👋🙋♂️🙋♀️ Hello and Welcome to my C++20 Cucumber interpreter 🥒🚀✨
- Getting Started
- Implementing Steps
- Custom Parameter Types
- Catalog
- Disclaimer
- Found A Bug? Need A Feature?
Tested compilers: GCC 13, Clang17 and MSVC 19
Conan (2.x) Recipe: https://github.com/ThoSe1990/cwt-cucumber-conan. I didn't push this project to conancenter yet, but you can just create the Conan package locally:
git clone https://github.com/ThoSe1990/cwt-cucumber-conan.git
cd package
conan create . --version 2.4 --user cwt --channel stable
Thanks to Jörg Kreuzberger, who has contributed and tested a lot lately, which has improved this project a lot.
Let us start with Cucumber. First of all build the project on your machine with CMake (you can use the Docker from this directory if you want):
cmake -S . -B ./build
cmake --build ./build -j12
Provide nlohmann-json 3.10.5
, in order to enable the Catalog as json export. You can do so yourself or use conan 2.x
, so that CMake's find_package(nlohmann_json)
succeeds:
conan install . -of ./build --build missing
cmake -S . -B ./build -DCMAKE_TOOLCHAIN_FILE="./build/conan_toolchain.cmake"
cmake --build ./build -j12
Now, consider a simple box where we can put items. To run Cucumber tests, we create a scenario:
Feature: My first feature
This is my cucumber-cpp hello world
Scenario: First Scenario
Given An empty box
When I place 2 x "apple" in it
Then The box contains 2 items
Now execute the first example from the example directory:
./build/bin/example ./examples/features/1_first_scenario.feature
Feature: My first feature ./examples/features/1_first_scenario.feature:2
Scenario: First Scenario ./examples/features/1_first_scenario.feature:5
[ PASSED ] An empty box ./examples/features/1_first_scenario.feature:6
[ PASSED ] I place 2 x "apple" in it ./examples/features/1_first_scenario.feature:7
[ PASSED ] The box contains 2 items ./examples/features/1_first_scenario.feature:8
1 Scenarios (1 passed)
3 Steps (3 passed)
Include cucumber.hpp
and start implementing your steps. In this example, we will use a simple box class, which should be self-explanatory:
#include "cucumber.hpp"
#include "box.hpp"
GIVEN(init_box, "An empty box")
{
const box& my_box = cuke::context<box>();
cuke::equal(my_box.items_count(), 0);
}
WHEN(add_item, "I place {int} x {string} in it")
{
const std::size_t count = CUKE_ARG(1);
const std::string item = CUKE_ARG(2);
cuke::context<box>().add_items(item, count);
}
THEN(check_box_size, "The box contains {int} items")
{
const int items_count = CUKE_ARG(1);
const box& my_box = cuke::context<box>();
cuke::equal(my_box.items_count(), items_count);
}
If custom error messages for steps are necessary, we can append a string which replaces the default error message:
THEN(check_box_size, "The box contains {int} items")
{
const int items_count = CUKE_ARG(1);
const box& my_box = cuke::context<box>();
cuke::equal(my_box.items_count(), items_count, "Items count is wrong at this point!");
}
We can use the backslash between two words and parenthesis around words or characters in order to use alternative words in step definitions. Consider Following step:
THEN(alternative_words, "{int} item(s) is/are {string}")
{
const std::size_t count = CUKE_ARG(1);
const std::string item = CUKE_ARG(2);
cuke::equal(count, cuke::context<box>().count(item));
}
And now I can write this Scenario:
Scenario: Alternative Words
Given An empty box
When I place 1 x "banana" in it
Then 1 item is "banana"
And I place 1 x "banana" in it
Then 2 items are "banana"
To implement a step, use a define with a function name and a step name. There is no technical difference between these definitions:
GIVEN(function_name, "your step goes here")
WHEN(function_name, "your step goes here")
THEN(function_name, "your step goes here")
- Alternatively just use
STEP(function_name, "your step goes here")
Use CUKE_ARG(..)
to access a value in a step. Start with index 1
for the first value. I have overloaded the underlying function with the implicit conversion operator. So always declare a variable to access values:
STEP(your_function, "A {string}, an {int} and a {float}")
{
std::string str = CUKE_ARG(1);
int i = CUKE_ARG(2);
float f = CUKE_ARG(3);
// ...
}
The values are implemted as Cucumber expressions and currently supported are:
- Integer values:
{byte}
,{short}
,{int}
,{long}
, - Floating point values:
{float}
,{double}
- Strings (in quotes
".."
{string}
- Word (w/o quotes)
{word}
-> read value asstd::string
in step definition - Anonymous, any character sequence
{}
-> read value asstd::string
in step definition - Custom Parameter Types:
{your type comes here}
-> check Custom Parameter Types for examples and how to use them
The scenario context or cuke::context
stores an object for the duration of a scenario. After the scenario is finished, the object is destroyed. One instance of each type can be stored in the scenario context.
Recall these steps from the first example:
WHEN(add_item, "I place {int} x {string} in it")
{
const std::size_t count = CUKE_ARG(1);
const std::string item = CUKE_ARG(2);
cuke::context<box>().add_items(item, count);
}
THEN(check_box_size, "The box contains {int} items")
{
const int items_count = CUKE_ARG(1);
const box& my_box = cuke::context<box>();
cuke::equal(my_box.items_count(), items_count);
}
When initializing your object, you can pass all arguments to the context:
// forwards 1,2,3 to your object:
cuke::context<some_object>(1,2,3);
// access or default initialize your object:
cuke::context<some_object>();
In a scenario outline you can define variables and run a scenario with different values:
Feature: My first feature
Scenario Outline: First Scenario Outline
Given An empty box
When I place <count> x <item> in it
Then The box contains <count> items
Examples:
| count | item |
| 1 | "apple" |
| 2 | "bananas" |
Which gives this output:
./build/bin/example ./examples/features/2_scenario_outline.feature
Feature: My first feature ./examples/features/2_scenario_outline.feature:2
Scenario Outline: First Scenario Outline ./examples/features/2_scenario_outline.feature:5
[ PASSED ] An empty box ./examples/features/2_scenario_outline.feature:6
[ PASSED ] I place 1 x "apple" in it ./examples/features/2_scenario_outline.feature:7
[ PASSED ] The box contains 1 items ./examples/features/2_scenario_outline.feature:8
Scenario Outline: First Scenario Outline ./examples/features/2_scenario_outline.feature:5
[ PASSED ] An empty box ./examples/features/2_scenario_outline.feature:6
[ PASSED ] I place 2 x "bananas" in it ./examples/features/2_scenario_outline.feature:7
[ PASSED ] The box contains 2 items ./examples/features/2_scenario_outline.feature:8
2 Scenarios (2 passed)
6 Steps (6 passed)
A background is a set of steps (or a single step) which are the first steps of every Scenario
in a Feature
. After the feature definition add Background
, see ./examples/features/3_background.feature
:
Feature: We always need apples!
Background: Add an apple
Given An empty box
When I place 1 x "apple" in it
Scenario: Apples Apples Apples
When I place 1 x "apple" in it
Then The box contains 2 items
Scenario: Apples and Bananas
When I place 1 x "apple" in it
And I place 1 x "banana" in it
Then The box contains 3 items
Append a doc string to a defined step in your feature file:
Scenario: Doc string with quotes
When There is a doc string:
"""
This is a docstring with quotes
after a step
"""
There is no parameter in the step definition needed.
To access a doc string use CUKE_DOC_STRING()
:
// There is no parameter needed in your step
WHEN(doc_string, "There is a doc string:")
{
const std::string& str = CUKE_DOC_STRING();
// or get each line in a vector:
const std::vector<std::string>& v = CUKE_DOC_STRING();
}
Similar to doc strings, you can append tables to a defined step. Then there are three different options to access the values. To create a table in your step definition use:
const cuke::table& t = CUKE_TABLE();
or as copycuke::table t = CUKE_TABLE();
You can directly access the elements with the operator[]
. But the underlying value is a cuke::value
which you have to cast accordingly with as
or to _string
.
const cuke::table& t = CUKE_TABLE();
t[0][0].to_string();
t[0][1].as<int>();
// ...
First we look at a raw table. This means there is no header line or identifiers for the given values:
Scenario: Adding items with raw
Given An empty box
When I add all items with the raw function:
| apple | 2 |
| strawberry | 3 |
| banana | 5 |
Then The box contains 10 items
You can iterate over this table with raw()
:
WHEN(add_table_raw, "I add all items with the raw function:")
{
// create a table
const cuke::table& t = CUKE_TABLE();
// with the raw function you iterate over all rows
for (const auto& row : t.raw())
{
// and with the operator[] you get access to each cell in each row
cuke::context<box>().add_items(row[0].to_string(), row[1].as<long>());
}
}
With an additional header in the table we can make this table more descriptive:
Scenario: Adding items with hashes
Given An empty box
When I add all items with the hashes function:
| ITEM | QUANTITY |
| apple | 3 |
| banana | 6 |
Then The box contains 9 items
You can now iterate over the table using hashes()
and access the elements with string literals:
WHEN(add_table_hashes, "I add all items with the hashes function:")
{
const cuke::table& t = CUKE_TABLE();
for (const auto& row : t.hashes())
{
cuke::context<box>().add_items(row["ITEM"].to_string(), row["QUANTITY"].as<long>());
}
}
Another more descriptive way works for key value pairs, or rows hash. The first column describes the element, the second holds the element:
Scenario: Adding items with rows_hash
Given An empty box
When I add the following item with the rows_hash function:
| ITEM | really good apples |
| QUANTITY | 3 |
Then The box contains 3 items
And with cuke::table::pair hash_rows = t.rows_hash();
you can create this hash map. The access to each element is again by the string literal.
WHEN(add_table_rows_hash, "I add the following item with the rows_hash function:")
{
const cuke::table& t = CUKE_TABLE();
// cuke::table::pair is just an alias for a std::unordered_map which gets created in rows.hash()
cuke::table::pair hash_rows = t.rows_hash();
cuke::context<box>().add_items(hash_rows["ITEM"].to_string(), hash_rows["QUANTITY"].as<long>());
}
Use the terminal option -t
or --tags
to provide tags. This will then check the given condition with tagged scenario and execute them accordingly. Consider this feature file:
Feature: Scenarios with tags
@apples @bananas
Scenario: Apples and Bananas
Given An empty box
When I place 2 x "apple" in it
And I place 2 x "banana" in it
Then The box contains 4 items
Scenario Outline:
Given An empty box
When I place <count> x <fruit> in it
Then The box contains <expected> items
@oranges
Examples:
| count | fruit | expected |
| 1 | "orange" | 1 |
@oranges @others
Examples:
| count | fruit | expected |
| 3 | "oranges" | 3 |
| 1 | "some stuff" | 1 |
And when we run this with tags, we can control which scenarios are executed.
This executes both scenarios:
./build/bin/example ./examples/features/4_tags.feature -t "@apples or @bananas"
And this would just execute the second scenario due to the and
condition:
./build/bin/example ./examples/features/4_tags.feature -t "@apples and @bananas"
Note: Tags can be applied to Feature:
, Scenario:
, Scenario Outline:
and Examples:
. The tags are inherited to the next child.
Hooks are executed before and after each scenario or step. The implementation is pretty straightforward. You can have multiple hooks of the same type. All of them are executed at their time.
// ./examples/hooks.cpp:
BEFORE(before)
{
// this runs before every scenario
}
AFTER(after)
{
// this runs after every scenario
}
BEFORE_STEP(before_step)
{
// this runs before every step
}
AFTER_STEP(after_step)
{
// this runs after every step
}
BEFORE_ALL(before_all)
{
// this runs at program start
}
AFTER_ALL(after_all)
{
// this runs at program end
}
You can add a tag expression to your hook. Use
BEFORE_T(name, "tags come here")
for a tagged hook before a scenrioAFTER_T(name, "tags come here")
for a tagged hook after a scenario
This means a tagged hook is executed, when a scenario fulfills the given condition. You can pass in any logical expression to a tagged hook:
// ./examples/hooks.cpp
AFTER_T(dispatch_box, "@ship or @important")
{
std::cout << "The box is shipped!" << std::endl;
}
Feature: Scenarios with tags
@ship
Scenario: We want to ship cucumbers
Given An empty box
When I place 1 x "cucumber" in it
Then The box contains 1 items
@important
Scenario: Important items must be shipped immediately
Given An empty box
When I place 2 x "important items" in it
Then The box contains 2 items
And now we can see that our box was shipped:
./build/bin/example ./examples/features/5_tagged_hooks.feature
Feature: Scenarios with tags ./examples/features/5_tagged_hooks.feature:1
Scenario: We want to ship cucumbers ./examples/features/5_tagged_hooks.feature:4
[ PASSED ] An empty box ./examples/features/5_tagged_hooks.feature:5
[ PASSED ] I place 1 x "cucumber" in it ./examples/features/5_tagged_hooks.feature:6
[ PASSED ] The box contains 1 items ./examples/features/5_tagged_hooks.feature:7
The box is shipped!
Scenario: Important items must be shipped immediately ./examples/features/5_tagged_hooks.feature:10
[ PASSED ] An empty box ./examples/features/5_tagged_hooks.feature:11
[ PASSED ] I place 2 x "important items" in it ./examples/features/5_tagged_hooks.feature:12
[ PASSED ] The box contains 2 items ./examples/features/5_tagged_hooks.feature:13
The box is shipped!
2 Scenarios (2 passed)
6 Steps (6 passed)
We can use cuke::skip_scenario()
inside a hook BEFORE_T
in order to skip a Scenario
. This can be useful when using a @skip
, @ignore
or @wip
tag:
BEFORE_T(skip, "@skip")
{
cuke::skip_scenario();
}
Note: The hook AFTER_ALL
still will be executed. The hook AFTER
the skipped Scenario
is not called.
If you only want to run single scenarios, you can append the appropriate line to the feature file:
This runs a Scenario in Line 6:
./build/bin/example ./examples/features/box.feature:6
This runs each Scenario in line 6, 11, 14:
./build/bin/example ./examples/features/box.feature:6:11:14
If you want to execute all feature files in a directory (and subdirectory), just pass the directory as argument:
./build/bin/example ./examples/features
CWT-Cucumber supports custom parameter types. This means that we can define custom expressions in our steps to introduce custom types (so to speak) and make the step definition more understandable.
In general: A custom parameter type is an individually defined type that we can use in the step definition. So we give the parameter a function name (as in a step), give the custom type a meaningful name, a description and a regex pattern. Then we implement a callback to consume the capture groups from the regex pattern. Here we use CUSTOM_PARAMETER(function-name, "{here goes your type}", "regex pattern", "description")
.
- Function-name: A defined function name (same as in steps)
- Custom-Type: Define the type you want, with curly braces as string
- Regex-Pattern: The regex pattern to match the step, you can use raw string literals, which makes it easier to write regex pattern (see below)
- Description: A string value to give a meaning full description. This will be printed to the catalog and has no effect on the scenarios.
In order to access the capture groups, use CUKE_PARAM_ARG(index)
where the index starts at 1 from left to right.
Note: You must explicitly return the dedicated type in the callback. The implementation uses type erasure and does not know which type will be used later.
Find all implementations in examples/step_definition.cpp
and the examples in examples/features/8_custom_parameters.feature
.
Lets define a type {pair of integers}
, which will create a std::pair<int,int>
:
CUSTOM_PARAMETER(custom, "{pair of integers}", R"(var1=(\d+), var2=(\d+))", "a pair of integers")
{
int var1 = CUKE_PARAM_ARG(1);
int var2 = CUKE_PARAM_ARG(2);
return std::make_pair(var1, var2);
}
And now we can use this type in our step definition. And note this type no uses a single index with CUKE_ARG(1)
and returns a std::pair<int,int>
:
WHEN(custom_par_when, "this is {pair of integers}")
{
std::pair<int, int> p = CUKE_ARG(1);
cuke::context<const std::pair<int, int>>(p);
}
And a possible scenario looks like this:
Scenario: An example
When this is var1=123, var2=99
Then their values are 123 and 99
A more complex example is defined below. We want to parse an event (which is represented as string) and two dates. Therefore we add following regex patterns:
- Event in single quotes:
'(.*?)'
and - The dates in the format:
from November 11, 2024 to December 12, 2024
:from ([A-Za-z]+) (\d{1,2}), (\d{4}) to ([A-Za-z]+) (\d{1,2}), (\d{4})
.
{event}
is fairly easy here:
CUSTOM_PARAMETER(custom_event, "{event}", R"('(.*?)')", "a custom event")
{
return CUKE_PARAM_ARG(1).to_string();
}
{date}
requires two structs and in with it we want to date_range
:
struct date
{
int day;
std::string month;
int year;
};
struct date_range
{
date start;
date end;
};
In order to create the date_range
we implement now the custom parameter:
CUSTOM_PARAMETER(
custom_date,
"{date}",
R"(from ([A-Za-z]+) (\d{1,2}), (\d{4}) to ([A-Za-z]+) (\d{1,2}), (\d{4}))",
"a custom date pattern")
{
date begin;
begin.month = CUKE_PARAM_ARG(1).to_string();
begin.day = int(CUKE_PARAM_ARG(2));
begin.year = CUKE_PARAM_ARG(3).as<int>();
date end;
end.month = CUKE_PARAM_ARG(4).to_string();
end.day = static_cast<int>(CUKE_PARAM_ARG(5));
end.year = CUKE_PARAM_ARG(6).as<int>();
return date_range{begin, end};
}
Note: When using structs, we have to be explicit about the types. You can either use static_casts or bracket initialization or you call .as<..>()
on CUKE_PARAM_ARG
.
And this is now our step we want to use, where we put an arbitrary string in front to describe the date range:
WHEN(using_date, "{event} is {date}")
{
std::string event = CUKE_ARG(1);
date_range dr = CUKE_ARG(2);
cuke::context<date_range>(dr);
}
Note: that the first argument here returns a string where we get the event and the second argument is date_range
.
Now we can use this for a Scenario (check the full implementation in examples/step_definition.cpp
):
Scenario: Date example
When 'The public festival in town' is from April 25, 2025 to Mai 13, 2025
Then The beginning month is April and the ending month is Mai
There are two options in order to create a string value. Some compiler have problems with can not find the correct string type. Therefore we have two options to create a string value from a regex capture:
Option 1: Dedicated to_string()
function:
CUKE_PARAM_ARG(..).to_string();
Option 2: Initialize a std::string
std::string str = CUKE_PARAM_ARG(..)
Run the compiled executable with the option --steps-catalog
or --steps-catalog-json
in order to print all implemented steps to stdout or a file. To print to a file just append a filepath to --steps-catalog
. This does not write to .feature
files and overwrites existing files.
./build/bin/example --steps-catalog
Step Definitions (catalog):
---------------------------
Given An empty box
When A {word} and {}
When I place {int} x {string} in it
When There is a doc string:
When There is a doc string as vector:
When I add all items with the raw function:
When I add all items with the hashes function:
When I add the following item with the rows_hash function:
Then They will match {string} and {string}
Then The {int}. item is {string}
Then The box contains {int} item(s)
Then {int} item(s) is/are {string}
Step doc string:
./build/bin/example --steps-catalog-json
{
"steps_catalog": [
{
"definition": "An empty box",
"type": "Given",
"var_types": []
},
{
"definition": "A {word} and {}",
"type": "When",
"var_types": [
"word",
"anonymous"
]
},
{
"definition": "I place {int} x {string} in it",
"type": "When",
"var_types": [
"int",
"string"
]
},
{
"definition": "There is a doc string:",
"type": "When",
"var_types": []
},
{
"definition": "There is a doc string as vector:",
"type": "When",
"var_types": []
},
{
"definition": "I add all items with the raw function:",
"type": "When",
"var_types": []
},
{
"definition": "I add all items with the hashes function:",
"type": "When",
"var_types": []
},
{
"definition": "I add the following item with the rows_hash function:",
"type": "When",
"var_types": []
},
{
"definition": "They will match {string} and {string}",
"type": "Then",
"var_types": [
"string",
"string"
]
},
{
"definition": "The {int}. item is {string}",
"type": "Then",
"var_types": [
"int",
"string"
]
},
{
"definition": "The box contains {int} item(s)",
"type": "Then",
"var_types": [
"int"
]
},
{
"definition": "{int} item(s) is/are {string}",
"type": "Then",
"var_types": [
"int",
"string"
]
},
{
"definition": "doc string:",
"type": "Step",
"var_types": []
}
]
}
This is a fun/educational project for me. After reading "Crafting Interpreters" by Robert Nystorm, I wanted a meaningful, not too complex example. Since I really like and use Cucumber, this seems like a perfect project to me. I'm not getting paid for this and I do all the implementation in my free time aside from work.
Don't hesitate and open an Issue.
Cheers 🍻
2024 Coding with Thomas
https://www.codingwiththomas.com/