Features and design goals
doctest has been designed from the start to be as light and unintrusive as possible. These key features should be kept.
It is modeled after Catch which is currently the most popular alternative for testing in C++ - check out the differences.
Unintrusive (transparent):
- everything testing-related can be removed from the binary executable by defining the
DOCTEST_CONFIG_DISABLE
identifier - very small and easy to integrate - single header
- Extremely low footprint on compile times - below 10ms of compile time overhead for including the header in a file
- The fastest possible assertion macros - 50k asserts can compile for under 30 seconds (even under 10 sec)
- doesn't drag any headers when included (except for in the translation unit where the library gets implemented)
- everything is in the
doctest
namespace (and the implementation details are in a nested detail
namespace) - all macros have prefixes - some by default have unprefixed versions as well but that is optional - see configuration
- 0 warnings even with the most aggresive flags (on all tested compilers!!!)
-Weverything -pedantic
for clang-Wall -Wextra -pedantic
and >> over 35 << other warnings not covered by these flags for GCC!!! - see here/W4
for MSVC (/Wall
is too much there - even their own headers produce thousands of warnings with that option)
- doesn't error on unrecognized command line options and supports prefixes for interop with client command line parsing
- can set options procedurally and not deal with passing
argc
/argv
from the command line - doesn't leave warnings disabled after itself
Extremely portable:
- Standards compliant C++98 code - should work with any C++98 compiler
- tested with GCC: 4.4, 4.5, 4.6, 4.7, 4.8, 4.9, 5, 6
- tested with Clang: 3.4, 3.5, 3.6, 3.7, 3.8, 3.9
- tested with MSVC: 2008, 2010, 2012, 2013, 2015, 2017
- per-commit tested on travis and appveyor CI services
- warnings as errors even on the most aggressive warning levels - see here
- all tests have their output compared to reference output of a previous known good run
- all tests built and ran in Debug/Release and also in 32/64 bit modes
- all tests ran through valgrind under Linux/OSX
- all tests ran through address and UB sanitizers under Linux/OSX
- tests are ran in more than 300 different configurations on UNIX (Linux + OSX) on travis CI
- tests are ran in a total of 24 different configurations on Windows on appveyor CI
This allows the library to be used in more ways than any other - tests can be written directly in the production code!
- This makes the barrier for writing tests much lower - you don't have to: 1. make a separate source file 2. include a bunch of stuff in it 3. add it to the build system and 4. add it to source control - You can just write the tests for a class or a piece of functionality at the bottom of its source file - or even header file!
- Tests in the production code can be thought of as documentation or up-to-date comments - showing how an API is used
- Testing internals that are not exposed through the public API and headers becomes easier!
- Test-driven development in C++ has never been easier!
The library can be used like any other if you don't like the idea of mixing production code and tests.
Other features:
- really easy to get started - it's just 1 header file - see the tutorial
- very light, unintrusive and portable - see the sections above - and also the benchmarks
- offers a way to remove everything testing-related from the binary with the
DOCTEST_CONFIG_DISABLE
macro - tests are registered automatically - no need to add them to a collection manually
- supports subcases for easy setup/teardown of tests (also supports the retro test fixtures with classes)
- supports logging macros for capturing local variables and strings - as a message for when an assert fails - with lazy stringification and no allocations when possible!
- output from all compilers on all platforms is the same - byte by byte
- binaries (exe/dll) can use the test runner of another binary - so tests end up in a single registry - example
- supports BDD style tests
- only one core assertion macro for comparisons - standard C++ operators are used for the comparison (less than, equal, greater than...) - yet the full expression is decomposed and left and right values of the expression are logged
- assertion macros for exceptions - if something should or shouldn't throw
- floating point comparison support - see the
Approx()
helper - powerful mechanism for stringification of user types - including exceptions!
- tests can be grouped in test suites
- can be used without exceptions and rtti - checkout
DOCTEST_CONFIG_NO_EXCEPTIONS
- powerful command line with lots of options
- tests can be filtered based on their name/file/test suite using wildcards
- can filter subcases using wildcards and by specifying the nesting levels for which those filters should work
- failures can (optionally) break into the debugger on Windows and Mac
- integration with the output window of Visual Studio for failing tests
- a
main()
can be provided when implementing the library with the DOCTEST_CONFIG_IMPLEMENT_WITH_MAIN
identifier - can write tests in headers - they will still be registered only once in the executable/shared object
- range-based execution of tests - see the range_based_execution example (the run.py script)
- colored output in the console
- controlling the order of test execution
- different
doctest::Context
s can be created and ran many times within a single execution of the program - ability to query if code is currently being ran in a test -
doctest::isRunningInTest()
There is a list of planned features which are all important and big - see the roadmap.
Home