ploeh blog danish software design

Watch out for this code smell with some unit testing frameworks.

In a previous article you saw this state-based integration test:

[Theory]
[InlineData(1049, 19, 00, "juliad@example.net", "Julia Domna", 5)]
[InlineData(1130, 18, 15, "x@example.com", "Xenia Ng", 9)]
[InlineData( 956, 16, 55, "kite@example.edu", null, 2)]
[InlineData( 433, 17, 30, "shli@example.org", "Shanghai Li", 5)]
public async Task PostValidReservationWhenDatabaseIsEmpty(
    int days,
    int hours,
    int minutes,
    string email,
    string name,
    int quantity)
{
    var at = DateTime.Now.Date + new TimeSpan(days, hours, minutes, 0);
    var db = new FakeDatabase();
    var sut = new ReservationsController(
        new SystemClock(),
        new InMemoryRestaurantDatabase(Grandfather.Restaurant),
        db);
 
    var dto = new ReservationDto
    {
        Id = "B50DF5B1-F484-4D99-88F9-1915087AF568",
        At = at.ToString("O"),
        Email = email,
        Name = name,
        Quantity = quantity
    };
    await sut.Post(dto);
 
    var expected = new Reservation(
        Guid.Parse(dto.Id),
        DateTime.Parse(dto.At, CultureInfo.InvariantCulture),
        new Email(dto.Email),
        new Name(dto.Name ?? ""),
        dto.Quantity);
    Assert.Contains(expected, db.Grandfather);
}

This was the test after I improved it. Still, I wasn't satisfied with it. It has several problems. Take a few moments to consider it. Can you identify any problems? Which ones?

The code shown here is part of the sample code base that accompanies my book Code That Fits in Your Head.

Size #

I know that you're not familiar with all the moving parts. You don't know how ReservationDto or Reservation are implemented. You don't know what InMemoryRestaurantDatabase is, or how ReservationsController behaves. Still, the issues I have in mind aren't specific to a particular code base.

I feel that the method is verging on being too big. Quantifiably, it doesn't fit in an 80x24 box, but that's just an arbitrary threshold anyway. Still, I think it's grown to a size that makes me uncomfortable.

If you aren't convinced, think of this code example as a stand-in for something larger. In the above test, a reservation contains five smaller values (Id, At, Email, Name, and Quantity). How would a similar test look if the object in question contains ten or twenty values?

In the decades I've been programming and consulting, I've seen plenty of code bases. Data objects made from twenty fields are hardly unusual.

What would a similar test look like if the dto and the expected object required twenty smaller values?

The test would be too big.

Primitive obsession #

A test like this one contains a mix of essential behaviour and implementation details. The behaviour that it verifies is that when you Post a valid dto, the data makes it all the way to the database.

Exactly how the dto or the expected value are constructed is less relevant for the test. Yet it's intermingled with the test of behaviour. The signal-to-noise ratio in the test isn't all that great. What can you do to improve things?

As given, it seems difficult to do much. The problem is primitive obsession. While this is a parametrised test, all the method parameters are primitives: integers and strings. The makes it hard to introduce useful abstractions.

In C# (and probably other languages as well) parametrised tests often suffer from primitive obsession. The most common data source is an attribute (AKA annotation), like xUnit.net's [InlineData] attribute. This isn't a limitation of xUnit.net, but rather of .NET attributes. You can only create attributes with primitive values and arrays.

What is a limitation of xUnit.net (and the other mainstream .NET testing frameworks, as far as I know) is that tests aren't first-class values. In Haskell, by contrast, it's easy to write parametrised tests using the normal language constructs exactly because tests are first-class values. (I hope that the next version of xUnit.net will support tests as first-class values.)

Imagine that instead of only five constituent fields, you'd have to write a parametrised test for objects with twenty primitive values. As long as you stick with attribute-based data sources, you'll be stuck with primitive values.

Granted, attributes like [InlineData] are lightweight, but over the years, my patience with them has grown shorter. They lock me into primitive obsession, and I don't appreciate that.

Essential test #

While tests as first-class values aren't an option in xUnit.net, you can provide other data sources for the [Theory] attribute than [InlineData]. It's not as lightweight, but it breaks the primitive obsession and re-enables normal code design techniques. It enables you to reduce the test itself to its essence. You no longer have to think in primitives, but can instead express the test unshackled by constraints. As a first pass, I'd like the test to look like this:

[Theory]
[ClassData(typeof(PostValidReservationWhenDatabaseIsEmptyTestCases))]
public async Task PostValidReservationWhenDatabaseIsEmpty(
    ReservationDto validDto, Reservation expected)
{
    var db = new FakeDatabase();
    var sut = new ReservationsController(
        new SystemClock(),
        new InMemoryRestaurantDatabase(Grandfather.Restaurant),
        db);
 
    await sut.Post(validDto);
 
    Assert.Contains(expected, db.Grandfather);
}

This version of the test eliminates the noise. How validDto and expected are constructed is an implementation detail that has little bearing on the behaviour being tested.

For a reader of the code, it's should now be clearer what's at stake here: If you Post a validDto the expected reservation should appear in the database.

Reducing the test code to its essentials made me realise something that hitherto had escaped me: that I could name the DTO by role. Instead of just dto, I could call the parameter validDto.

Granted, I could also have done that previously, but I didn't think of it. There's was so much noise in that test that I didn't stop to consider whether dto sufficiently communicated the role of that variable.

The less code, the easier it becomes to think such things through, I find.

In any case, the test code now much more succinctly expresses the essence of the desired behaviour. Notice how I started my refactoring by writing the desired test code. I've yet to implement the data source. Now that the data source expresses test data as full objects, I'm not so concerned with whether or not that's going to be possible. Of course it's possible.

Object data source #

You can define data sources for xUnit.net as classes or methods. In C# I usually reach for the [ClassData] option, since an object (in C#, that is) gives me better options for further decomposition. For example, I can define a class and delegate the details to helper methods:

private class PostValidReservationWhenDatabaseIsEmptyTestCases :
    TheoryData<ReservationDto, Reservation>
{
    public PostValidReservationWhenDatabaseIsEmptyTestCases()
    {
        AddWithName(1049, 19, 00, "juliad@example.net", "Julia Domna", 5);
        AddWithName(1130, 18, 15, "x@example.com", "Xenia Ng", 9);
        AddWithoutName(956, 16, 55, "kite@example.edu", 2);
        AddWithName(433, 17, 30, "shli@example.org", "Shanghai Li", 5);
    }
 
    // More members here...

Here, I'm taking advantage of xUnit.net's built-in TheoryData<T1, T2> base class, but that's just a convenience. All you have to do is to implement IEnumerable<object[]>.

As you can see, the constructor adds the four test cases by calling two private helper methods. Here's the first of those:

private const string id = "B50DF5B1-F484-4D99-88F9-1915087AF568";
 
private void AddWithName(
    int days,
    int hours,
    int minutes,
    string email,
    string name,
    int quantity)
{
    var at =
        DateTime.Now.Date + new TimeSpan(days, hours, minutes, 0);
    Add(new ReservationDto
        {
            Id = id,
            At = at.ToString("O"),
            Email = email,
            Name = name,
            Quantity = quantity
        },
        new Reservation(
            Guid.Parse(id),
            at,
            new Email(email),
            new Name(name),
            quantity));
}

The other helper method is almost identical, although it has a slight variation when it comes to the reservation name:

private void AddWithoutName(
    int days,
    int hours,
    int minutes,
    string email,
    int quantity)
{
    var at =
        DateTime.Now.Date + new TimeSpan(days, hours, minutes, 0);
    Add(new ReservationDto
        {
            Id = id,
            At = at.ToString("O"),
            Email = email,
            Quantity = quantity
        },
        new Reservation(
            Guid.Parse(id),
            at,
            new Email(email),
            new Name(""),
            quantity));
}

In total, this refactoring results in more code, so how is this an improvement?

The paradox of decomposition #

In object-oriented design, decomposition tends to lead to more code. If you want to isolate and make reusable a particular piece of behaviour, you'll usually introduce an interface or a base class. Even stateless functions need a static class to define them. (To be fair, functional programming isn't entirely devoid of such overhead associated with decomposition, but the cost tends to smaller.) This leads to more code, compared with the situation before decomposition.

This is a situation you may also encounter if you attempt to refactor to design patterns, or follow the SOLID principles. You'll have more code than when you started. This often leads to resistance to such 'code bloat'.

It's fine to resist code bloat. It's also fine to dislike 'complexity for complexity's sake'. Try to evaluate each potential code change based on advantages and disadvantages. I'm not insisting that the above refactoring is objectively better. I did feel, however, that I had a problem that I ought to address, and that this was a viable alternative. The result is more code, but each piece of code is smaller and simpler.

You can, conceivably, read the test method itself to get a feel for what it tests, even if you don't know all the implementation details. You can read the four statements in the PostValidReservationWhenDatabaseIsEmptyTestCases constructor without, I hope, understanding all the details about the two helper methods. And you can read AddWithName without understanding how AddWithoutName works, and vice versa, because these two methods don't depend on each other.

Conclusion #

In this article, I've described how the use of code annotations for parametrised tests tend to pull in the direction of primitive obsession. This is a force worth keeping an eye on, I think.

You saw how to refactor to class-based test data generation. This enables you to use objects instead of primitives, thus opening your design palette. You can now use all your usual object-oriented or functional design skills to factor the code in a way that's satisfactory.

Was it worth it in this case? Keep in mind that the original problem was already marginal. While the code didn't fit in a 80x24 box, it was only 33 lines of code (excluding the test data). Imagine, however, that instead of a five-field reservation, you'd be dealing with a twenty-field data class, and such a refactoring begins to look more compelling.

Is the code now perfect? It still isn't. I'm a little put off by the similarity of AddWithName and AddWithoutName. I'm also aware that there's a trace of production code duplicated in the test case, in the way that the test code duplicates how a valid ReservationDto relates to a Reservation. I'm on the fence whether I should do anything about this.

At the moment I'm inclined to heed the rule of three. The duplication is still too insubstantial to warrant refactoring, but it's worth keeping an eye on.

Comments

A typical future test maintenance problem.

In a recent article I showed a unit test and parenthetically mentioned that it might have a future maintenance problem. Here's a more recent version of the same test. Can you tell what the future issue might be?

[Theory]
[InlineData("2023-11-24 19:00", "juliad@example.net", "Julia Domna", 5)]
[InlineData("2024-02-13 18:15", "x@example.com", "Xenia Ng", 9)]
[InlineData("2023-08-23 16:55", "kite@example.edu", null, 2)]
[InlineData("2022-03-18 17:30", "shli@example.org", "Shanghai Li", 5)]
public async Task PostValidReservationWhenDatabaseIsEmpty(
    string at, string email, string name, int quantity)
{
    var db = new FakeDatabase();
    var sut = new ReservationsController(
        new SystemClock(),
        new InMemoryRestaurantDatabase(Grandfather.Restaurant),
        db);
 
    var dto = new ReservationDto
    {
        Id = "B50DF5B1-F484-4D99-88F9-1915087AF568",
        At = at,
        Email = email,
        Name = name,
        Quantity = quantity
    };
    await sut.Post(dto);
 
    var expected = new Reservation(
        Guid.Parse(dto.Id),
        DateTime.Parse(dto.At, CultureInfo.InvariantCulture),
        new Email(dto.Email),
        new Name(dto.Name ?? ""),
        dto.Quantity);
    Assert.Contains(expected, db.Grandfather);
}

To be honest, there's more than one problem with this test, but presently I'm going to focus on one of them.

Since you don't know the details of the implementation, you may not be able to tell what the problem might be. It's not a trick question. On the other hand, you might still be able to guess, just from the clues available in the above code listing.

The code shown here is part of the sample code base that accompanies my book Code That Fits in Your Head.

Sooner or later #

Here are some clues to consider: I'm writing this article in the beginning of 2021. Consider the dates supplied via the [InlineData] attributes. Seen from 2021, they're all in the future.

Notice, as well, that the sut takes a SystemClock dependency. You don't know the SystemClock class (it's a proprietary class in this code base), but from the name I'm sure that you can imagine what it represents.

From the perspective of early 2021, all dates are going to be in the future for more than a year. What is going to happen, though, once the test runs after March 18, 2022?

That test case is going to fail.

You can't tell from the above code listing, but the system under test rejects reservations in the past. Once March 18, 2022 has come and gone, the reservation at "2022-03-18 17:30" is going to be in the past. The sut will reject the reservation, and the assertion will fail.

You have to be careful with tests that rely on the system clock.

Test Double? #

The fundamental problem is that the system clock is non-deterministic. A typical reaction to non-determinism in unit testing is to introduce a Test Double of some sort. Instead of using the system clock, you could use a Stub as a stand-in for the real time.

This is possible here as well. The ReservationsController class actually depends on an IClock interface that SystemClock implements. You could define a test-specific ConstantClock implementation that would always return a constant date and time. This would actually work, but would rely on an implementation detail.

At the moment, the ReservationsController only calls Clock.GetCurrentDateTime() a single time to get the current time. As soon as it has that value, it passes it to a pure function, which implements the business logic:

var now = Clock.GetCurrentDateTime();
if (!restaurant.MaitreD.WillAccept(now, reservations, reservation))
    return NoTables500InternalServerError();

A ConstantClock would work, but only as long as the ReservationsController only calls Clock.GetCurrentDateTime() once. If it ever began to call this method multiple times to detect the passing of time, using a constant time value would mostly likely again break the test. This seems brittle, so I don't want to go that way.

Relative time #

Working with the system clock in automated tests is easier if you deal with relative time. Instead of defining the test cases as absolute dates, express them as days into the future. Here's one way to refactor the test:

[Theory]
[InlineData(1049, 19, 00, "juliad@example.net", "Julia Domna", 5)]
[InlineData(1130, 18, 15, "x@example.com", "Xenia Ng", 9)]
[InlineData( 956, 16, 55, "kite@example.edu", null, 2)]
[InlineData( 433, 17, 30, "shli@example.org", "Shanghai Li", 5)]
public async Task PostValidReservationWhenDatabaseIsEmpty(
    int days,
    int hours,
    int minutes,
    string email,
    string name,
    int quantity)
{
    var at = DateTime.Now.Date + new TimeSpan(days, hours, minutes, 0);
    var db = new FakeDatabase();
    var sut = new ReservationsController(
        new SystemClock(),
        new InMemoryRestaurantDatabase(Grandfather.Restaurant),
        db);
 
    var dto = new ReservationDto
    {
        Id = "B50DF5B1-F484-4D99-88F9-1915087AF568",
        At = at.ToString("O"),
        Email = email,
        Name = name,
        Quantity = quantity
    };
    await sut.Post(dto);
 
    var expected = new Reservation(
        Guid.Parse(dto.Id),
        DateTime.Parse(dto.At, CultureInfo.InvariantCulture),
        new Email(dto.Email),
        new Name(dto.Name ?? ""),
        dto.Quantity);
    Assert.Contains(expected, db.Grandfather);
}

The absolute dates always were fairly arbitrary, so I just took the current date and converted the dates to a number of days into the future. Now, the first test case will always be a date 1,049 days (not quite three years) into the future, instead of November 24, 2023.

The test is no longer a failure waiting to happen.

Conclusion #

Treating test cases that involve time and date as relative to the current time, instead of as absolute values, is usually a good idea if the system under test depends on the system clock.

It's always a good idea to factor as much code as you can as pure functions, like the above WillAccept method. Pure functions don't depend on the system clock, so here you can safely pass absolute time and date values. Pure functions are intrinsically testable.

Still, as the test pyramid suggests, relying exclusively on unit tests isn't a good idea. The test shown in this article isn't really a unit test, but rather a state-based integration test. It relies on both the system clock and a Fake database. Expressing the test cases for this test as relative time values effectively addresses the problem introduced by the system clock.

There are plenty of other problems with the above test. One thing that bothers me is that the 'fix' made the line count grow. It didn't quite fit into a 80x24 box before, but now it's even worse! I should do something about that, but that's a topic for another article.

A proof of concept of cross-branch testing for compiled languages.

Hillel Wayne recently published an article called Cross-Branch Testing. It outlines an approach to a class of problems that are hard to test. He mentions computer vision and simulations, among others. I can add that it's also difficult to write intuitive tests of convex hulls and Conway's game of life.

Hillel Wayne calls these rho problems, 'just because'. I'm totally going to run with that term.

In the article, he outlines an approach where you test an iteration of rho code against a 'last known good' snapshot. He uses git worktree to set up a snapshot of the reference implementation. He then writes a property that compares the refactored code's behaviour against the reference.

The example code is in Python, which is a language that I don't know. As far as I can tell, it works because Python is 'lightweight' enough that you can load and execute source code directly. I found that the approach makes much sense, but I wondered how it would apply for statically typed, compiled languages. I decided to create a proof of concept in F#.

Test cases from Python #

My first problem was to port Hillel Wayne's example rho problem to F#. The function f doesn't have any immediate mathematical properties; nor is its behaviour intuitive. While I think that I understand what each line of code in f means, I don't really know Python. Since one of the properties of rho problems is that bugs can be subtle, I didn't trust myself to be able to port the Python code to F# without some test cases.

To solve that problem, I first found an online Python interpreter and pasted the f function into it. I then wrote code to print the output of a function call:

print(f'1, 2, 3, { f(1, 2, 3) }')

This line of code produces this output:

1, 2, 3, True

In other words, I could produce comma-separated values of input and actual output.

Hillel Wayne wrote properties using Hypothesis, which, it seems, by default runs each property 200 times.

In F# I'm going to use FsCheck, so I first used F# Interactive with FsCheck to produce 200 Python print statements like the above:

> Arb.Default.Int32().Generator
|> Gen.three
|> Gen.map (fun (x, y, z) -> sprintf "print(f'%i, %i, %i, { f(%i, %i, %i) }')" x y z x y z)
|> Gen.sample 100 200
|> List.iter (printfn "%s");;
print(f'-77, 67, 84, { f(-77, 67, 84) }')
print(f'58, -46, 3, { f(58, -46, 3) }')
print(f'21, 13, 94, { f(21, 13, 94) }')
...

This is a throwaway data pipeline that starts with an FsCheck integer generator, creates a triple from it, turns that triple into a Python print statement, and finally writes 200 of those to the console. The above code listing only shows the first three lines of output, while the rest are indicated by an ellipsis.

I copied those 200 print statements over to the online Python interpreter and ran the code. That produced 200 comma-separated values like these:

-77, 67, 84, False
58, -46, 3, False
21, 13, 94, True
...

These can serve as test cases for porting the Python code to F#.

Port to F# #

The next step is to write a parametrised test, using a provisional implementation of f:

[<Theory; MemberData(nameof fTestCases)>]
let ``test f`` x y z expected = expected =! f x y z

This test uses xUnit.net 2.4.1 and Unquote 5.0.0. As you can tell, apart from the annotations, it's a true one-liner. It calls the f function with the three supplied arguments x, y, and z and compares the return value with the expected value.

The code uses the new nameof feature of F# 5. fTestCases is a function in the same module that holds the test:

// unit -> seq<obj []>
let fTestCases () =
    use strm = typeof<Anchor>.Assembly.GetManifestResourceStream streamName
    use rdr = new StreamReader (strm)
    let s = rdr.ReadToEnd ()
    s.Split Environment.NewLine |> Seq.map csvToTestCase

It reads an embedded resource stream of test cases, like the above comma-separated values. Even though the values are in a text file, it's easier to embed the file in the test assembly, because it nicely dispenses with the problem of copying a text file to the appropriate output directory when the code compiles. That would, however, be an valid alternative.

Anchor is a dummy type to support typeof, and streamName is just a string constant that identifies the name of the stream.

The csvToTestCase function converts each line of comma-separated values to test cases for the [<Theory>] attribute:

// string -> obj []
let csvToTestCase (csv : string) =
    let values = csv.Split ','
    [|
        values.[0] |> Convert.ToInt32 |> box
        values.[1] |> Convert.ToInt32 |> box
        values.[2] |> Convert.ToInt32 |> box
        values.[3] |> Convert.ToBoolean |> box
    |]

It's not the safest code I could write, but this is, after all, a proof of concept.

The most direct port of the Python code I could produce is this:

// f : int -> int -> int -> bool
let f (x : int) (y : int) (z : int) =
    let mutable mx = bigint x
    let mutable my = bigint y
    let mutable mz = bigint z
    let mutable out = 0I
    for i in [0I..9I] do
        out <- out * mx + abs (my * mz - i * i)
        let x' = mx
        let y' = my
        let z' = mz
        mx <- y' + 1I
        my <- z'
        mz <- x'
    abs out % 100I < 10I

As F# code goes, it's disagreeable, but it passes all 200 test cases, so this will serve as an initial implementation. The out variable can grow to values that overflow even 64-bit integers, so I had to convert to bigint to get all test cases to pass.

If I make the same mutation to the code that Hillel Wayne did (abs out % 100I < 9I) two test cases fail. This gives me some confidence that I have a degree of problem coverage comparable to his.

Test oracle #

Now that a reference implementation exists, we can use it as a test oracle for refactorings. You can, for example, add a little test-only utility to your program portfolio:

open Prod
open FsCheck
 
[<EntryPoint>]
let main argv =
    Arb.Default.Int32().Generator
    |> Gen.three
    |> Gen.sample 100 200
    |> List.iter (fun (x, y, z) -> printfn "%i, %i, %i, %b" x y z (f x y z))
 
    0 // return an integer exit code

Notice that the last step in the pipeline is to output the values of each x, y, and z, as well as the result of calling f x y z.

This is a command-line executable that uses FsCheck to produce new test cases by calling the f function. It looks similar to the above one-off script that produced Python code, but this one instead just produces comma-separated values. You can run it from the command line to produce a new sample of test cases:

$ ./foracle
29, -48, -78, false
-8, -25, 13, false
-74, 34, -68, true
...

As above, I've used an ellipsis to indicate that in reality, 200 lines of comma-separated values scroll by.

When you use Bash, you can even pipe the output straight to a file:

$ ./foracle > csv.txt

You can now take the new comma-separated values and update the test values that the above test f test uses.

In other words, you use version n of f as a test oracle for version n + 1. When iteration n + 1 is a function of iteration n, you have a so-called dynamic system, so I think that we can call this technique dynamic test oracles.

The above foracle program is just a proof of concept. You could make it more flexible by making it take command-line arguments that would let you control the sample size and FsCheck's size parameter (the hard-coded 100 in the above code listing).

Refactoring #

With the confidence instilled by the test cases, we can now refactor the f function:

// f : int -> int -> int -> bool
let f (x : int) (y : int) (z : int) =
    let imp (x, y, z, out) i =
        let out = out * x + abs (y * z - i * i)
        y + 1I, z, x, out
    let (_, _, _, out) = List.fold imp (bigint x, bigint y, bigint z, 0I) [0I..9I]
    abs out % 100I < 10I

Instead of all those mutable variables, the function is, after all, just a left fold. Phew, I feel better now.

Conclusion #

This article demonstrated a proof of concept where you use a known good version (n) of the code as a test oracle for the next version (n + 1). In interpreted languages, you may be able to load two versions of the code base side by side, but that's rarely practical in a statically typed compiled language like F#. Instead, I used a utility program to generate test cases that can be used as a data source for a parametrised test.

The example rho problem takes only integers as input, and returns a simple Boolean value, so in this case it's trivial to encode each test case as a line of comma-separated values. For (real) problems that may involve more complex types, it'd be better to use another serialisation format, such as JSON or XML.

An outstanding issue is whether it's possible to implement shrinking behaviour when tests fail. Currently, the proof of concept just uses a set of serialised test cases. Normally, when a property-based testing framework like FsCheck detects a counter-example, it'll shrink the counter-example to values that are easier to understand than the original. This proof of concept doesn't do that. I'm not sure if a framework like FsCheck currently contains enough extensibility points to enable that sort of behaviour. I'll leave that question open for any reader interested in taking on that problem.

Comments

An F# port of the previous Haskell proof of concept.

This article is part of a short article series on applicative validation with a twist. The twist is that validation, when it fails, should return not only a list of error messages; it should also retain that part of the input that was valid.

In the previous article you saw a Haskell proof of concept that demonstrated how to compose the appropriate applicative functor with a suitable semigroup to make validation work as desired. In this article, you'll see how to port that proof of concept to F#.

Data definitions #

Like in the previous article, we're going to need some types. These are essentially direct translations of the corresponding Haskell types:

type Input = { Name : string option; DoB : DateTime option; Address : string option}
type ValidInput = { Name : string; DoB : DateTime; Address : string }

The Input type plays the role of the input we'd like to validate, while ValidInput presents validated data.

If you're an F# fan, you can bask in the reality that F# records are terser than Haskell records. I like both languages, so I have mixed feelings about this.

Computation expression #

Haskell's main workhorse is its type class system. F# doesn't have that, but it has computation expressions, which in F# 5 got support for applicative functors. That's just what we need, and it turns out that there isn't a lot of code we have to write to make all of this work.

To recap from the Haskell proof of concept: We need a Result-like container that returns a tuple for errors. One element of the tuple should be a an endomorphism, which forms a monoid (and therefore also a semigroup). The other element should be a list of error messages - another monoid. In F# terms we'll write it as (('b -> 'b) * 'c list).

That's a tuple, and since tuples form monoids when their elements do the Error part of Result supports accumulation.

To support an applicative computation expression, we're going to need a a way to merge two results together. This is by far the most complicated piece of code in this article, all six lines of code:

module Result =
    // Result<'a       ,(('b -> 'b) * 'c list)> ->
    // Result<'d       ,(('b -> 'b) * 'c list)> ->
    // Result<('a * 'd),(('b -> 'b) * 'c list)>
    let merge x y =
        match x, y with
        | Ok xres, Ok yres -> Ok (xres, yres)
        | Error (f, e1s), Error (g, e2s)  -> Error (f >> g, e2s @ e1s)
        | Error e, Ok _ -> Error e
        | Ok _, Error e -> Error e

The merge function composes two input results together. The results have Ok types called 'a and 'd, and if they're both Ok values, the return value is an Ok tuple of 'a and 'd.

If one of the results is an Error value, it beats an Ok value. The only moderately complex operations is when both are Error values.

Keep in mind that an Error value in this instance contains a tuple of the type (('b -> 'b) * 'c list). The first element is an endomorphism 'b -> 'b and the other element is a list. The merge function composes the endomorphism f and g by standard function composition (the >> operator), and concatenates the lists with the standard @ list concatenation operator.

Because I'm emulating how the original forum post's code behaves, I'm concatenating the two lists with the rightmost going before the leftmost. It doesn't make any other difference than determining the order of the error list.

With the merge function in place, the computation expression is a simple matter:

type ValidationBuilder () =
    member _.BindReturn (x, f) = Result.map f x
    member _.MergeSources (x, y) = Result.merge x y

The last piece is a ValidationBuilder value:

[<AutoOpen>]
module ComputationExpressions =
    let validation = ValidationBuilder ()

Now, whenever you use the validation computation expression, you get the desired functionality.

Validators #

Before we can compose some validation functions, we'll need to have some validators in place. These are straightforward translations of the Haskell validation functions, starting with the name validator:

// Input -> Result<string,((Input -> Input) * string list)>
let validateName ({ Name = name } : Input) =
    match name with
    | Some n when n.Length > 3 -> Ok n
    | Some _ ->
        Error (
            (fun (args : Input) -> { args with Name = None }),
            ["no bob and toms allowed"])
    | None -> Error (id, ["name is required"])

When the name is too short, the endomorphism resets the Name field to None.

The date-of-birth validation function works the same way:

// DateTime -> Input -> Result<DateTime,((Input -> Input) * string list)>
let validateDoB (now : DateTime) ({ DoB = dob } : Input) =
    match dob with
    | Some d when d > now.AddYears -12 -> Ok d
    | Some _ ->
        Error (
            (fun (args : Input) -> { args with DoB = None }),
            ["get off my lawn"])
    | None -> Error (id, ["dob is required"])

Again, like in the Haskell proof of concept, instead of calling DateTime.Now from within the function, I'm passing now as an argument to keep the function pure.

The address validation concludes the set of validators:

// Input -> Result<string,(('a -> 'a) * string list)>
let validateAddress ({ Address = address }: Input) =
    match address with
    | Some a -> Ok a
    | None -> Error (id, ["add1 is required"])

The inferred endomorphism type here is the more general 'a -> 'a, but it's compatible with Input -> Input.

Composition #

All three functions have compatible Error types, so they ought to compose with the applicative computation expression to produce the desired behaviour:

// DateTime -> Input -> Result<ValidInput,(Input * string list)>
let validateInput now args =
    validation {
        let! name = validateName args
        and! dob = validateDoB now args
        and! address = validateAddress args
        return { Name = name; DoB = dob; Address = address }
    }
    |> Result.mapError (fun (f, msgs) -> f args, msgs)

The validation expression alone produces a Result<ValidInput,((Input -> Input) * string list)> value. To get an Input value in the Error tuple, we need to 'run' the Input -> Input endomorphism. The validateInput function does that by applying the endomorphism f to args when mapping the error with Result.mapError.

Tests #

To test that the validateInput works as intended, I first copied all the code from the original forum post. I then wrote eight characterisation tests against that code to make sure that I could reproduce the desired functionality.

I then wrote a parametrised test against the new function:

[<Theory; ClassData(typeof<ValidationTestCases>)>]
let ``Validation works`` input expected =
    let now = DateTime.Now
    let actual = validateInput now input
    expected =! actual

The ValidationTestCases class is defined like this:

type ValidationTestCases () as this =
    inherit TheoryData<Input, Result<ValidInput, Input * string list>> ()

This class produces a set of test cases, where each test case contains an input value and the expected output. To define the test cases, I copied the eight characterisation tests I'd already produced and adjusted them so that they fit the simpler API of the validateInput function. Here's a few examples:

let eightYearsAgo = DateTime.Now.AddYears -8
do this.Add (
    { Name = Some "Alice"; DoB = Some eightYearsAgo; Address = None },
    Error (
        { Name = Some "Alice"; DoB = Some eightYearsAgo; Address = None },
        ["add1 is required"]))
 
do this.Add (
    { Name = Some "Alice"; DoB = Some eightYearsAgo; Address = Some "x" },
    Ok ({ Name = "Alice"; DoB = eightYearsAgo; Address = "x" }))

The first case expects an Error value because the Input value has no address. The other test case expects an Ok value because all input is fine.

I copied all eight characterisation tests over, so now I have those eight tests, as well as the modified eight tests for the applicative-based API shown here. All sixteen tests pass.

Conclusion #

I find this solution to the problem elegant. It's always satisfying when you can implement what at first glance looks like custom behaviour using universal abstractions.

Besides the aesthetic value, I also believe that this keeps a team more productive. These concepts of monoids, semigroups, applicative functors, and so on, are concepts that you only have to learn once. Once you know them, you'll recognise them when you run into them. This means that there's less code to understand.

An ad-hoc implementation as the original forum post suggested (even though it looked quite decent) always puts the onus on a maintenance developer to read and understand even more one-off infrastructure code.

With an architecture based on universal abstractions and well-documented language features, a functional programmer that knows these things will be able to pick up what's going on without much trouble. Specifically, (s)he will recognise that this is 'just' applicative validation with a twist.

This article is the December 28 entry in the F# Advent Calendar in English 2020.

Next: A C# port of validation with partial round trip.

Which Semigroup best addresses the twist in the previous article?

This article is part of a short article series on applicative validation with a twist. The twist is that validation, when it fails, should return not only a list of error messages; it should also retain that part of the input that was valid.

In this article, I'll show how I did a quick proof of concept in Haskell.

Data definitions #

You can't use the regular Either instance of Applicative for validation because it short-circuits on the first error. In other words, you can't collect multiple error messages, even if the input has multiple issues. Instead, you need a custom Applicative instance. You can easily write such an instance yourself, but there are a couple of libraries that already do this. For this prototype, I chose the validation package.

import Data.Bifunctor
import Data.Time
import Data.Semigroup
import Data.Validation

Apart from importing Data.Validation, I also need a few other imports for the proof of concept. All of them are well-known. I used no language extensions.

For the proof of concept, the input is a triple of a name, a date of birth, and an address:

data Input = Input {
  inputName :: Maybe String,
  inputDoB :: Maybe Day,
  inputAddress :: Maybe String }
  deriving (Eq, Show)

The goal is actually to parse (not validate) Input into a safer data type:

data ValidInput = ValidInput {
  validName :: String,
  validDoB :: Day,
  validAddress :: String }
  deriving (Eq, Show)

If parsing/validation fails, the output should report a collection of error messages and return the Input value with any valid data retained.

Looking for a Semigroup #

My hypothesis was that validation, even with that twist, can be implemented elegantly with an Applicative instance. The validation package defines its Validation data type such that it's an Applicative instance as long as its error type is a Semigroup instance:

Semigroup err => Applicative (Validation err)

The question is: which Semigroup can we use?

Since we need to return both a list of error messages and a modified Input value, it sounds like we'll need a product type of some sorts. A tuple will do; something like (Input, [String]). Is that a Semigroup instance, though?

Tuples only form semigroups if both elements give rise to a semigroup:

(Semigroup a, Semigroup b) => Semigroup (a, b)

The second element of my candidate is [String], which is fine. Lists are Semigroup instances. But what about Input? Can we somehow combine two Input values into one? It's not entirely clear how we should do that, so that doesn't seem too promising.

What we need to do, however, is to take the original Input and modify it by (optionally) resetting one or more fields. In other words, a series of functions of the type Input -> Input. Aha! There's the semigroup we need: Endo Input.

So the Semigroup instance we need is (Endo Input, [String]), and the validation output should be of the type Validation (Endo Input, [String]) a.

Validators #

Cool, we can now implement the validation logic; a function for each field, starting with the name:

validateName :: Input -> Validation (Endo Input, [String]) String
validateName (Input (Just name) _ _) | length name > 3 = Success name
validateName (Input (Just _) _ _) =
  Failure (Endo $ \x -> x { inputName = Nothing }, ["no bob and toms allowed"])
validateName _ = Failure (mempty, ["name is required"])

This function reproduces the validation logic implied by the forum question that started it all. Notice, particularly, that when the name is too short, the endomorphism resets inputName to Nothing.

The date-of-birth validation function works the same way:

validateDoB :: Day -> Input -> Validation (Endo Input, [String]) Day
validateDoB now (Input _ (Just dob) _) | addGregorianYearsRollOver (-12) now < dob =
  Success dob
validateDoB _ (Input _ (Just _) _) =
  Failure (Endo $ \x -> x { inputDoB = Nothing }, ["get off my lawn"])
validateDoB _ _ = Failure (mempty, ["dob is required"])

Again, the validation logic is inferred from the forum question, although I found it better keep the function pure by requiring a now argument.

The address validation is the simplest of the three validators:

validateAddress :: Monoid a => Input -> Validation (a, [String]) String
validateAddress (Input _ _ (Just a)) = Success a
validateAddress _ = Failure (mempty, ["add1 is required"])

This one's return type is actually more general than required, since I used mempty instead of Endo id. This means that it actually works for any Monoid a, which also includes Endo Input.

Composition #

All three functions return Validation (Endo Input, [String]), which has an Applicative instance. This means that we should be able to compose them together to get the behaviour we're looking for:

validateInput :: Day -> Input -> Either (Input, [String]) ValidInput
validateInput now args =
  toEither $
  first (first (`appEndo` args)) $
  ValidInput <$> validateName args <*> validateDoB now args <*> validateAddress args

That compiles, so it probably works.

Sanity check #

Still, it'd be prudent to check. Since this is only a proof of concept, I'm not going to set up a test suite. Instead, I'll just start GHCi for some ad-hoc testing:

λ> now <- localDay <&> zonedTimeToLocalTime <&> getZonedTime
λ> validateInput now & Input Nothing Nothing Nothing
Left (Input {inputName = Nothing, inputDoB = Nothing, inputAddress = Nothing},
      ["name is required","dob is required","add1 is required"])
λ> validateInput now & Input (Just "Bob") Nothing Nothing
Left (Input {inputName = Nothing, inputDoB = Nothing, inputAddress = Nothing},
      ["no bob and toms allowed","dob is required","add1 is required"])
λ> validateInput now & Input (Just "Alice") Nothing Nothing
Left (Input {inputName = Just "Alice", inputDoB = Nothing, inputAddress = Nothing},
      ["dob is required","add1 is required"])
λ> validateInput now & Input (Just "Alice") (Just & fromGregorian 2002 10 12) Nothing
Left (Input {inputName = Just "Alice", inputDoB = Nothing, inputAddress = Nothing},
      ["get off my lawn","add1 is required"])
λ> validateInput now & Input (Just "Alice") (Just & fromGregorian 2012 4 21) Nothing
Left (Input {inputName = Just "Alice", inputDoB = Just 2012-04-21, inputAddress = Nothing},
      ["add1 is required"])
λ> validateInput now & Input (Just "Alice") (Just & fromGregorian 2012 4 21) (Just "x")
Right (ValidInput {validName = "Alice", validDoB = 2012-04-21, validAddress = "x"})

In order to make the output more readable, I've manually edited the GHCi session by adding line breaks to the output.

It looks like it's working like it's supposed to. Only the last line successfully parses the input and returns a Right value.

Conclusion #

Before I started this proof of concept, I had an inkling of the way this would go. Instead of making the prototype in F#, I found it more productive to do it in Haskell, since Haskell enables me to compose things together. I particularly appreciate how a composition of types like (Endo Input, [String]) is automatically a Semigroup instance. I don't have to do anything. That makes the language great for prototyping things like this.

Now that I've found the appropriate semigroup, I know how to convert the code to F#. That's in the next article.

Next: An F# demo of validation with partial data round trip.

Comments

newtype Perhaps a = Perhaps { runPerhaps :: Maybe  a } deriving (Eq, Show)
 
instance Semigroup (Perhaps a) where
  Perhaps Nothing <> _ = Perhaps Nothing
  _ <> Perhaps Nothing = Perhaps Nothing
  Perhaps (Just x) <> _ = Perhaps (Just x)

validateName :: Input -> Validation (Endo Input, [String]) String

A validation problem with a twist.

Until recently, I thought that data validation was a solved problem: Use an applicative functor. I then encountered a forum question that for a few minutes shook my faith.

After brief consideration, though, I realised that all is good. Validation, even with a twist, is successfully modelled with an applicative functor. Faith in computer science restored.

The twist #

Usually, when you see a demo of applicative validation, the result of validating is one of two: either a parsed result, or a collection of error messages.

λ> validateReservation $ ReservationJson "2017-06-30 19:00:00+02:00" 4 "Jane Doe" "j@example.com"
Validation (Right (Reservation {
    reservationDate = 2017-06-30 19:00:00 +0200,
    reservationQuantity = 4,
    reservationName = "Jane Doe",
    reservationEmail = "j@example.com"}))

λ> validateReservation $ ReservationJson "2017/14/12 6pm" 4.1 "Jane Doe" "jane.example.com"
Validation (Left ["Not a date.","Not a positive integer.","Not an email address."])

λ> validateReservation $ ReservationJson "2017-06-30 19:00:00+02:00" (-3) "Jane Doe" "j@example.com"
Validation (Left ["Not a positive integer."])

(Example from Applicative validation.)

What if, instead, you're displaying an input form? When users enter data, you want to validate it. Imagine, for the rest of this short series of articles that the input form has three fields: name, date of birth, and address. Each piece of data has associated validation rules.

If you enter a valid name, but an invalid date of birth, you want to clear the input form's date of birth, but not the name. It's such a bother for a user having to retype valid data just because a single field turned out to be invalid.

Imagine, for example, that you want to bind the form to a data model like this F# record type:

type Input = { Name : string option; DoB : DateTime option; Address : string option}

Each of these three fields is optional. We'd like validation to work in the following way: If validation fails, the function should return both a list of error messages, and also the Input object, with valid data retained, but invalid data cleared.

One of the rules implied in the forum question is that names must be more than three characters long. Thus, input like this is invalid:

{ Name = Some "Tom"; DoB = Some eightYearsAgo; Address = Some "x" }

Both the DoB and Address fields, however, are valid, so, along with error messages, we'd like our validation function to return a partially wiped Input value:

{ Name = None; DoB = Some eightYearsAgo; Address = Some "x" }

Notice that both DoB and Address field values are retained, while Name has been reset.

A final requirement: If validation succeeds, the return value should be a parsed value that captures that validation took place:

type ValidInput = { Name : string; DoB : DateTime; Address : string }

That requirement is straightforward. That's how you'd usually implement application validation. It's the partial data round-trip that seems to throw a spanner in the works.

How should we model such validation?

Theory, applied #

There's a subculture of functional programming that draws heavily on category theory. This is most prevalent in Haskell. I've been studying category theory in an attempt to understand what it's all about. I even wrote a substantial article series about some design patterns and how they relate to theory.

One thing I learned after I'd named that article series is that most of the useful theoretical concepts come from abstract algebra, with the possible exception of monads.

People often ask me: does all that theory have any practical use?

Yes, it does, as it turns out. It did, for example, enable me to identify a solution to the above twist in five to ten minutes.

It's a discussion that I often have, particularly with the always friendly F# community. Do you have to understand functors, monads, etcetera to be a productive F# developer?

To anyone who wants to learn F# I'd respond: Don't worry about that at the gate. Find a good learning resource and dive right in. It's a friendly language that you can learn gradually.

Sooner or later, though, you'll run into knotty problems that you may struggle to address. I've seen this enough times that it looks like a pattern. The present forum question is just one example. A beginner or intermediate F# programmer will typically attempt to solve the problem in an ad-hoc manner that may or may not be easy to maintain. (The solution proposed by the author of that forum question doesn't, by the way, look half bad.)

To be clear: there's nothing wrong with being a beginner. I was once a beginner programmer, and I'm still a beginner in multiple ways. What I'm trying to argue here is that there is value in knowing theory. With my knowledge of abstract algebra and how it applies to functional programming, it didn't take me long to identify a solution. I'll get to that later.

Before I outline a solution, I'd like to round off the discussion of applied theory. That question about monads comes up a lot. Do I have to understand functors, monads, etcetera to be a good F# developer?

I think it's like asking Do I have to understand polymorphism, design patterns, the SOLID principles, etcetera to be a good object-oriented programmer?

Those are typically not the first topics people are taught about OOD. I would assert, however, that understanding such topics do help. They may not be required to get started with OOP, but knowing them makes you a better programmer.

I think the same is true for functional programming. It's just a different skill set that makes you better in that paradigm.

Solution outline #

When you know a bit of theory, you may know that validation can be implemented with an applicative sum type like Either (AKA Option), with one extra requirement.

Either has two dimensions, left or right (success or failure, ok or error, etcetera). The applicative nature of it already supplies a way to compose the successes, but what if there's more than one validation error?

In my article about applicative validation I showed how to collect multiple error messages in a list. Lists, however, form a monoid, so I typed the validation API to be that flexible.

In fact, all you need is a semigroup. When I wrote the article on applicative validation, Haskell's Semigroup type class wasn't yet a supertype of Monoid, and I (perhaps without sufficient contemplation) just went with Monoid.

What remains is that applicative validation can collect errors for any semigroup of errors. All we need to solve the above validation problem with a twist, then, is to identify a suitable semigroup.

I don't want to give away everything in this article, so I'm going to leave you with this cliffhanger. Which semigroup solves the problem? Read on.

• A Haskell proof of concept of validation with partial data round trip
• An F# demo of validation with partial data round trip
• A C# port of validation with partial round trip

After the Haskell article, I'll show how it translates to F#. You can skip the Haskell article if you like.

Conclusion #

I still think that validation is a solved problem. It's always interesting when such a belief for a moment is challenged, and satisfying to discover that it still holds.

This is, after all, not proof of anything. Perhaps tomorrow, someone will throw another curve ball that I can't catch. If that happens, I'll have to update my beliefs. Until then, I'll consider validation a solved problem.

Next: A Haskell proof of concept of validation with partial data round trip.

Is it ever okay to branch and loop in a unit test?

When I coach development organisations about unit testing and test-driven development, there's often a sizeable group of developers who don't see the value of unit testing. Some of the arguments they typically use are worth considering.

A common complaint is that it's difficult to see the wisdom in writing code to prevent defects in code. That's not an unreasonable objection.

We have scant scientific knowledge about software engineering, but the little we know suggests that the number of defects is proportional to lines of code. The more lines of code, the more defects.

If that's true, adding more code - even when it's test code - seems like a bad idea.

Reasons to trust test code #

First, we should consider the possibility that the correlation between lines of code and defects doesn't mean that defects are evenly distributed. As Adam Tornhill argues in Your Code as a Crime Scene, defects tend to cluster in hotspots.

You can have a large proportion of your code base which is, for all intents and purpose, bug-free, and hotspots where defects keep spawning.

If this is true, adding test code isn't a problem if you can keep it bug-free.

That, however, sounds like a chicken-and-the-egg kind of problem. How can you know that test code is bug-free without tests?

I've previously answered that question. In short, you can trust a test for two reasons:

• You've seen it fail (haven't you?)
• It's simple

That's what this article is actually about.

What's in a name? #

I was working with an online restaurant reservation system (example code), and had written this test:

[Theory]
[InlineData("2023-11-24 19:00", "juliad@example.net", "Julia Domna", 5)]
[InlineData("2024-02-13 18:15", "x@example.com", "Xenia Ng", 9)]
public async Task PostValidReservationWhenDatabaseIsEmpty(
    string at,
    string email,
    string name,
    int quantity)
{
    var db = new FakeDatabase();
    var sut = new ReservationsController(db);
 
    var dto = new ReservationDto
    {
        At = at,
        Email = email,
        Name = name,
        Quantity = quantity
    };
    await sut.Post(dto);
 
    var expected = new Reservation(
        DateTime.Parse(dto.At, CultureInfo.InvariantCulture),
        dto.Email,
        dto.Name,
        dto.Quantity);
    Assert.Contains(expected, db);
}

This is a state-based test that verifies that a valid reservation makes it to the database. The test has a cyclomatic complexity of 1, and I've seen it fail, so all is good. (It may, in fact, contain a future maintenance problem, but that's a topic for another article.)

The code shown here is part of the sample code base that accompanies my book Code That Fits in Your Head.

What constitutes a valid reservation? At the very least, we should demand that At is a valid date and time, and that Quantity is a positive number. The restaurant would like to be able to email a confirmation to the user, so an email address is also required. Email addresses are notoriously difficult to validate, so we'll just require that the the string isn't null.

What about the Name? I thought about this a bit and decided that, according to Postel's law, the system should accept null names. The name is only a convenience; the system doesn't need it, it's just there so that when you arrive at the restaurant, you can say "I have a reservation for Julia" instead of giving an email address to the maître d'hôtel. But then, if you didn't supply a name when you made the reservation, you can always state your email address when you arrive. To summarise, the name is just a convenience, not a requirement.

This decision meant that I ought to write a test case with a null name.

That turned out to present a problem. I'd defined the Reservation class so that it didn't accept null arguments, and I think that's the appropriate design. Null is just evil and has no place in my domain models.

That's not a problem in itself. In this case, I think it's acceptable to convert a null name to the empty string.

Copy and paste #

Allow me to summarise. If you consider the above unit test, I needed a third test case with a null name. In that case, expected should be a Reservation value with the name "". Not null, but "".

As far as I can tell, you can't easily express that in PostValidReservationWhenDatabaseIsEmpty without increasing its cyclomatic complexity. Based on the above introduction, that seems like a no-no.

What's the alternative? Should I copy the test and adjust the single line of code that differs? If I did, it would look like this:

[Theory]
[InlineData("2023-08-23 16:55", "kite@example.edu", null, 2)]
public async Task PostValidReservationWithNullNameWhenDatabaseIsEmpty(
    string at,
    string email,
    string name,
    int quantity)
{
    var db = new FakeDatabase();
    var sut = new ReservationsController(db);
 
    var dto = new ReservationDto
    {
        At = at,
        Email = email,
        Name = name,
        Quantity = quantity
    };
    await sut.Post(dto);
 
    var expected = new Reservation(
        DateTime.Parse(dto.At, CultureInfo.InvariantCulture),
        dto.Email,
        "",
        dto.Quantity);
    Assert.Contains(expected, db);
}

Apart from the values in the [InlineData] attribute and the method name, the only difference from PostValidReservationWhenDatabaseIsEmpty is that expected has a hard-coded name of "".

This is not acceptable.

There's a common misconception that the DRY principle doesn't apply to unit tests. I don't see why this should be true. The DRY principle exists because copy-and-paste code is difficult to maintain. Unit test code is also code that you have to maintain. All the rules about writing maintainable code also apply to unit test code.

Branching in test #

What's the alternative? One option (that shouldn't be easily dismissed) is to introduce a Test Helper to perform the conversion from a nullable name to a non-nullable name. Such a helper would have a cyclomatic complexity of 2, but could be unit tested in isolation. It might even turn out that it'd be useful in the production code.

Still, that seems like overkill, so I instead made the taboo move and added branching logic to the existing test to see how it'd look:

[Theory]
[InlineData("2023-11-24 19:00", "juliad@example.net", "Julia Domna", 5)]
[InlineData("2024-02-13 18:15", "x@example.com", "Xenia Ng", 9)]
[InlineData("2023-08-23 16:55", "kite@example.edu", null, 2)]
public async Task PostValidReservationWhenDatabaseIsEmpty(
    string at,
    string email,
    string name,
    int quantity)
{
    var db = new FakeDatabase();
    var sut = new ReservationsController(db);
 
    var dto = new ReservationDto
    {
        At = at,
        Email = email,
        Name = name,
        Quantity = quantity
    };
    await sut.Post(dto);
 
    var expected = new Reservation(
        DateTime.Parse(dto.At, CultureInfo.InvariantCulture),
        dto.Email,
        dto.Name ?? "",
        dto.Quantity);
    Assert.Contains(expected, db);
}

Notice that the expected name is now computed as dto.Name ?? "". Perhaps you think about branching instructions as relating exclusively to keywords such as if or switch, but the ?? operator is also a branching instruction. The test now has a cyclomatic complexity of 2.

Is that okay?

To branch or not to branch #

I think that in this case, it's okay to slightly increase the cyclomatic complexity of the test. It's not something I just pull out of my hat, though. I think it's possible to adjust the above heuristics to embrace this sort of variation.

To be clear, I consider this an advanced practice. If you're just getting started with unit testing, try to keep tests simple. Keep the cyclomatic complexity at 1.

Had I been in the above situation a couple of years ago, I might not have considered this option. About a year ago, though, I watched John Hughes' presentation Building on developers' intuitions to create effective property-based tests. When he, about 15 minutes in, wrote a test with a branching instruction, I remember becoming quite uncomfortable. This lasted for a while until I understood where he was going with it. It's truly an inspiring and illuminating talk; I highly recommend it.

How it relates to the problem presented here is through coverage. While the PostValidReservationWhenDatabaseIsEmpty test now has a cyclomatic complexity of 2, it's a parametrised test with three test cases. Two of these cover one branch, and the third covers the other.

What's more important is the process that produced the test. I added one test case at a time, and for each case, I saw the test fail.

Specifically, when I added the third test case with the null name, I first added the branching expression dto.Name ?? "" and ran the two existing tests. They still both passed, which bolstered my belief that they both exercised the left branch of that expression. I then added the third case and saw that it (and only it) failed. This supported my belief that the third case exercised the right branch of ??.

Branching in unit tests isn't something I do lightly. I still believe that it could make the test more vulnerable to future changes. I'm particularly worried about making a future change that might shift one or more of these test cases into false negatives in the form of tautological assertions.

Conclusion #

As you can tell, when I feel that I'm moving onto thin ice, I move deliberately. If there's one thing I've learned from decades of professional programming it's that my brain loves jumping to conclusions. Moving slowly and deliberately is my attempt at countering this tendency. I believe that it enables me to go faster in the long run.

I don't think that branching in unit tests should be common, but I believe that it may be occasionally valid. The key, I think, is to guarantee that each branch in the test is covered by a test case. The implication is that there must be at least as many test cases as the cyclomatic complexity. In other words, the test must be a parametrised test.

Comments

		[Theory]
		[InlineData("2023-11-24 19:00", "juliad@example.net", "Julia Domna", 5)]
		[InlineData("2024-02-13 18:15", "x@example.com",      "Xenia Ng",    9)]
		public async Task PostValidReservationWithNameWhenDatabaseIsEmpty
		           (string at,          string email,         string name,   int quantity) =>
			PostValidReservationWhenDatabaseIsEmpty(at, email, name, expectedName: name, quantity);

		[Fact]
		public async Task PostValidReservationWithoutNameWhenDatabaseIsEmpty() =>
			PostValidReservationWhenDatabaseIsEmpty(
				at          : "2023-11-24 19:00",
				email       : "juliad@example.net",
				name        : null,
				expectedName: "",
				quantity    : 5);

		private async Task PostValidReservationWhenDatabaseIsEmpty(
			    string at,
			    string email,
			    string name,
			    string expectedName,
			    int    quantity)
		{
		    var db = new FakeDatabase();
		    var sut = new ReservationsController(db);
		 
		    var dto = new ReservationDto
		    {
		        At       = at,
		        Email    = email,
		        Name     = name,
		        Quantity = quantity,
		    };
		    await sut.Post(dto);
		 
		    var expected = new Reservation(
		        DateTime.Parse(dto.At, CultureInfo.InvariantCulture),
		        dto.Email,
		        expectedName, // /!\ Not `dto.Name`,
		        dto.Quantity);
		    Assert.Contains(expected, db);
		}

				

private class PostValidReservationWhenDatabaseIsEmptyTestCases :
    TheoryData<ReservationDto, Reservation>
{
    public PostValidReservationWhenDatabaseIsEmptyTestCases()
    {
        AddWithName(new DateTime(2023, 11, 24, 19, 0, 0), "juliad@example.net", "Julia Domna", 5);
        AddWithName(new DateTime(2024, 2, 13, 18, 15, 0), "x@example.com", "Xenia Ng", 9);
        AddWithoutName(new DateTime(2023, 8, 23, 16, 55, 0), "kite@example.edu", 2);
    }
 
    private void AddWithName(DateTime at, string email, string name, int quantity)
    {
        Add(new ReservationDto
            {
                At = at.ToString("O"),
                Email = email,
                Name = name,
                Quantity = quantity
            },
            new Reservation(at, email, name, quantity));
    }
 
    private void AddWithoutName(DateTime at, string email, int quantity)
    {
        Add(new ReservationDto { At = at.ToString("O"), Email = email, Quantity = quantity },
            new Reservation(at, email, "", quantity));
    }
}
 
 
[Theory, ClassData(typeof(PostValidReservationWhenDatabaseIsEmptyTestCases))]
public async Task PostValidReservationWhenDatabaseIsEmpty(
    ReservationDto dto, Reservation expected)
{
    var db = new FakeDatabase();
    var sut = new ReservationsController(db);
 
    await sut.Post(dto);
 
    Assert.Contains(expected, db);
}

Consider naming variables according to their role, instead of their type.

My recent article on good names might leave you with the impression that I consider good names unimportant. Not at all. That article was an attempt at delineating the limits of naming. Good names aren't the panacea some people seem to imply, but they're still important.

As the cliché goes, naming is one of the hardest problems in software development. Perhaps it's hard because you have to do it so frequently. Every time you create a variable, you have to name it. It's also an opportunity to add clarity to a code base.

A common naming strategy is to name objects after their type:

Reservation? reservation = dto.Validate(id);

or:

Restaurant? restaurant = await RestaurantDatabase.GetRestaurant(restaurantId);

There's nothing inherently wrong with a naming scheme like this. It often makes sense. The reservation variable is a Reservation object, and there's not that much more to say about it. The same goes for the restaurant object.

In some contexts, however, objects play specific roles. This is particularly prevalent with primitive types, but can happen to any type of object. It may help the reader if you name the variables according to such roles.

In this article, I'll show you several examples. I hope these examples are so plentiful and varied that they can inspire you to come up with good names. Most of the code shown here is part of the sample code base that accompanies my book Code That Fits in Your Head.

A variable introduced only to be named #

In a recent article I showed this code snippet:

private bool SignatureIsValid(string candidate, ActionExecutingContext context)
{
    var sig = context.HttpContext.Request.Query["sig"];
    var receivedSignature = Convert.FromBase64String(sig.ToString());
 
    using var hmac = new HMACSHA256(urlSigningKey);
    var computedSignature = hmac.ComputeHash(Encoding.ASCII.GetBytes(candidate));
 
    var signaturesMatch = computedSignature.SequenceEqual(receivedSignature);
    return signaturesMatch;
}

Did you wonder about the signaturesMatch variable? Why didn't I just return the result of SequenceEqual, like the following?

private bool SignatureIsValid(string candidate, ActionExecutingContext context)
{
    var sig = context.HttpContext.Request.Query["sig"];
    var receivedSignature = Convert.FromBase64String(sig.ToString());
 
    using var hmac = new HMACSHA256(urlSigningKey);
    var computedSignature = hmac.ComputeHash(Encoding.ASCII.GetBytes(candidate));
 
    return computedSignature.SequenceEqual(receivedSignature);
}

Visual Studio even offers this as a possible refactoring that it'll do for you.

The inclusion of the signaturesMatch variable was a conscious decision of mine. I felt that directly returning the result of SequenceEqual was a bit too implicit. It forces readers to make the inference themselves: Ah, the two arrays contain the same sequence of bytes; that must mean that the signatures match!

Instead of asking readers to do that work themselves, I decided to do it for them. I hope that it improves readability. It doesn't change the behaviour of the code one bit.

Test roles #

When it comes to unit testing, there's plenty of inconsistent terminology. One man's mock object is another woman's test double. Most of the jargon isn't even internally consistent. Do yourself a favour and adopt a consistent pattern language. I use the one presented in xUnit Test Patterns.

For instance, the thing that you're testing is the System Under Test (SUT). This can be a pure function or a static method, but when it's an object, you're going to create a variable. Consider naming it sut. A typical test also defines other variables. Naming one of them sut clearly identifies which of them is the SUT. It also protects the tests against the class in question being renamed.

[Fact]
public void ScheduleSingleReservationCommunalTable()
{
    var table = Table.Communal(12);
    var sut = new MaitreD(
        TimeSpan.FromHours(18),
        TimeSpan.FromHours(21),
        TimeSpan.FromHours(6),
        table);
 
    var r = Some.Reservation;
    var actual = sut.Schedule(new[] { r });
 
    var expected = new[] { new TimeSlot(r.At, table.Reserve(r)) };
    Assert.Equal(expected, actual);
}

The above test follows my AAA.formatting heuristic. In all, it defines five variables, but there can be little doubt about which one is the sut.

The table and r variables follow the mainstream practice of naming variables after their type. They play no special role, so that's okay. You may balk at such a short variable name as r, and that's okay. In my defence, I follow Clean Code's N5 heuristic for long and short scopes. A variable name like r is fine when it only spans three lines of code (four, if you also count the blank line).

Consider also using the variable names expected and actual, as in the above example. In many unit testing frameworks, those are the argument names for the assertion. For instance, in xUnit.net (which the above test uses) the Assert.Equals overloads are defined as Equal<T>(T expected, T actual). Using these names for variables makes the roles clearer, I think.

The other #

The above assertion relies on structural equality. The TimeSlot class is immutable, so it can safely override Equals (and GetHashCode) to implement structural equality:

public override bool Equals(object? obj)
{
    return obj is TimeSlot other &&
           At == other.At &&
           Tables.SequenceEqual(other.Tables);
}

I usually call the downcast variable other because, from the perspective of the instance, it's the other object. I usually use that convention whenever an instance interacts with another object of the same type. Among other examples, this happens when you model objects as semigroups and monoids. The Angle struct, for example, defines this binary operation:

public Angle Add(Angle other)
{
    return new Angle(this.degrees + other.degrees);
}

Again, the method argument is in the role as the other object, so naming it other seems natural.

Here's another example from a restaurant reservation code base:

public bool Overlaps(Seating other)
{
    if (other is null)
        throw new ArgumentNullException(nameof(other));
 
    return Start < other.End && other.Start < End;
}

The Overlaps method is an instance method on the Seating class. Again, other seems natural.

Candidates #

The Overlaps method looks like a predicate, i.e. a function that returns a Boolean value. In the case of that method, other indicates the role of being the other object, but it also plays another role. It makes sense to me to call predicate input candidates. Typically, you have some input that you want to evaluate as either true or false. I think it makes sense to think of such a parameter as a 'truth candidate'. You can see one example of that in the above SignatureIsValid method.

There, the string parameter is a candidate for having a valid signature.

Here's another restaurant-related example:

public bool WillAccept(
    DateTime now,
    IEnumerable<Reservation> existingReservations,
    Reservation candidate)
{
    if (existingReservations is null)
        throw new ArgumentNullException(nameof(existingReservations));
    if (candidate is null)
        throw new ArgumentNullException(nameof(candidate));
    if (candidate.At < now)
        return false;
    if (IsOutsideOfOpeningHours(candidate))
        return false;
 
    var seating = new Seating(SeatingDuration, candidate.At);
    var relevantReservations =
        existingReservations.Where(seating.Overlaps);
    var availableTables = Allocate(relevantReservations);
    return availableTables.Any(t => t.Fits(candidate.Quantity));
}

Here, the reservation in question is actually not yet a reservation. It might be rejected, so it's a candidate reservation.

You can also use that name in TryParse methods, as shown in this article.

Data Transfer Objects #

Another name that I like to use is dto for Data Transfer Objects (DTOs). The benefit here is that as long as dto is unambiguous in context, it makes it easier to distinguish between a DTO and the domain model you might want to turn it into:

[HttpPost("restaurants/{restaurantId}/reservations")]
public async Task<ActionResult> Post(int restaurantId, ReservationDto dto)
{
    if (dto is null)
        throw new ArgumentNullException(nameof(dto));
 
    var id = dto.ParseId() ?? Guid.NewGuid();
    Reservation? reservation = dto.Validate(id);
    if (reservation is null)
        return new BadRequestResult();
 
    var restaurant = await RestaurantDatabase.GetRestaurant(restaurantId);
    if (restaurant is null)
        return new NotFoundResult();
 
    return await TryCreate(restaurant, reservation);
}

By naming the input parameter dto, I keep the name reservation free for the domain object, which ought to be the more important object of the two.

"A Data Transfer Object is one of those objects our mothers told us never to write."

Time #

You often need to make decisions based on the current time or date. In .NET the return value from DateTime.Now is a DateTime value. Typical variable names are dateTime, date, time, or dt, but why not call it now?

private async Task<ActionResult> TryCreate(Restaurant restaurant, Reservation reservation)
{
    using var scope = new TransactionScope(TransactionScopeAsyncFlowOption.Enabled);
 
    var reservations = await Repository.ReadReservations(restaurant.Id, reservation.At);
    var now = Clock.GetCurrentDateTime();
    if (!restaurant.MaitreD.WillAccept(now, reservations, reservation))
        return NoTables500InternalServerError();
 
    await Repository.Create(restaurant.Id, reservation).ConfigureAwait(false);
 
    scope.Complete();
 
    return Reservation201Created(restaurant.Id, reservation);
}

This is the TryCreate method called by the above Post method. Here, DateTime.Now is hidden behind Clock.GetCurrentDateTime() in order to make execution repeatable, but the idea remains: the variable represents the current time or date, or, with a bit of good will, now.

Notice that the WillAccept method (shown above) also uses now as a parameter name. That value's role is to represent now as a concept.

When working with time, I also sometimes use the variable names before and after. This is mostly useful in integration tests:

[Fact]
public async Task GetCurrentYear()
{
    using var api = new LegacyApi();
 
    var before = DateTime.Now;
    var response = await api.GetCurrentYear();
    var after = DateTime.Now;
 
    response.EnsureSuccessStatusCode();
    var actual = await response.ParseJsonContent<CalendarDto>();
    AssertOneOf(before.Year, after.Year, actual.Year);
    Assert.Null(actual.Month);
    Assert.Null(actual.Day);
    AssertLinks(actual);
}

While you can inject something like a Clock dependency in order to make your SUT deterministic, in integration tests you might want to see behaviour when using the system clock. You can often verify such behaviour by surrounding the test's Act phase with two calls to DateTime.Now. This gives you the time before and after the test exercised the SUT.

When you do that, however, be careful with the assertions. If such a test runs at midnight, before and after might be two different dates. If it runs on midnight December 31, it might actually be two different years! That's the reason that the test passes as long as the actual.Year is either of before.Year and after.Year.

Invalid values #

While integration tests often test happy paths, unit tests should also exercise error paths. What happens when you supply invalid input to a method? When you write such tests, you can identify the invalid values by naming the variables or parameters accordingly:

[Theory]
[InlineData(null)]
[InlineData("")]
[InlineData("bas")]
public async Task PutInvalidId(string invalidId)
{
    var db = new FakeDatabase();
    var sut = new ReservationsController(
        new SystemClock(),
        new InMemoryRestaurantDatabase(Some.Restaurant),
        db);
 
    var dummyDto = new ReservationDto
    {
        At = "2024-06-25 18:19",
        Email = "colera@example.com",
        Name = "Cole Aera",
        Quantity = 2
    };
    var actual = await sut.Put(invalidId, dummyDto);
 
    Assert.IsAssignableFrom<NotFoundResult>(actual);
}

Here, the invalid input represent an ID. To indicate that, I called the parameter invalidId.

The system under test is the Put method, which takes two arguments:

public Task<ActionResult> Put(string id, ReservationDto dto)

When testing an error path, it's important to keep other arguments well-behaved. In this example, I want to make sure that it's the invalidId that causes the NotFoundResult result. Thus, the dto argument should be as well-behaved as possible, so that it isn't going to be the source of divergence.

Apart from being well-behaved, that object plays no role in the test. It just needs to be there to make the code compile. xUnit Test Patterns calls such an object a Dummy Object, so I named the variable dummyDto as information to any reader familiar with that pattern language.

Derived class names #

The thrust of all of these examples is that you don't have to name variables after their types. You can extend this line of reasoning to class inheritance. Just because a base class is called Foo it doesn't mean that you have to call a derived class SomethingFoo.

This is something of which I have to remind myself. For example, to support integration testing with ASP.NET you'll need a WebApplicationFactory<TEntryPoint>. To override the default DI Container configuration, you'll have to derive from this class and override its ConfigureWebHost method. In an example I've previously published I didn't spend much time thinking about the class name, so RestaurantApiFactory it was.

At first, I named the variables of this type factory, or something equally devoid of information. That bothered me, so instead tried service, which I felt was an improvement, but still too vapid. I then adopted api as a variable name, but then realised that that also suggested a better class name. So currently, this defines my self-hosting API:

public sealed class SelfHostedApi : WebApplicationFactory<Startup>

Here's how I use it:

[Fact]
public async Task ReserveTableAtNono()
{
    using var api = new SelfHostedApi();
    var client = api.CreateClient();
    var dto = Some.Reservation.ToDto();
    dto.Quantity = 6;
 
    var response = await client.PostReservation("Nono", dto);
 
    var at = Some.Reservation.At;
    await AssertRemainingCapacity(client, at, "Nono", 4);
    await AssertRemainingCapacity(client, at, "Hipgnosta", 10);
}

The variable is just called api, but the reader can tell from the initialisation that this is an instance of the SelfHostedApi class. I like how that communicates that this is an integration test that uses a self-hosted API. It literally says that.

This test also uses the dto naming convention. Additionally, you may take note of the variable and property called at. That's another name for a date and time. I struggled with naming this value, until Karsten Strøbæk suggested that I used the simple word at: reservation.At indicates the date and time of the reservation without being encumbered by awkward details about date and time. Should we call it date? time? dateTime? No, just call it at. I find it elegant.

Conclusion #

Sometimes, a Reservation object is just a reservation, and that's okay. At other times, it's the actual value, or the expected value. If it represents an invalid reservation in a test case, it makes sense to call the variable invalidResevation.

Giving variables descriptive names improves code quality. You don't have to write comments as apologies for poor readability if a better name communicates what the comment would have said.

Consider naming variables (and classes) for the roles they play, rather than their types.

On the other hand, when variable names are in the way, consider point-free code.

Comments

public Reservation(Guid id, DateTime at, Email email, Name name, int quantity)

Good names are important, but insufficient, for code maintainability.

You should give the building blocks of your code bases descriptive names. It's easier to understand the purpose of a library, module, class, method, function, etcetera if the name contains a clue about the artefact's purpose. This is hardly controversial, and while naming is hard, most teams I visit agree that names are important.

Still, despite good intentions and efforts to name things well, code bases deteriorate into unmaintainable clutter.

Clearly, good names aren't enough.

Tenuousness of names #

A good name is tenuous. First, naming is hard, so while you may have spent some effort coming up with a good name, other people may misinterpret it. Because they originate from natural language, names are as ambiguous as language. (Terse operators, on the other hand...)

Another maintainability problem with names is that implementation may change over time, but the names remain constant. Granted, modern IDEs make it easy to rename methods, but developers rarely adjust names when they adjust behaviour. Even the best names may become misleading over time.

These weakness aren't the worst, though. In my experience, a more fundamental problem is that all it takes is one badly named 'wrapper object' before the information in a good name is lost.

In the figure, the inner object is well-named. It has a clear name and descriptive method names. All it takes before this information is lost, however, is another object with vague names to 'encapsulate' it.

An attempt at a descriptive method name #

Here's an example. Imagine an online restaurant reservation system. One of the features of this system is to take reservations and save them in the database.

A restaurant, however, is a finite resource. It can only accommodate a certain number of guests at the same time. Whenever the system receives a reservation request, it'll have to retrieve the existing reservations for that time and make a decision. Can it accept the reservation? Only if it can should it save the reservation.

How do you model such an interaction? How about a descriptive name? How about TrySave? Here's a possible implementation:

public async Task<bool> TrySave(Reservation reservation)
{
    if (reservation is null)
        throw new ArgumentNullException(nameof(reservation));
 
    var reservations = await Repository
        .ReadReservations(
            reservation.At,
            reservation.At + SeatingDuration)
        .ConfigureAwait(false);
    var availableTables = Allocate(reservations);
    if (!availableTables.Any(t => reservation.Quantity <= t.Seats))
        return false;
 
    await Repository.Create(reservation).ConfigureAwait(false);
    return true;
}

There's an implicit naming convention in .NET that methods with the Try prefix indicate an operation that may or may not succeed. The return value of such methods is either true or false, and they may also have out parameters if they optionally produce a value. That's not the case here, but I think one could make the case that TrySave succinctly describes what's going on.

All is good, then?

A vague wrapper #

After our conscientious programmer meticulously designed and named the above TrySave method, it turns out that it doesn't meet all requirements. Users of the system file a bug: the system accepts reservations outside the restaurant's opening hours.

The original programmer has moved on to greener pastures, so fixing the bug falls on a poor maintenance developer with too much to do. Having recently learned about the open-closed principle, our new protagonist decides to wrap the existing TrySave in a new method:

public async Task<bool> Check(Reservation reservation)
{
    if (reservation is null)
        throw new ArgumentNullException(nameof(reservation));
 
    if (reservation.At < DateTime.Now)
        return false;
    if (reservation.At.TimeOfDay < OpensAt)
        return false;
    if (LastSeating < reservation.At.TimeOfDay)
        return false;
 
    return await Manager.TrySave(reservation).ConfigureAwait(false);
}

This new method first checks whether the reservation is within opening hours and in the future. If that's not the case, it returns false. Only if these preconditions are fulfilled does it delegate the decision to that TrySave method.

Notice, however, the name. The bug was urgent, and our poor programmer didn't have time to think of a good name, so Check it is.

Caller's perspective #

How does this look from the perspective of calling code? Here's the Controller action that handles the pertinent HTTP request:

public async Task<ActionResult> Post(ReservationDto dto)
{
    if (dto is null)
        throw new ArgumentNullException(nameof(dto));
 
    Reservation? r = dto.Validate();
    if (r is null)
        return new BadRequestResult();
 
    var isOk = await Manager.Check(r).ConfigureAwait(false);
    if (!isOk)
        return new StatusCodeResult(StatusCodes.Status500InternalServerError);
 
    return new NoContentResult();
}

Try to forget the code you've just seen and imagine that you're looking at this code first. You'd be excused if you miss what's going on. It looks as though the method just does a bit of validation and then checks 'something' concerning the reservation.

There's no hint that the Check method might perform the significant side effect of saving the reservation in the database.

You'll only learn that if you read the implementation details of Check. As I argue in my Humane Code video, if you have to read the source code of an object, encapsulation is broken.

Such code doesn't fit in your brain. You'll struggle as you try keep track of all the things that are going on in the code, all the way from the outer boundary of the application to implementation details that relate to databases, third-party services, etcetera.

Straw man? #

You may think that this is a straw man argument. After all, wouldn't it be better to edit the original TrySave method?

Perhaps, but it would make that class more complex. The TrySave method has a cyclomatic complexity of only 3, while the Check method has a complexity of 5. Combining them might easily take them over some threshold.

Additionally, each of these two classes have different dependencies. As the TrySave method implies, it relies on both Repository and SeatingDuration, and the Allocate helper method (not shown) uses a third dependency: the restaurant's table configuration.

Likewise, the Check method relies on OpensAt and LastSeating. If you find it better to edit the original TrySave method, you'd have to combine these dependencies as well. Each time you do that, the class grows until it becomes a God object.

It's rational to attempt to separate things in multiple classes. It also, on the surface, seems to make unit testing easier. For example, here's a test that verifies that the Check method rejects reservations before the restaurant's opening time:

[Fact]
public async Task RejectReservationBeforeOpeningTime()
{
    var r = new Reservation(
        DateTime.Now.AddDays(10).Date.AddHours(17),
        "colaera@example.com",
        "Cole Aera",
        1);
    var mgrTD = new Mock<IReservationsManager>();
    mgrTD.Setup(mgr => mgr.TrySave(r)).ReturnsAsync(true);
    var sut = new RestaurantManager(
        TimeSpan.FromHours(18),
        TimeSpan.FromHours(21),
        mgrTD.Object);
 
    var actual = await sut.Check(r);
 
    Assert.False(actual);
}

By replacing the TrySave method by a test double, you've ostensibly decoupled the Check method from all the complexity of the TrySave method.

To be clear, this style of programming, with lots of nested interfaces and tests with mocks and stubs is far from ideal, but I still find it better than a big ball of mud.

Alternative #

A better alternative is Functional Core, Imperative Shell, AKA impureim sandwich. Move all impure actions to the edge of the system, leaving only referentially transparent functions as the main implementers of logic. It could look like this:

[HttpPost]
public async Task<ActionResult> Post(ReservationDto dto)
{
    if (dto is null)
        throw new ArgumentNullException(nameof(dto));
 
    var id = dto.ParseId() ?? Guid.NewGuid();
    Reservation? r = dto.Validate(id);
    if (r is null)
        return new BadRequestResult();
 
    var reservations = await Repository.ReadReservations(r.At).ConfigureAwait(false);
    if (!MaitreD.WillAccept(DateTime.Now, reservations, r))
        return NoTables500InternalServerError();
 
    await Repository.Create(r).ConfigureAwait(false);
 
    return Reservation201Created(r);
}

Nothing is swept under the rug here. WillAccept is a pure function, and while it encapsulates significant complexity, the only thing you need to understand when you're trying to understand the above Post code is that it returns either true or false.

Another advantage of pure functions is that they are intrinsically testable. That makes unit testing and test-driven development easier.

Even with a functional core, you'll also have an imperative shell. You can still test that, too, such as the Post method. It isn't referentially transparent, so you might be inclined to use mocks and stubs, but I instead recommend state-based testing with a Fake database.

Conclusion #

Good names are important, but don't let good names, alone, lull you into a false sense of security. All it takes is one vaguely named wrapper object, and all the information in your meticulously named methods is lost.

This is one of many reasons I try to design with static types instead of names. Not that I dismiss the value of good names. After all, you'll have to give your types good names as well.

Types are more robust in the face of inadvertent changes; or, rather, they tend to resist when we try to do something stupid. I suppose that's what lovers of dynamically typed languages feel as 'friction'. In my mind, it's entirely opposite. Types keep me honest.

Unfortunately, most type systems don't offer an adequate degree of safety. Even in F#, which has a great type system, you can introduce impure actions into what you thought was a pure function, and you'd be none the wiser. That's one of the reasons I find Haskell so interesting. Because of the way IO works, you can't inadvertently sweep surprises under the rug.

Comments

Evolving REST API URLs when cool URIs don't change.

More than one reader reacted to my article on fit URLs by asking about bookmarks and original URLs. Daniel Sklenitzka's question is a good example:

"I see how signing the URLs prevents clients from retro-engineering the URL templates, but how does it help preventing breaking changes? If the client stores the whole URL instead of just the ID and later the URL changes because a restaurant ID is added, the original URL is still broken, isn't it?"

The rules of HTTP #

I agree with the implicit assumption that clients are allowed to bookmark links. It seems, then, like a breaking change if you later change your internal URL scheme. That seems to imply that the bookmarked URL is gone, breaking a tenet of the HTTP protocol: Cool URIs don't change.

REST APIs are supposed to play by the rules of HTTP, so it'd seem that once you've published a URL, you can never retire it. You can, on the other hand, change its behaviour.

Let's call such URLs legacy URLs. Keep them around, but change them to return 301 Moved Permanently responses.

The rules of REST go both ways. The API is expected to play by the rules of HTTP, and so are the clients. Clients are not only expected to follow links, but also redirects. If a legacy URL starts returning a 301 Moved Permanently response, a well-behaved client doesn't break.

Reverse proxy #

As I've previously described, one of the many benefits of HTTP-based services is that you can put a reverse proxy in front of your application servers. I've no idea how to configure or operate NGINX or Varnish, but from talking to people who do know, I get the impression that they're quite scriptable.

Since the above ideas are independent of actual service implementation or behaviour, it's a generic problem that you should seek to address with general-purpose software.

Imagine that a reverse proxy is configured with a set of rules that detects legacy URLs and knows how to forward them. Clearly, the reverse proxy must know of the REST API's current URL scheme to be able to do that. You might think that this would entail leaking an implementation detail, but just as I consider any database used by the API as part of the overall system, I'd consider the reverse proxy as just another part.

Redirecting with ASP.NET #

If you don't have a reverse proxy, you can also implement redirects in code. It'd be better to use something like a reverse proxy, because that would mean that you get to delete code from your code base, but sometimes that's not possible.

The code shown here is part of the sample code base that accompanies my book Code That Fits in Your Head.

In ASP.NET, you can return 301 Moved Permanently responses just like any other kind of HTTP response:

[Obsolete("Use Get method with restaurant ID.")]
[HttpGet("calendar/{year}/{month}")]
public ActionResult LegacyGet(int year, int month)
{
    return new RedirectToActionResult(
        nameof(Get),
        null,
        new { restaurantId = Grandfather.Id, year, month },
        permanent: true);
}

This LegacyGet method redirects to the current Controller action called Get by supplying the arguments that the new method requires. The Get method has this signature:

[HttpGet("restaurants/{restaurantId}/calendar/{year}/{month}")]
public async Task<ActionResult> Get(int restaurantId, int year, int month)

When I expanded the API from a single restaurant to a multi-tenant system, I had to grandfather in the original restaurant. I gave it a restaurantId, but in order to not put magic constants in the code, I defined it as the named constant Grandfather.Id.

Notice that I also adorned the LegacyGet method with an [Obsolete] attribute to make it clear to maintenance programmers that this is legacy code. You might argue that the Legacy prefix already does that, but the [Obsolete] attribute will make the compiler emit a warning, which is even better feedback.

Regression test #

While legacy URLs may be just that: legacy, that doesn't mean that it doesn't matter whether or not they work. You may want to add regression tests.

If you implement redirects in code (as opposed to a reverse proxy), you should also add automated tests that verify that the redirects work:

[Theory]
[InlineData("http://localhost/calendar/2020?sig=ePBoUg5gDw2RKMVWz8KIVzF%2Fgq74RL6ynECiPpDwVks%3D")]
[InlineData("http://localhost/calendar/2020/9?sig=ZgxaZqg5ubDp0Z7IUx4dkqTzS%2Fyjv6veDUc2swdysDU%3D")]
public async Task BookmarksStillWork(string bookmarkedAddress)
{
    using var api = new LegacyApi();
 
    var actual = await api.CreateDefaultClient().GetAsync(new Uri(bookmarkedAddress));
 
    Assert.Equal(HttpStatusCode.MovedPermanently, actual.StatusCode);
    var follow = await api.CreateClient().GetAsync(actual.Headers.Location);
    follow.EnsureSuccessStatusCode();
}

This test interacts with a self-hosted service at the HTTP level. LegacyApi is a test-specific helper class that derives from WebApplicationFactory<Startup>.

The test uses URLs that I 'bookmarked' before I evolved the URLs to a multi-tenant system. As you can tell from the host name (localhost), these are bookmarks against the self-hosted service. The test first verifies that the response is 301 Moved Permanently. It then requests the new address and uses EnsureSuccessStatusCode as an assertion.

Conclusion #

When you evolve fit URLs, it could break clients that may have bookmarked legacy URLs. Consider leaving 301 Moved Permanently responses at those addresses.