New and hopefully improved documentation for matchers

This commit is contained in:
Martin Hořeňovský 2020-03-01 15:48:28 +01:00
parent 121f04ffcf
commit 64d7f9b98a
No known key found for this signature in database
GPG Key ID: DE48307B8B0D381A
2 changed files with 307 additions and 121 deletions

View File

@ -1,77 +1,123 @@
<a id="top"></a>
# Matchers
Matchers are an alternative way to do assertions which are easily extensible and composable.
This makes them well suited to use with more complex types (such as collections) or your own custom types.
Matchers were first popularised by the [Hamcrest](https://en.wikipedia.org/wiki/Hamcrest) family of frameworks.
**Contents**<br>
[Using Matchers](#using-matchers)<br>
[Built-in matchers](#built-in-matchers)<br>
[Writing custom matchers (old style)](#writing-custom-matchers-old-style)<br>
[Writing custom matchers (new style)](#writing-custom-matchers-new-style)<br>
## In use
Matchers, as popularized by the [Hamcrest](https://en.wikipedia.org/wiki/Hamcrest)
framework are an alternative way to write assertions, useful for tests
where you work with complex types or need to assert more complex
properties. Matchers are easily composable and users can write their
own and combine them with the Catch2-provided matchers seamlessly.
Matchers are introduced with the `REQUIRE_THAT` or `CHECK_THAT` macros, which take two arguments.
The first argument is the thing (object or value) under test. The second part is a match _expression_,
which consists of either a single matcher or one or more matchers combined using `&&`, `||` or `!` operators.
For example, to assert that a string ends with a certain substring:
## Using Matchers
```c++
using Catch::Matchers::EndsWith; // or Catch::EndsWith
std::string str = getStringFromSomewhere();
REQUIRE_THAT( str, EndsWith( "as a service" ) );
```
Matchers are most commonly used in tandem with the `REQUIRE_THAT` or
`CHECK_THAT` macros. The `REQUIRE_THAT` macro takes two arguments,
the first one is the input (object/value) to test, the second argument
is the matcher itself.
The matcher objects can take multiple arguments, allowing more fine tuning.
The built-in string matchers, for example, take a second argument specifying whether the comparison is
case sensitive or not:
```c++
REQUIRE_THAT( str, EndsWith( "as a service", Catch::CaseSensitive::No ) );
```
And matchers can be combined:
```c++
REQUIRE_THAT( str,
EndsWith( "as a service" ) ||
(StartsWith( "Big data" ) && !Contains( "web scale" ) ) );
```
_The combining operators do not take ownership of the matcher objects.
This means that if you store the combined object, you have to ensure that
the matcher objects outlive its last use. What this means is that code
like this leads to a use-after-free and (hopefully) a crash:_
For example, to assert that a string ends with the "as a service"
substring, you can write the following assertion
```cpp
using Catch::Matchers::EndsWith;
REQUIRE_THAT( getSomeString(), EndsWith("as a service") );
```
Individual matchers can also be combined using the C++ logical
operators, that is `&&`, `||`, and `!`, like so:
```cpp
using Catch::Matchers::EndsWith;
using Catch::Matchers::Contains;
REQUIRE_THAT( getSomeString(),
EndsWith("as a service") && Contains("web scale"));
```
The example above asserts that the string returned from `getSomeString`
_both_ ends with the suffix "as a service" _and_ contains the string
"web scale" somewhere.
Both of the string matchers used in the examples above live in the
`catch_matchers_string.h` header, so to compile the code above also
requires `#include <catch2/matchers/catch_matchers_string.h>`.
**IMPORTANT**: The combining operators do not take ownership of the
matcher objects being combined. This means that if you store combined
matcher object, you have to ensure that the matchers being combined
outlive its last use. What this means is that the following code leads
to a use-after-free (UAF):
```cpp
#include <catch2/catch_test_macros.hpp>
#include <catch2/matchers/catch_matchers_string.h>
TEST_CASE("Bugs, bugs, bugs", "[Bug]"){
std::string str = "Bugs as a service";
auto match_expression = Catch::EndsWith( "as a service" ) ||
(Catch::StartsWith( "Big data" ) && !Catch::Contains( "web scale" ) );
auto match_expression = Catch::Matchers::EndsWith( "as a service" ) ||
(Catch::Matchers::StartsWith( "Big data" ) && !Catch::Matchers::Contains( "web scale" ) );
REQUIRE_THAT(str, match_expression);
}
```
## Built in matchers
Catch2 provides some matchers by default. They can be found in the
`Catch::Matchers::foo` namespace and are imported into the `Catch`
namespace as well.
## Built-in matchers
There are two parts to each of the built-in matchers, the matcher
type itself and a helper function that provides template argument
deduction when creating templated matchers. As an example, the matcher
for checking that two instances of `std::vector` are identical is
`EqualsMatcher<T>`, but the user is expected to use the `Equals`
helper function instead.
Every matcher provided by Catch2 is split into 2 parts, a factory
function that lives in the `Catch::Matchers` namespace, and the actual
matcher type that is in some deeper namespace and should not be used by
the user. In the examples above, we used `Catch::Matchers::Contains`.
This is the factory function for the
`Catch::Matchers::StdString::ContainsMatcher` type that does the actual
matching.
Out of the box, Catch2 provides the following matchers:
### String matchers
The string matchers are `StartsWith`, `EndsWith`, `Contains`, `Equals` and `Matches`. The first four match a literal (sub)string against a result, while `Matches` takes and matches an ECMAScript regex. Do note that `Matches` matches the string as a whole, meaning that "abc" will not match against "abcd", but "abc.*" will.
### `std::string` matchers
Each of the provided `std::string` matchers also takes an optional second argument, that decides case sensitivity (by-default, they are case sensitive).
Catch2 provides 5 different matchers that work with `std::string`,
* `StartsWith(std::string str, CaseSensitive)`,
* `EndsWith(std::string str, CaseSensitive)`,
* `Contains(std::string str, CaseSensitive)`,
* `Equals(std::string str, CaseSensitive)`, and
* `Matches(std::string str, CaseSensitive)`.
The first three should be fairly self-explanatory, they succeed if
the argument starts with `str`, ends with `str`, or contains `str`
somewhere inside it.
The `Equals` matcher matches a string if (and only if) the argument
string is equal to `str`.
Finally, the `Matches` matcher performs an ECMASCript regex match using
`str` against the argument string. It is important to know that
the match is performed agains the string as a whole, meaning that
the regex `"abc"` will not match input string `"abcd"`. To match
`"abcd"`, you need to use e.g. `"abc.*"` as your regex.
The second argument sets whether the matching should be case-sensitive
or not. By default, it is case-sensitive.
> `std::string` matchers live in `catch2/matchers/catch_matchers_string.h`
### Vector matchers
Catch2 currently provides 5 built-in matchers that work on `std::vector`.
_Vector matchers have been deprecated in favour of the generic
range matchers with the same functionality._
Catch2 provides 5 built-in matchers that work on `std::vector`.
These are
* `Contains` which checks whether a specified vector is present in the result
@ -81,40 +127,82 @@ These are
* `Approx` which checks whether the result is "approx-equal" (order matters, but comparison is done via `Approx`) to a specific vector
> Approx matcher was [introduced](https://github.com/catchorg/Catch2/issues/1499) in Catch 2.7.2.
An example usage:
```cpp
std::vector<int> some_vec{ 1, 2, 3 };
REQUIRE_THAT(some_vec, Catch::Matchers::UnorderedEquals(std::vector<int>{ 3, 2, 1 }));
```
This assertions will pass, because the elements given to the matchers
are a permutation of the ones in `some_vec`.
> vector matchers live in `catch2/matchers/catch_matchers_vector.h`
### Floating point matchers
Catch2 provides 3 matchers for working with floating point numbers. These
are `WithinAbsMatcher`, `WithinUlpsMatcher` and `WithinRelMatcher`.
The `WithinAbsMatcher` matcher accepts floating point numbers that are
within a certain distance of target. It should be constructed with the
`WithinAbs(double target, double margin)` helper.
Catch2 provides 3 matchers that target floating point numbers. These
are:
The `WithinUlpsMatcher` matcher accepts floating point numbers that are
within a certain number of [ULPs](https://en.wikipedia.org/wiki/Unit_in_the_last_place)
of the target. Because ULP comparisons need to be done differently for
`float`s and for `double`s, there are two overloads of the helpers for
this matcher, `WithinULP(float target, int64_t ULPs)`, and
`WithinULP(double target, int64_t ULPs)`.
The `WithinRelMatcher` matcher accepts floating point numbers that are
_approximately equal_ with the target number with some specific tolerance.
In other words, it checks that `|lhs - rhs| <= epsilon * max(|lhs|, |rhs|)`,
with special casing for `INFINITY` and `NaN`. There are _4_ overloads of
the helpers for this matcher, `WithinRel(double target, double margin)`,
`WithinRel(float target, float margin)`, `WithinRel(double target)`, and
`WithinRel(float target)`. The latter two provide a default epsilon of
machine epsilon * 100.
* `WithinAbs(double target, double margin)`,
* `WithinUlps(FloatingPoint target, uint64_t maxUlpDiff)`, and
* `WithinRel(FloatingPoint target, FloatingPoint eps)`.
> `WithinRel` matcher was introduced in Catch 2.10.0
### Generic matchers
Catch also aims to provide a set of generic matchers. Currently this set
contains only a matcher that takes arbitrary callable predicate and applies
it onto the provided object.
Because of type inference limitations, the argument type of the predicate
has to be provided explicitly. Example:
`WithinAbs` creates a matcher that accepts floating point numbers whose
difference with `target` is less than the `margin`.
`WithinULP` creates a matcher that accepts floating point numbers that
are no more than `maxUlpDiff`
[ULPs](https://en.wikipedia.org/wiki/Unit_in_the_last_place)
away from the `target` value. The short version of what this means
is that there is no more than `maxUlpDiff - 1` representeable floating
point numbers between the argument for matching and the `target` value.
`WithinRel` creates a matcher that accepts floating point numbers that
are _approximately equal_ with the `target` with tolerance of `eps.`
Specifically, it matches if
`|arg - target| <= eps * max(|arg|, |target|)` holds. If you do not
specify `eps`, `std::numeric_limits<FloatingPoint>::epsilon * 100`
is used as the default.
In practice, you will often want to combine multiple of these matchers,
together for an assertion, because all 3 options have edge cases where
they behave differently than you would expect. As an example, under
the `WithinRel` matcher, a `0.` only ever matches a `0.` (or `-0.`),
regardless of the relative tolerance specified. Thus, if you want to
handle numbers that are "close enough to 0 to be 0", you have to combine
it with the `WithinAbs` matcher.
For example, to check that our computation matches known good value
within 0.1%, or is close enough (no different to 5 decimal places)
to zero, we would write this assertion:
```cpp
REQUIRE_THAT( computation(input),
Catch::Matchers::WithinRel(expected, 0.001)
|| Catch::Matchers::WithinAbs(0, 0.000001) );
```
> floating point matchers live in `catch2/matchers/catch_matchers_floating.h`
### Miscellaneous matchers
Catch2 also provides some matchers and matcher utilities that do not
quite fit into other categories.
The first one of them is the `Predicate(Callable pred, std::string description)`
matcher. It creates a matcher object that calls `pred` for the provided
argument. The `description` argument allows users to set what the
resulting matcher should self-describe as if required.
Do note that you will need to explicitly specify the type of the
argument, like in this example:
```cpp
REQUIRE_THAT("Hello olleH",
Predicate<std::string>(
@ -123,84 +211,175 @@ REQUIRE_THAT("Hello olleH",
);
```
The second argument is an optional description of the predicate, and is
used only during reporting of the result.
> the predicate matcher lives in `catch2/matchers/catch_matchers_generic.hpp`
### Exception matchers
Catch2 also provides an exception matcher that can be used to verify
that an exception's message exactly matches desired string. The matcher
is `ExceptionMessageMatcher`, and we also provide a helper function
`Message`.
The other miscellaneous matcher utility is exception matching.
The matched exception must publicly derive from `std::exception` and
the message matching is done _exactly_, including case.
> `ExceptionMessageMatcher` was introduced in Catch 2.10.0
#### Matching exceptions
Catch2 provides an utility macro for asserting that an expression
throws exception of specific type, and that the exception has desired
properties. The macro is `REQUIRE_THROWS_MATCHES(expr, ExceptionType, Matcher)`.
> `REQUIRE_THROWS_MATCHES` macro lives in `catch2/matchers/catch_matchers.h`
Catch2 currently provides only one matcher for exceptions,
`Message(std::string message)`. `Message` checks that the exception's
message, as returned from `what` is exactly equal to `message`.
Example use:
```cpp
REQUIRE_THROWS_MATCHES(throwsDerivedException(), DerivedException, Message("DerivedException::what"));
```
## Custom matchers
It's easy to provide your own matchers to extend Catch or just to work with your own types.
Note that `DerivedException` in the example above has to derive from
`std::exception` for the example to work.
You need to provide two things:
1. A matcher class, derived from `Catch::MatcherBase<T>` - where `T` is the type being tested.
The constructor takes and stores any arguments needed (e.g. something to compare against) and you must
override two methods: `match()` and `describe()`.
2. A simple builder function. This is what is actually called from the test code and allows overloading.
> the exception message matcher lives in `catch2/matchers/catch_matchers_exception.hpp`
Here's an example for asserting that an integer falls within a given range
(note that it is all inline for the sake of keeping the example short):
## Writing custom matchers (old style)
The old style of writing matchers has been introduced back in Catch
Classic. To create an old-style matcher, you have to create your own
type that derives from `Catch::Matchers::MatcherBase<ArgT>`, where
`ArgT` is the type your matcher works for. Your type has to override
two methods, `bool match(ArgT const&) const`,
and `std::string describe() const`.
As the name suggests, `match` decides whether the provided argument
is matched (accepted) by the matcher. `describe` then provides a
human-oriented description of what the matcher does.
We also recommend that you create factory function, just like Catch2
does, but that is mostly useful for template argument deduction for
templated matchers (assuming you do not have CTAD available).
To combine these into an example, let's say that you want to write
a matcher that decides whether the provided argument is a number
within certain range. We will call it `IsBetweenMatcher<T>`:
```c++
// The matcher class
class IntRange : public Catch::MatcherBase<int> {
int m_begin, m_end;
public:
IntRange( int begin, int end ) : m_begin( begin ), m_end( end ) {}
#include <catch2/catch_test_macros.hpp>
#include <catch2/matchers/catch_matchers.h>
// ...
// Performs the test for this matcher
bool match( int const& i ) const override {
return i >= m_begin && i <= m_end;
template <typename T>
class IsBetweenMatcher : public Catch::Matchers::MatcherBase<T> {
T m_begin, m_end;
public:
IsBetweenMatcher(T begin, T end) : m_begin(begin), m_end(end) {}
bool match(T const& in) const override {
return in >= m_begin && in <= m_end;
}
// Produces a string describing what this matcher does. It should
// include any provided data (the begin/ end in this case) and
// be written as if it were stating a fact (in the output it will be
// preceded by the value under test).
virtual std::string describe() const override {
std::string describe() const override {
std::ostringstream ss;
ss << "is between " << m_begin << " and " << m_end;
return ss.str();
}
};
// The builder function
inline IntRange IsBetween( int begin, int end ) {
return IntRange( begin, end );
template <typename T>
IsBetweenMatcher<T> IsBetween(T begin, T end) {
return { begin, end };
}
// ...
// Usage
TEST_CASE("Integers are within a range")
{
CHECK_THAT( 3, IsBetween( 1, 10 ) );
TEST_CASE("Numbers are within range") {
// infers `double` for the argument type of the matcher
CHECK_THAT(3., IsBetween(1., 10.));
// infers `int` for the argument type of the matcher
CHECK_THAT(100, IsBetween(1, 10));
}
```
Running this test gives the following in the console:
Obviously, the code above can be improved somewhat, for example you
might want to `static_assert` over the fact that `T` is an arithmetic
type... or generalize the matcher to cover any type for which the user
can provide a comparison function object.
Note that while any matcher written using the old style can also be
written using the new style, combining old style matchers should
generally compile faster. Also note that you can combine old and new
style matchers arbitrarily.
> `MatcherBase` lives in `catch2/matchers/catch_matchers.h`
## Writing custom matchers (new style)
The new style of writing matchers has been introduced in Catch2 v3.
To create a new-style matcher, you have to create your own type that
derives from `Catch::Matchers::MatcherGenericBase`. Your type has to
also provide two methods, `bool match( ... ) const` and overriden
`std::string describe() const`.
Unlike with old-style matchers, there are no requirements on how
the `match` member function takes its argument. This means that the
argument can be taken by value or by mutating reference, but also that
the matcher's `match` member function can be templated.
This allows you to write more complex matcher, such as a matcher that
can compare one range-like (something that responds to `begin` and
`end`) object to another, like in the following example:
```cpp
#include <catch2/catch_test_macros.hpp>
#include <catch2/matchers/catch_matchers_templates.hpp>
// ...
template<typename Range>
struct EqualsRangeMatcher : Catch::Matchers::MatcherGenericBase {
EqualsRangeMatcher(Range const& range):
range{ range }
{}
template<typename OtherRange>
bool match(OtherRange const& other) const {
using std::begin; using std::end;
return std::equal(begin(range), end(range), begin(other), end(other));
}
std::string describe() const override {
return "Equals: " + Catch::rangeToString(range);
}
private:
Range const& range;
};
template<typename Range>
auto EqualsRange(const Range& range) -> EqualsRangeMatcher<Range> {
return EqualsRangeMatcher<Range>{range};
}
TEST_CASE("Combining templated matchers", "[matchers][templated]") {
std::array<int, 3> container{{ 1,2,3 }};
std::array<int, 3> a{{ 1,2,3 }};
std::vector<int> b{ 0,1,2 };
std::list<int> c{ 4,5,6 };
REQUIRE_THAT(container, EqualsRange(a) || EqualsRange(b) || EqualsRange(c));
}
```
/**/TestFile.cpp:123: FAILED:
CHECK_THAT( 100, IsBetween( 1, 10 ) )
with expansion:
100 is between 1 and 10
```
Do note that while you can rewrite any matcher from the old style to
a new style matcher, combining new style matchers is more expensive
in terms of compilation time. Also note that you can combine old style
and new style matchers arbitrarily.
> `MatcherGenericBase` lives in `catch2/matchers/catch_matchers_templates.hpp`
---

View File

@ -61,6 +61,13 @@
* `ListeningReporter` is now final
### Improvements
* Matchers have been extended with the ability to use different signatures of `match` (#1307, #1553, #1554, #1843)
* This includes having templated `match` member function
* See the [rewritten Matchers documentation](matchers.md#top) for details
### Fixes
* The `INFO` macro no longer contains superfluous semicolon (#1456)
* The `--list*` family of command line flags now return 0 on success (#1410, #1146)