IRepository: one size does not fit all

I’ve been spending a lot of time putting TDD/DDD into practice lately, and obsessing over all sorts of small-yet-important details as I try concepts and patterns for myself.

One pattern that has recently become popular in mainstream .NET is IRepository<T>. Originally documented in PoEAA, a repository is an adapter that can read and write objects from the database as if it were an in-memory collection like an array. This is called persistence ignorance (PI), and lets you forget about underlying database/ORM semantics.

Somewhere along the line however, someone realized that with generics you can create a single, shared interface that will support pretty much any operation you could ever want to do with any sort of object. They usually look something like this:

public interface IRepository<T>
    T GetById(int id);
    IEnumerable<T> GetAll();
    IEnumerable<T> FindAll(IDictionary<string, object> propertyValuePairs);
    T FindOne(IDictionary<string, object> propertyValuePairs);
    T SaveOrUpdate(T entity);
    void Delete(T entity);

If you need to add any custom behaviour, you can simply subclass it — e.g. IProjectRepository : IRepository — and add new methods there. That’s pretty handy for a quick forms-over-LINQ-to-SQL application.

However, I don’t believe this is satisfactory for applications using domain-driven design (DDD), as repositories can vary greatly in their capabilities. For example, some repositories will allow aggregates to be added, but not deleted (like legal records). Others might be completely read-only. Trying to shoehorn them all into a one-size-fits-all IRepository<T> interface will simple result in a lot of leaky abstractions: you could end up with a Remove() method that is available but always throws InvalidOperationException, or developer team rules like “do not never call Save() on RepositoryX”. That would be pretty bad, so what else can we do instead?

Possibility #2: no common repository interface at all

The first alternative is simply dropping the IRepository<T> base, and make a totally custom repository for each aggregate.

public interface IProjectRepository
    Project GetById(int id);
    void Delete(Project entity);
    // ... etc

This is good, because we won’t inherit a bunch of crap we don’t want, but similar repositories will have a lot of duplicate code. Making method names consistent is now left to the developer — we could end up with one repository having Load(), another Get(), another Retrieve() etc. This isn’t very good.

Thinking about it, a lot of repositories are going to fit into one or two broad categories that share a few common methods — those that support reading and writing, and those that only support reading. What if we extracted these categories out into semi-specialized base interfaces?

Possibility #3: IReadOnlyRepository and IMutableRepository

What if we provided base interfaces for common repository types like this:

This is better, but still doesn’t satisfy all needs. Providing a GetAll() method might be helpful for a small collection of aggregates that we enumerate over often, but it wouldn’t make so much sense for a database of a million customers. We still need to be able to include/exclude standard capabilities at a more granular level.

Possibility #4: Repository Traits

Let’s create a whole load of little fine-grained interfaces — one for each standard method a repository might have.

public interface ICanGetAll<T>
    IEnumerable<T> GetAll();

public interface ICanGetById<TEntity, TKey>
    TEntity GetById(TKey id);

public interface ICanRemove<T>
    void Remove(T entity);

public interface ICanSave<T>
    void Save(T entity);

public interface ICanGetCount
    int GetCount();

// ...etc

I am calling these Repository Traits. When defining a repository interface for a particular aggregate, you can explicitly pick and choose which methods it should have, as well as adding your own custom ones:

public interface IProjectRepository :
    ICanGetById<Project, int>,
    IEnumerable<Project> GetProjectsForUser(User user);

This lets you define repositories that can do as little or as much as they need to, and no more. If you recognize a new trait that may be shared by several repositories — e.g., ICanDeleteAll — all you need to do is define and implement a new interface.

Side note: concrete repositories can still have generic bases

Out of interest, here’s what my concrete PersonRepository looks like:

There’s very little new code in it, because all of the common repository traits are already satisfied by a generic, one-size-fits-all NHibernateRepository base class (which must be completely hidden from external callers!). The IPersonRepository just defines which subset of its methods are available to the domain layer.

Gallio, the framework-agnostic test runner for .NET

Over the past week, I’ve been converting a bunch of unit tests written with NUnit to Microsoft’s MSTest framework, the tool of choice for this particular project.

I’m not a big fan of either the MSTest framework, nor Visual Studio’s in-built test runner. It feels like you’re running unit tests in Access. Plus, the last thing my Visual Studio needs is to be cluttered up with more windows and toolbars.

Anyway, here’s where Gallio comes in. It’s a open source, framework-agnostic test runner that supports MbUnit, MSTest, NBehave, NUnit, and xUnit.Net frameworks. Running tests is quite similar to NUnit — you add test assembly paths to a .gallio project file and fire them up in Icarus, Gallio’s graphical test runner:

As well as being a nicer test runner for MSTest, Gallio is ideal for managing tests in a mixed-framework environment. It also integrates with MSBuild, NAnt, CruiseControl.NET for automated testing during your build process.

Nested FOR XML results with SQL Server’s PATH mode

Today, while doing some work on a highly data (not object) driven .NET application, I needed a query output as XML from the application’s SQL Server 2005 database. I wanted:

  • Nicely formatted and properly mapped XML (e.g. no <row> elements as found in FOR XML RAW mode)
  • To be able to easily map columns to XML elements and attributes
  • A single root node, so I can load it into an XmlDocument without having to create the root node myself
  • Nested child elements
  • Not to have to turn my elegant little query into a huge mess of esoteric T-SQL (as with [Explicit!1!Mode])

I discovered that all this is surprisingly easy to achieve all of these things with SQL Server 2005′s FOR XML PATH mode. (I say surprising, because I’ve tried this sort of thing with FOR XML AUTO a few times before under SQL Server 2000, and gave up each time).

Here’s a quick example I’ve created using the venerable AdventureWorks example database, with comments against all the important bits:

    -- Map columns to XML attributes/elements with XPath selectors.
    category.ProductCategoryID AS '@id',
    category.Name AS '@name',
        -- Use a sub query for child elements.
            ProductID AS '@id',
            Name AS '@name',
            ListPrice AS '@price'
            ProductCategoryID = category.ProductCategoryID
            XML PATH('product'), -- The element name for each row.
            TYPE -- Column is typed so it nests as XML, not text.
    AS 'products' -- The root element name for this child collection.
    SalesLT.ProductCategory category
    XML PATH('category'), -- The element name for each row.
    ROOT('categories'-- The root element name for this result set.

As you can see, we’ve mapped columns to attributes/elements with XPath selectors, and set row and root element names with PATH() and ROOT() respectively.

Plus, by specifying my own names for everything, I was also able to address the difference in capitalization, prefixing and pluralization style between the AdventureWorks’ database table names and common XML.

Running this query produces output in the following format. Note the root nodes for both outer and child collections:

  <category id="4" name="Accessories" />
  <category id="24" name="Gloves">
      <product id="858" name="Half-Finger Gloves, S" price="24.4900" />
      <product id="859" name="Half-Finger Gloves, M" price="24.4900" />
      <product id="860" name="Half-Finger Gloves, L" price="24.4900" />
      <product id="861" name="Full-Finger Gloves, S" price="37.9900" />
      <product id="862" name="Full-Finger Gloves, M" price="37.9900" />
      <product id="863" name="Full-Finger Gloves, L" price="37.9900" />
  <category id="35" name="Helmets">
      <product id="707" name="Sport-100 Helmet, Red" price="34.9900" />
      <product id="708" name="Sport-100 Helmet, Black" price="34.9900" />
      <product id="711" name="Sport-100 Helmet, Blue" price="34.9900" />

Strategies for resource-based 404 errors in ASP.NET MVC

There are already a couple of patterns for handling 404 errors in ASP.NET MVC.

  • For invalid routes, you can add a catch-all {*url} route to match “anything else” that couldn’t be handled by any other route.
  • For invalid controllers and controller actions, you can implement your own IControllerFactory with inbuilt error handling.

But what happens when a user invokes a valid action on a valid controller, and the requested resource — product, article, sub-category etc — doesn’t exist? Let’s investigate some of the options available.

Classic ASP.NET custom errors: easy, but not that great

The easiest option is to use ASP.NET’s custom errors feature, redirecting users to special pages according to the status code of the exception that was caught.

All you need to do is to uncomment the following section from your web.config, and create a new file called FileNotFound.htm:

<customErrors mode="On">
  <error statusCode="404" redirect="FileNotFound.htm" />

Then you can start throwing 404 HttpExceptions from your action method.

// Get information about a product.
public ActionResult Detail(int? id)
    // Look up the product by ID.
    Product product = products.FirstOrDefault(p => p.Id == id);
    // If it wasn't found, throw a 404 HttpException.
    if (product == null)
        throw new HttpException(404, "Product not found.");
    return View(product);

When a user requests the detail for a non-existent product, here’s what they’ll see with a static FileNotFound.htm 404 page (style copied from the application’s MasterPage):

This method is very easy to implement but has a few disadvantages — particularly from the user’s point of view.

  • The redirection and error page return HTTP status codes 302 Found and 200 OK respectively. While most humans might not notice the difference between 302 Found and 404 Not Found, search engines certainly appreciate the distinction when indexing content. Additionally, under REST principles, redirecting to a different path is not an appropriate response when an particular resource cannot be found. The server should simply return a 404 not found status code.
  • The path to the requested page is passed externally, and the user is redirected to a traditional .html or .aspx page. This looks noisy and unprofessional when used alongside ASP.NET MVC’s elegant paths. The path redirection also makes it difficult for users to retype and correct badly-entered URLs.
  • These error pages are handled outside the ASP.NET MVC framework. ViewData, Models and Filters are not available — data access, logging and master pages will be implemented differently to the rest of your web site.

While they do arguably work, ASP.NET custom errors are clearly not the best tool for the job.

ASP.NET MVC HandleErrorAttribute: getting warmer

The next option is the ASP.NET MVC framework’s built in HandleErrorAttribute, which renders a special error view when uncaught exceptions are thrown from a controller action.

// My custom exception type for 404 errors.
public class ResourceNotFoundException : Exception {}


// Render the ResourceNotFound view when a ResourceNotFoundException is thrown.
[HandleError(ExceptionType = typeof(ResourceNotFoundException),
    View = "ResourceNotFound")]
public class ProductsController : Controller
    // Get information about a product.
    public ActionResult Detail(int? id)
        // Look up the product by ID.
        Product product = products.FirstOrDefault(p => p.Id == id);
        // If it wasn't found, throw a ResourceNotFoundException.
        if (product == null)
            throw new ResourceNotFoundException();
        return View(product);

To achieve this, I defined my own ResourceNotFoundException class, and decorated my controller class with a HandleErrorAttribute. Actually you would probably decorate it twice — one for all exceptions, and one specifically for the ResourceNotFoundException type.

Anyway, in this example, when an action throws a ResourceNotFoundException, the HandleErrorAttribute filter will catch it and render a view called ResourceNotFound. If you don’t specify a view, HandleErrorAttribute will look for a default view called Error. I have chosen to use a different one, because I don’t want 404 errors reported in the same way as general application errors.

This renders better results — no URL redirection, and the code is contained within the ASP.NET MVC framework. However, there’s still a problem.

As you can see, the invalid resource request is coming back as a 500 Server Error. This isn’t by any flaw of the HandleErrorAttribute class; it’s simply designed to handle application errors, not 404s. In fact, it will explicitly ignore any exceptions that have an HTTP status code other than 500.

While it works pretty well, ASP.NET MVC’s HandleErrorAttribute isn’t really suited as a 404 handler.

Custom IExceptionAttribute

HandleErrorAttribute is a filter — a feature of the ASP.NET MVC framework that allows you to execute before or after a controller action is called, or when an exception is thrown. They’re great for checking authentication, logging page requests, exception handling and other cross-cutting concerns.

Here’s a filter I’ve created — it’s effectively a clone of the built in HandleErrorAttribute, except that it ignores everything except ResourceNotFoundExceptions, and it sets the HTTP status code to 404 instead of 500.

public class HandleResourceNotFoundAttribute : FilterAttribute, IExceptionFilter
    public void OnException(ExceptionContext filterContext)
        Controller controller = filterContext.Controller as Controller;
        if (controller == null || filterContext.ExceptionHandled)
        Exception exception = filterContext.Exception;
        if (exception == null)
        // Action method exceptions will be wrapped in a
        // TargetInvocationException since they're invoked using
        // reflection, so we have to unwrap it.
        if (exception is TargetInvocationException)
            exception = exception.InnerException;
        // If this is not a ResourceNotFoundException error, ignore it.
        if (!(exception is ResourceNotFoundException))
        filterContext.Result = new ViewResult()
            TempData = controller.TempData,
            ViewName = View
        filterContext.ExceptionHandled = true;
        filterContext.HttpContext.Response.StatusCode = 404;

As with HandleErrorAttribute, you simply decorate the controller class with a HandleResourceNotFound attribute.

// Render the ResourceNotFound view when a ResourceNotFoundException is thrown.
public class ProductsController : Controller
    // Get information about a product.
    public ActionResult Detail(int? id)
        // Look up the product by ID.
        Product product = products.FirstOrDefault(p => p.Id == id);
        // If it wasn't found, throw a ResourceNotFoundException.
        if (product == null)
            throw new ResourceNotFoundException();
        return View(product);

This filter produces very similar output to that of the HandleErrorAttribute, but with the correct 404 HTTP status code. This is exactly what we want — an easy-to-generate 404 status code, raised and presented within the ASP.NET MVC framework.

In part 2 of this article, I’ll discuss methods for improving the quality of these error pages, and why a single generic 404 error page isn’t sufficient for a modern, rich web application.

Three common ASP.NET MVC URL routing issues

Following on from my post last week about consistent ASP.NET MVC URLs, here are a few tips that helped when I was first getting started with ASP.NET MVC and its URL routing system.

Have you got your routes registered in the correct order?

ASP.NET MVC checks incoming URL requests against routes in the order they were registered, so you need to add them in the order of most to least specific. Thus, even though you might have a very specific route defined later on that matches the incoming URL perfectly, ASP.NET MVC will choose whichever route matches first. You can’t just drop in extra rules at the bottom of the list and expect them to work.

For example, if you have ten routes, with an {*anything} pattern registered first, the other nine routes will always be ignored, because the anything route will always be matched first.

In a real scenario, one application I am working on uses the default ASP.NET MVC route for most pages, but uses some extra routing rules as well.

  • Ignore any request for the Charts directory, and let ASP.NET handle those URLs as normal.
  • Allow browsing one controller’s resources by name, instead of by numeric ID.
  • Anything else can go to the 404 handler.

Here’s the code I use to implement these rules, in the correct order:

public static void RegisterRoutes(RouteCollection routes)
	// Ignore dynamic image handlers e.g. Charts/Summary.ashx?year=2008&month=08

	// Allow browsing categories by name, instead of by ID.
	routes.MapRoute("Categories", "Categories/{action}/{name}",
		new { controller = "Categories", action = "Index", name = "" }

	// All other pages use the default route.
	routes.MapRoute("Default", "{controller}/{action}/{id}",
		new { controller = "Applications", action = "Index", id = "" }

	// Show a 404 error page for anything else.
	routes.MapRoute("Error", "{*url}",
		new { controller = "Error", action = "404" }

Have you set the default redirection in Default.aspx.cs?

So you’ve changed the default controller, but your application still directs to /Home. If you take a look in Default.aspx.cs, you’ll see the following:

public void Page_Load(object sender, System.EventArgs e)

This code gets called when a user browses to the directory where your application is installed, but doesn’t specify any controller (i.e. an empty URL path). It redirects requests to a default controller, which, in the case for a new ASP.NET MVC Web Application project, is called Home.

If you want to use something other than HomeController, this path needs to be changed as well.

Do your parameter names match?

Here’s another one that might catch you off-guard. The ASP.NET MVC routing system uses Reflection to assign parameter values by argument name when a controller action is called — not by argument index as I originally thought. For example:

routes.MapRoute("Product", "Product/{action}/{id}",
    new { controller = "Product", action = "Detail" });
public class ProductController : Controller
    public ActionResult Detail(int? productID) { .. }

Here, the product ID can’t be passed through automatically because the Detail() method’s parameter is called productID, where in the route it is defined as simply id.

When browsed to, the first URL will work, because you’ve explicitly specified a query string parameter with the right name. The second (desired) URL will not work, because the name of the parameter in the route doesn’t match that of the controller method.

If you want the parameter to be passed through automatically, the name in the controller action must match the name in the route.

C# 3.0’s var keyword: Jeff Atwood gets it all wrong

One of the new features of C# 3.0 is the var keyword, which can be used to implicitly declare a variable’s type. For example, instead of declaring the type, you can just write var instead, and the compiler will figure out what type name should be from the value assigned to it:

// A variable explicitly declared as a certain type.
string name = "Richard";

// A variable implicitly declared as the type being assigned to it.
var name = "Richard";

Note that a var isn’t a variant, so it’s quite different from say, a var in javascript. C# vars are strongly typed, so you can’t re-assign them with values of different types:

var name = "Richard"
name = 62; // error, can't assign an int to a string

Vars were introduced to support the use of anonymous types, which don’t have type names. For example:

var me = { Name: "Richard", Age: 22 };

But there’s nothing to stop you from using vars with regular types. In a recent article on, Jeff Atwood recommends using vars “whenever and wherever it makes [your] code more concise”. Guessing from his examples, this means everywhere you possibly can.

The logic behind this is that it improves your code by eliminating the redundancy of having to write type names twice — once for the variable declaration, and again for a constructor.

There’s always a tradeoff between verbosity and conciseness, but I have an awfully hard time defending the unnecessarily verbose way objects were typically declared in C# and Java.

BufferedReader br = new BufferedReader (new FileReader(name));

Who came up with this stuff?

Is there really any doubt what type of the variable br is? Does it help anyone, ever, to require another BufferedReader on the front of that line? This has bothered me for years, but it was an itch I just couldn’t scratch. Until now.

I don’t think he’s right on this one — my initial reaction is that using var everywhere demonstrates laziness by the programmer. And I don’t mean the good sort of laziness which works smarter, not harder, but the bad sort, which can’t be bothered writing proper variable declarations.

Jeff is right though – writing the type name twice does seem a little excessive. However, as Nicholas Paldino notes, the correct way to eliminate this redundancy would be to make the type name on the right hand-side optional, not the left:

While I agree with you that redundancy is not a good thing, the better way to solve this issue would have been to do something like the following:

MyObject m = new();

Or if you are passing parameters:

Person p = new("FirstName", "LastName");

Where in the creation of a new object, the compiler infers the type from the left-hand side, and not the right.

Microsoft’s C# language reference page for var also warns about the consequences of using var everywhere.

Overuse of var can make source code less readable for others. It is recommended to use var only when it is necessary, that is, when the variable will be used to store an anonymous type or a collection of anonymous types.

In the following example, the balance variable could now theoretically contain an Int32, Int64, Single (float), Double, Decimal, or even a ‘Balance’ object instance, depending on how the GetBalance() method works.

var balance = account.GetBalance();

This is rather confusing. Plus, unless you explicitly down cast, vars are always the concrete type being constructed. Let’s go back to Jeff’s BufferedReader example, of which he asks, “is there really any doubt what type of the variable br is?”

BufferedReader br = new BufferedReader (new FileReader(name));

Actually, yes there is, because polymorphism is used quite extensively in .NET’s IO libraries. The fact that br is being implemented with a BufferedReader is most likely irrelevant. All we need is something that satisfies a Reader base class contract. So br might actually look like this:

Reader br = new BufferedReader (new FileReader(name));

Just because a language allows you to do something, doesn’t mean it’s a good idea to do so. Sometimes new features adapt well to solving problems they weren’t designed for, but this is not one of those situations. Stick to using vars for what they were designed for!

Boost: How do I write a unit test for a signal?

Today, while writing some unit tests, I encountered a challenge. The user story was that, when a Person’s details are updated, the display should be updated to reflect the changes.

I’d implemented this feature using a signal on the person class that will be called whenever any details are updated:

class person
    // Set the Person's name.
    void name(const std::string & name)
        name_ = name;
    // An signal that will be called when the person's details are updated.
    boost::signal<void(const person & person)> updated;
    // The person's name.
    std::string name_;

This is a fairly standard application of an observer pattern that you might find in any MVC application.

But the question is, using the Boost unit test framework, how can I test if my signal has been called?

The mock signal handler

To test the signal handler, we’ll use a functor as a mock signal handler, that sets an internal flag when it gets called. In the functor’s destructor, we’ll do a test on the flag to make sure it got set:

struct mock_handler
    mock_handler(const person & expected_person) :
        has_been_called_(false), expected_person_(expected_person) {};
    // The signal handler function.
    void operator()(const person & person)
        has_been_called_ = true;
        BOOST_CHECK_EQUAL(&person == &expected_person_, true);
    // This handler must be called before it goes out of scope.
        BOOST_CHECK_EQUAL(has_been_called_, true);
    bool has_been_called_;
    const person & expected_person_;

The test case

Once we’ve written a mock, the test case is pretty simple. Note that I wrap my handler with a boost::ref, so that it doesn’t get copied.

// Test that setting a new name triggers the person.updated signal.
    person subject;
    mock_handler handler(subject);
    // Change the person's name, triggering the updated signal."Richard");

This works great. And if we comment out the updated signal call in person::name():

Running 1 test case... person_test.cpp(49): error in "setting_name_triggers_update_signal": check has_been_called_ == true failed [0 != 1]
*** 1 failure detected in test suite "tests"

..then the test case will fail accordingly.

Using the Boost Unit Test Framework with Xcode 3

When writing C++ code, I frequently use the Boost C++ libraries, which includes all sorts of great libraries to complement the C++ Standard Template Library (STL).

Recently I’ve been playing around with Boost’s Test library — in particular, the Unit Test Framework. Here’s a quick guide on how you can integrate it into a project in Xcode, Apple’s IDE.

You’ll need Xcode 3 and Boost 1.35 installed, and a C++ project to play with. For this example, I’m using a C++ Command Line Utility project.

Add a target for the tests executable

First, let’s create a new target for the executable that will run all our tests.

Right click on Targets, and click Add > New Target. Select Shell Tool from the list. Use “Tests” for the target’s name.

On the Build tab of the Target Info window, add Boost’s install paths to the Header and Library Search Paths. On this machine, I used MacPorts to install Boost, which uses /opt/local/include and opt/local/lib, respectively.

Right click on the Tests target, and click Add > Existing Frameworks. Browse and select libboost_unit_test_framework-mt-1_35.dylib from your library search path. Add it to your Tests target.

Now the Boost unit test framework is ready to be used from your Xcode project. Let’s use it!

Writing some test suites

My application has two classes, called a and b. They both have bugs. I’ve created a new group called Tests, and written a simple test suite for each of them.

Here’s my b_tests.cpp file. The examples I’ve used aren’t important, but the case and suite declarations are.

// Tests for the 'b' class.

	// Ensure that subtracting 6 from 8 gives 2.
	BOOST_CHECK_EQUAL(b::subtract(8, 6), 2);


First, make sure all your .cpp files are added to the Tests target. Then, to tie all the test files together into a single executable, we’ll use the test framework’s automatic main() feature in our own file called tests_main.cpp:

#define BOOST_TEST_MODULE my application tests

#include <boost/test/unit_test.hpp>

Set the active target to Tests. To check everything got wired up correctly, open the debugger window and hit Build and Go. You should see a bunch of tests fail.

Running tests as part of the build process

Our tests are set up, so let’s integrate them into the build process. Right-click on the Tests target, and click Add > New Run Script Build Phase. Paste the following into the script field (this will resolve to the tests executable):


To keep things obvious, I renamed our new Run Script phase to Run Tests.

Now let’s set up a new rule – the main target should only build after tests have been built and run successfully. Right-click on the MyApp target, click Get Info, and add Tests to the Direct Dependencies list.

Change the active target to MyApp, and hit Build. The tests should fail and return an error code, which Xcode will pick up as a script error.

If the tests don’t succeed, the build will fail, which is exactly what we want.

Parsing the test results

So far, we’ve got the tests running, and integrated into the build process, but the test results output are a bit rough. Let’s fix that.

Xcode will automatically parse script output if it’s prefixed the right way. Unfortunately, none of the Boost Unit Test Framework’s run-time parameters can produce this format. So, we’re going to have to roll up our sleeves and write our own custom unit_test_log_formatter instead.

struct xcode_log_formatter :
	public boost::unit_test::output::compiler_log_formatter
	// Produces an Xcode-friendly message prefix.
	void print_prefix(std::ostream& output,
		boost::unit_test::const_string file_name, std::size_t line)
		output << file_name << ':' << line << ": error: ";

I’ve chucked this in a file called xcode_log_formatter.hpp in the Tests group. In tests_main.cpp, we’ll use a global fixture to tell the unit test framework to use our Xcode formatter instead of the default.

// Set up the unit test framework to use an xcode-friendly log formatter.
struct xcode_config
		unit_test_log.set_formatter( new xcode_log_formatter );

	~xcode_config() {}

// Call our fixture.

After we’ve got this wired up, the test results look much better:

As you can see, instead of reporting it as a generic script error, Xcode has now recognised both test failures as two individual errors. And instead of hunting around through different files trying to find the line where a test broke, you can simply click on an error, and Xcode will find and highlight it for you.

This makes for much easier development, and allows you to manage unit test failures as easily as standard compile errors.

Is the 80 character line limit still relevant?

Traditionally, it’s always been standard practice for programmers to wrap long lines of code so they don’t span more than 80 characters across the screen.


This is because, back in the bad old days, most computer terminals could only display 25 rows of 80 columns of text on screen at once. Any lines that were longer would simply trail off out of sight. To ensure this didn’t happen, programmers split up long lines of code so none of them exceeded 80 characters.

Today, however, it’s pretty unlikely that you or anyone will still be writing code on an 80-column-width terminal. So why do we keep limiting our code to support them?

The answer, of course, is that we don’t. An 80 character limit has no relevance any more with modern computer displays. The three-year-old Powerbook I am writing this post on, for example, can easily display over 200 characters across the screen, at a comfortable 10 point font size. That’s two and a half VT100s!

The reason this standard has stuck around all these years is because of the other benefits it provides.


Long lines that span too far across the monitor are hard to read. This is typography 101. The shorter your line lengths, the less your eye has to travel to see it.

If your code is narrow enough, you can fit two files on screen, side by side, at the same time. This can be very useful if you’re comparing files, or watching your application run side-by-side with a debugger in real time.

Plus, if you write code 80 columns wide, you can relax knowing that your code will be readable and maintainable on more-or-less any computer in the world.

Another nice side effect is that snippets of narrow code are much easier to embed into documents or blog posts.


Constraining the width of your code can sometimes require you to break up lines in unnatural places, making the code look awkward and disjointed. This particular problem is worse with languages like Java and .NET, that tend to use long, descriptive identifier names.

Plus, the amount of usable space for code is also by impacted by tab width. For example, if you’re using 8-space tabs and an 80-column page width, code within a class, a method, and an if statement will already have almost a third of the available space taken for indentation.


Why 80? At work, my current project team uses a 120-character limit. We’ve all got 24″ wide-screen LCD displays, and 120 characters seems to be a good fit for our .NET/Visual Studio development environment, while still leaving ample whitespace.

There are a few factors you should think about, however. The average length of a line of code depends on what language and libraries you’re using. C generally has much shorter identifier names, and subsequently much shorter lines than, say, a .NET language.

It also depends on what sort of project you’re working on. For private and internal projects, use discretion. Find out what works best for your team, and follow it.

For open-source projects, or other situations where you don’t know who’s going to be reading your source code, tradition dictates that you stick with 80.

Another possibility is to make the limit a guideline, rather than a concrete rule. Sometimes you might not care if a particular line continues out of sight. A long string literal, for example, isn’t going to cause the end of the world if you can’t see the whole thing on screen at once.

It may sound pedantic, but if you do decide to use something different, make sure everyone knows the rule, and obeys it. When there are unclear or conflicting rules, chaos ensues. You can end up with hilarious games like formatting tennis, where every time a developer works on a piece of code, they first waste time reformatting the whole thing to reflect their own preferred coding style.

Why bother?

Some of you might wonder why anyone would worry about such trivial details like the length of a line of code. And that’s cool. But, if like me, you believe that code isn’t finished until it not only works well, but looks beautiful too, balancing style with practicality is very important.

Adding GMail’s Archive button to Microsoft Outlook 2007

Recently, I have been using GMail as my primary e-mail client. It’s simple to use, lightweight, fast and powerful. One feature I particularly like is the Archive button, which lets me clear messages out of my inbox with a single click. Instead of letting e-mails clutter up my inbox just in case I need them again, I can archive them as soon as I’ve read them. Fewer items in my inbox makes it easier to focus and prioritize tasks, which in turn makes me more productive.

In the high-pressure environment at work, where I frequently receive up to a hundred or so e-mails a day, everything I can do to make my life easier helps. Here is how I added an Archive button to Microsoft Outlook:

  1. Make a new folder in Outlook called Archive.
  2. Open the macro editor under Tools > Macros > Visual Basic Editor.
  3. Open ThisOutlookSession under Project1 in the Project Explorer window at the left side of the screen.
  4. Paste the following code:
    Option Explicit
    Public Sub ArchiveSelectedItems()
        MoveSelectedItemsToFolder "Archive"
    End Sub
    Private Sub MoveSelectedItemsToFolder(FolderName As String)
        On Error GoTo ErrorHandler
        Dim Namespace As Outlook.Namespace
        Set Namespace = Application.GetNamespace("MAPI")
        Dim Inbox As Outlook.MAPIFolder
        Set Inbox = Namespace.GetDefaultFolder(olFolderInbox)
        Dim Folder As Outlook.MAPIFolder
        Set Folder = Inbox.Folders(FolderName)
        If Folder Is Nothing Then
            MsgBox "The '" & FolderName & "' folder doesn't exist!", _
                vbOKOnly + vbExclamation, "Invalid Folder"
        End If
        Dim Item As Object
        For Each Item In Application.ActiveExplorer.Selection
            If Item.UnRead Then Item.UnRead = False
            Item.Move Folder
        Exit Sub
        MsgBox Error(Err)
    End Sub
  5. Save and close the editor.

Now the archive macro has been created. Let’s add a button that will trigger it:

  1. Click on View > Toolbars > Customize, and select the Commands tab.
  2. Scroll down to the Macros category.
  3. Drag the Project1.ThisOutlookSession.ArchiveSelected command to a toolbar.
  4. Right-click on your new button, and rename it to Archive Selected. Set it to display Text Only.
  5. Click Close.

All done!