width to something - I've just set it to the width of the form - 10 to allow for the ellipsis but
    // I guess this would need to depend on which control the label is to be displayed inside of
    int width = this.Width - 10;
    label1.Text = ShortenString(textToDraw, width, label1.Font);
}

The trick to solving this problem was to use the TextRenderer class and calling its MeasureText function. MeasureText doesn’t seem like the most obvious function to use but it seems to work.

Note that this example is for a C# winforms application. The same could be done for a web application although you’d have to use JavaScript to re-calculate the string when the browser or element is resized.

Has anyone else ever had to solve this particular problem? If so, I’m interested to hear how you did it!

All the gadgets I take with me on holiday… is it too much?

A few years ago I went on a 4 month long trip around the world and I took with me 1 film-based camera, 1 mp3 player and 1 gameboy. Not bad for 4 months I thought. Didn’t take long however before I was robbed and lost the mp3 player and gameboy. Without much choice in the matter, I was left carrying only one ‘electronic’ item, my camera. And guess what? I had a great holiday and didn’t miss any of the other stuff that I thought I would need.

More recently, Chris and I took a 3 week holiday to Canada and between us we took: 3 cameras (+ xtra sd-cards, chargers and cables), 1 laptop, 2 cellphones, 2 ipods (80gig video) and a GPS unit (which we hardly ever used). Hmm. Writing that made me laugh. At myself. You’d think that our bags have nothing but electronics in them. But seriously, at some point whilst packing our bags we figured that we’d need all of that stuff and well, to be honest, we did use it all in one way or another.

So what should you take with you when taking your next holiday? In reality, it all depends on where and when you’re going, how long you’re going for and what you’ll be doing when you’re there (oh and how much of a geek you are). Although there is no definitive answer, I thought I’d share some of the factors that affect my decision making when it comes to picking which gadgets to take with me on holiday.

• Take only what you really think you won’t be able to live without.
  If you don’t take your cellphone and you really need to make a phone call, you can always use a payphone. Alternatively, you could rent a cellphone. Many airports house shops that will rent you a cellphone for the duration of your stay. Or you could even buy a new cheap phone if you absolutely run out of other options. See? You really don’t need to take yours.
• Take small, stream-lined versions of all the electronics you think you’re going to need.
  If you can’t stand the idea of not having your own laptop whilst on holiday then make sure you take a small, light netbook with you, not a full-sized desktop replacement brick. You really don’t want to be carrying 5kg of solid expensive gear with you everywhere you go. You’ll end up a stooped old hunchback if you do.
• Try to detach yourself from caring too much if your stuff does get stolen or damaged.
  This is hard to do, especially if you have some sort of emotional attachment to your gadgets. If it’s irreplaceably special to you or would cost a lot of money to replace, I wouldn’t recommend taking it. Leave it at home, safe and sound. You can’t control how or when some scummy thief out there may take an interest in it and just take it from you. I’m particularly guilty of this and often worry about losing something of value. It just ends up frustrating me that I spend more time worrying about locking away anything of value than I do just relaxing, letting go and enjoying myself.
• If you have a digital camera and are taking lots of photos, back them up early and often.
  If something were to happen to your camera, you don’t want to lose all your photos as well. The simplest option is to take your camera into a shop that can write your photos to a CD/DVD (you’ll find these all over the world) or if you have another gadget with you, you could back them up there instead. For example, I backed some of my photos up onto my Ipod since it had plenty of free space on it.
• If you’re traveling with someone else and you’re both taking gadgets that have the same chargers / cables, take only one charger between you.
  Chances are that you’re not going to need to fully charge your gadget at exactly the same time as whoever else you’re traveling with so if you can, share one between you. Chris and I took two sets of Ipod chargers and cables with us on one of our long trips and ended up sending 1 set home to save space.

If you’re a traveler, what do you take with you?

Since I was about 5 years old and got my little hands on my first Commodore 64, I’ve been comfortable around computers. At first it was all about the games, and then it was about the BASIC programs that my mum and I wrote after school. Later on it was about mIRC, ICQ and horrible geocities pages. And now computers are my job, hobby and obsession. Computers have been in my life for well over 20 years in one way or another and I totally take them for granted. Oh, and did I mention that I’m a girl?

However, this is not the norm. The norm is that girls don’t play with computers and don’t work in IT. But why? Why are there so few girls in IT?

Well, the answer is that there is no one simple reason. It’s something that Universities and education ministries around the world have struggled with for years. For every 200 guys graduating from Computer Science at University, there is 1 girl. This may not be the same at every University around the world but it’s a pretty sad statistic. The same is true in the IT industry. Basically, girls are under-represented in IT almost everywhere you look.

Anyway, I thought I would list a few of the factors that I believe strongly contribute to why girls end up in IT:

• Programming is not taught in all-girls high schools
  This is obviously a pretty big reason why girls don’t take up computer science at University. I actually experienced this myself when I was told in no uncertain terms that “girls don’t program” after asking what programming languages we’d be learning in our Computer Studies course in an all girls’ college I went to in Wellington, New Zealand. The result of this is that any girls who may be interested in trying Computer Science or Computer Engineering at University are already starting off at a disadvantage, especially compared to boys who apart from being curious and probably having played computer games and scripted for years already, may well have been taught some form of programming in high school, simply because they are boys.
• Barbies versus binary
  Parents continue to buy their daughters dolls while their sons get computer games and game consoles. I don’t have children but I have friends who do and I have seen this happen time and time again. If children that young are seeing and being made part of this divide, how can we really expect it to be any different when those same children are trying to decide which career to follow?
• Working with computers is seen to be nerdy and uncool
  There is quite a strong stereotype around geeky computer programmers and it could be possible that a lot of girls, in their formative years, don’t feel that it would be cool to be associated with that sort of image. Peer pressure is a huge motivator for many children and adults, and if that peer pressure is telling girls not to spend time on computers but rather to spend that time reading fashion magazines then that is what they will do.
  
• Could it be that our brains are just wired different?
  I’m no psychologist or doctor so I don’t have the answer to this one. But obviously males and females are different in many other ways so it only follows that our brains probably work differently as well. However, I do believe that this is the last string in the list of reasons why girls aren’t in IT and should most definitely not be used as an excuse for it. As pointed out above, I think there are other reasons why girls don’t pursue IT careers that need to be looked at carefully before we fall back onto a reason that we can’t really prove or do anything about.

• You don’t have to be a social outcast to be a girl software developer
  There is a stereotype surrounding girls in IT – usually very geeky looking, perhaps overweight, no friends, bad skin, etc etc. This is not true and most of the girls I know in IT are very much the opposite – smart, intelligent and strong women. It doesn’t matter what you look like or how you dress; if you want to be a programmer, go for it.
• All guys working in IT are geeky and no fun to hang out with
  Again, this is complete rubbish. Guys who work in IT are for the most part smart, funny, kind and caring. I’ve been friends with geeky guys all my life and in my experience, they’ve been way cooler to hang out with than the mindless jocks who don’t even know how to turn a computer on.
• As a girl in IT, you’ll never be as good as the guys you work with
  Not true. If you’re good at what you do, people will realize that and treat you accordingly. Just make sure you keep up to date with the technologies that you use in your job and don’t be afraid to speak up if you’ve got some ideas you want to share. The guys in your team will come to trust and respect you and trust me, that’s a great feeling.
  

• If you have some geeky friends, ask them for help or advice.
  Asking people who already work in IT what it’s like for them may help you get a better idea of what it might be like for you. Ask them if they can show you some of the work they do, or help you get started with a small project of your own.
• Set up a blog or a small website for yourself.
  Creating a blog for yourself is a great way to start learning about IT. There are plenty of free blogging engines out there like Blogger, WordPress or Textpattern. This is a great introduction into some common but very useful technologies (HTML, JavaScript, CSS) and you’ll end up with a cool blog that you can use to express yourself online. Pretty cool huh?
• Keep up with some technologies that interest you by reading online blogs and articles.
  Whether you’re interested in web programming, desktop application development or linux scripting, there are many resources out there for you to keep an eye on. It’s always handy to be able to tell interviewers that you actively follow some geeky site – it shows that you’re truly interested in IT. Some good sites to check out are Coding Horror, Joel on Software, MSDN Blogs, however, there are many many more out there.
• Go get yourself some IT qualifications.
  If possible, go to University and do a Computer Science degree. However, if you can’t or don’t want to do that just yet, do a short course with a polytechnic or some equivalent. A qualification will not only give you the introduction you’ll need to start off in IT, but it will prove to those who interview you that you’re serious about what you want to do and that you’ve got the smarts to pull it off.
• Don’t let the lack of women in IT put you off.
  For me personally, being in IT has been a lot of fun. I’ve met heaps of really great people and have fit right in with all the development teams I’ve been on. IT guys will treat you like one of them and you’ll never feel left out.

I know this is not rocket science but it’s actually something I’d never had to do until recently. Normally I populate my drop down lists (and repeaters, etc etc) using data from databases, not XML.

However, fear not. Using XML as a datasource in .NET is super easy. Here is one way to do it:

1
2
3
4
5
6
7
8

XmlDataSource source = new XmlDataSource();
source.DataFile = Server.MapPath("Values.xml");
source.XPath = "ParentElement/ValueList/ValueItem";
 
myDropDownList.DataSource = source;
myDropDownList.DataTextField = "Text";
myDropDownList.DataValueField = "Value";
myDropDownList.DataBind();

And your xml file, Values.xml, may look something like:

<?xml version="1.0" encoding="utf-8"  standalone="yes" ?>
<ParentElement>
  <ValueList>
    <ValueItem Text="Item 1" Value="1" />
    <ValueItem Text="Item 2" Value="2" />
  </ValueList>
</ParentElement>

Happy coding!

The following article is part 4 and therefore the last in a 4 part series of articles about developing .NET applications using SSL. Part 1 in this series can be found here, part 2 can be found here and part 3 can be found here.

How to set up a web application to automatically redirect to HTTPS if accessed via HTTP when SSL is required

SSL can be “required” at the site level, folder level or even file level.

If any file that has SSL set to “required” is requested via HTTP, HTTP Error 403;4 “Forbidden” will be returned automatically by IIS.  As far as I can tell, there is no simply way to tell IIS to re-direct to the same page via HTTPS.  This effect can be achieved in two different ways and if “defence in depth” is the preferred approach then both solutions can be implemented together.  Note that both solutions require custom code to be written however.

Redirect with IIS and code:

1. Create a new Web Form (aspx) in your Web Application called RedirectToSSL (or similar).
2. In the Page_Load event handler, insert the following code:
   
   
   ^?View Code CSHARP

   1 2 3 4 5 6 7 8

   string redirectUrl = HttpContext.Current.Request.Url.Query; redirectUrl = redirectUrl.Substring((redirectUrl.LastIndexOf(';')+1) , redirectUrl.Length - (redirectUrl.LastIndexOf(';')+1));   System.UriBuilder secureUrlBuilder = new System.UriBuilder(redirectUrl); secureUrlBuilder.Scheme = System.Uri.UriSchemeHttps; secureUrlBuilder.Port = -1; // Use default port HttpContext.Current.Response.Redirect(secureUrlBuilder.Uri.ToString(), true);

3. The aspx needs to be at least 512 bytes, otherwise IIS won’t redirect to it. So make sure you don’t delete everything out of it, it should never get displayed anyway.
4. In IIS, right-click your Web Site and go to Properties > Custom Errors.
5. Select the “403;4” HTTP error in the list HTTP errors and click the Edit Properties button.
6. Select “URL” from the Message type drop-down list and enter “/RedirectToSSL.aspx” into the URL textbox, as shown below.
7. Press OK twice to close the dialogs down.
8. Restart the website and/or reset IIS to ensure all settings have been applied.
9. Open Internet Explorer and go to: https://<servername>/<securedpage> and verify that you are automatically redirected to the https version of the same page.

1. Insert the following code at the beginning of the Page_Load event handler of every page that needs to be secured:
   
   
   ^?View Code CSHARP

   1 2 3 4 5 6 7 8 9

   // Make sure the request has come through SSL (ie that HTTPS has been used) if (!HttpContext.Current.Request.IsSecureConnection) { System.Uri currentUrl = HttpContext.Current.Request.Url; System.UriBuilder secureUrlBuilder = new System.UriBuilder(currentUrl);   secureUrlBuilder.Scheme = System.Uri.UriSchemeHttps; secureUrlBuilder.Port = -1; HttpContext.Current.Response.Redirect(secureUrlBuilder.Uri.ToString(), true); }




• 

The following article is part 3 in a 4 part series of articles about developing .NET applications using SSL. Part 1 in this series can be found here and part 2 can be found here.

Installing Certificates into the correct Certificate Stores

Applications depend heavily on server SSL and client certificates being installed in the correct certificate stores. Certificates may need to be copied and pasted or exported and imported into stores in order to get the systems correctly working together in a secure manner.

Client application (with client certificate):

1. Local Computer > Trusted Root Authority store should contain the Root CA certificate of the Server SSL Certificate Issuer of the Web application.
2. Current User > Personal store should contain the Client Certificate for the Client application.
3. Current User > Trusted Root Authority store should contain the Root CA certificate of the Client Certificate Issuer of the Client application

1. Local Computer > Personal store should contain the Server SSL Certificate for the Web application.
2. Local Computer > Trusted Root Authority store should contain the Root CA certificate of the Server SSL Certificate Issuer of the Web application.
3. Local Computer > Trusted Root Authority store should contain the Root CA certificate of the Client Certificate Issuer of the Client application.
4. Current User > Trusted Root Authority store should contain the Root CA certificate of the Client Certificate Issuer of the Client application.

1. From the Start menu, select Run, type “mmc” and click OK. This will open the Microsoft Management Console.
2. Click on the File menu option then on “Add/Remove Snap-in”.
3. In the Add/Remove Snap-in dialog, click the Add button.
4. From the Add Standalone Snap-in dialog, select “Certificates” from the list of snap-ins, and click Add.
5. Select the “Computer account” option and click Add. Select “Local computer” and click Finish.
6. If you are logged in as the user that will be used to run the client application, add the “My user account” as well.
7. Now you will be able to look through the Certificates that are installed in all the different stores.

The following article is part 2 in a 4 part series of articles about developing .NET applications using SSL. Part 1 in this series can be found here.

Setting Up Client Certificates

Client certificates can be used to authenticate the client connecting to a web application. The client can either be a browser or a custom written client application.

Requesting a client certificate:

1. Open IE (Firefox didn’t work so well for me…) and type in: http://<certauthservername>/certsrv/
2. Click on the “Request a certificate” link.
3. Click on the “Web Browser Certificate” link.
4. Enter all required information into the form shown above click “Submit”

1. Log onto the Windows Server 2003 machine and open the Certification Authority dialog from Administrative Tools.
2. Expand the root certification authority “Pending Requests” folder.
3. The certificate you just requested should be in the “Pending Requests” folder. Right-click on this request and select All Tasks > Issue. The certificate request will be removed from the “Pending Requests” folder and appear in the “Issued Certificates”.

1. Go back to IE on your web server and type in http://<certauthservername>/certsrv/ again.
2. Click on the “View the status of a pending certificate request” link.
3. Click on the certificate request you want to view.
4. Click on the “Install this certificate” link.
5. Click “Yes” on the dialog that opens to install the client certificate.
6. Click “Yes” on the next dialog as well.

1. On the client machine, open a Microsoft Management Console with the Certificates snap-in. Export the client certificate without the private key. This will generate a .cer file. Copy this file over to the server hosting the web application.
2. On the server hosting the web application, right-click the website in IIS and select “Properties”.
3. Under Properties, go to the “Directory Security” tab.
4. Under the Secure Communications section, click the “Edit” button.
5. In the Secure Communications dialog, change the “Client Certificates” option to “Require client certificates” and check the “Enable client certificate mapping” checkbox.
6. Click on the “Edit” button in the “Enable client certificate mapping” section.
7. This will open up the Account Mappings dialog.
8. Click the “Add” button.
9. Select the client certificate .cer file recently exported from the server that hosts the client application or client browser.
10. Give the mapping a relevant name and choose the windows user that the web application will be run under.
11. Click OK.
12. Click OK to close the Account Mappings dialog and Secure Communications dialogs.
13. Remove all other authentication options in the IIS manager Directory Security setting by clicking on the “Edit” button under Anonymous access and authentication control in the Directory Security tab.
14. Uncheck all authentication options and press OK.
15. Open up a browser and verify that you cannot reach the web site unless you navigate to it from the machine that has the client certificate installed on it.