@ThijsWithaar is responsible for giving me the idea, but his PR
had couple of things that meant it was simpler to rewrite it than
to fix and merge it.
Supersedes and closes#1599
This is because otherwise the installations paths provided via
GNUInstallDirs become messed up and parts of the installation
package will end up in the wrong place.
Also it doesn't make much sense to force dependees to also install
our header alongside them.
Closes#1373
This means
* a new cmake option, `CATCH_BUILD_EXTRA_TESTS`, that conditionally
includes the ExtraTests subfolder
* building and running them on some of the Travis build images
* An example configuration test
In the future these should be extended to cover most of the
configuration options in Catch2, but this is a start.
This also goes for pkg-config installed by our CMake installation.
This includes
* Updating CMake version on Travis
* Adding a `Catch2` subfolder to the `single_include/` folder to
provide this include path both _inside_ the repository, and _outside_.
* Updated examples to build with the new paths
* Other general CMake cleanup
Android apparently does not support `std::to_string`, so we add a
small polyfill over it. Right now only the ULP matcher uses it,
but we have had plans to use it in `StringMaker<int>` and friends,
as it performs a lot better than `std::stringstream` based
stringification on MSVC.
See #1280 for more details
Unlike the relatively non-invasive old way of capturing stdout/stderr,
this new way is also able to capture output from C's stdlib functions
such as `printf`. This is done by redirecting stdout and stderr file
descriptors to a file, and then reading this file back.
This approach has two sizeable drawbacks:
1) Performance, obviously. Previously an installed capture made the
program run faster (as long as it was then discarded), because a call
to `std::cout` did not result in text output to the console. This new
capture method in fact forces disk IO. While it is likely that any
modern OS will keep this file in memory-cache and might never actually
issue the IO to the backing storage, it is still a possibility and
calls to the file system are not free.
2) Nonportability. While POSIX is usually assumed portable, and this
implementation relies only on a very common parts of it, it is no
longer standard C++ (or just plain C) and thus might not be available
on some obscure platforms. Different C libs might also implement the
relevant functions in a less-than-useful ways (e.g. MS's `tmpfile`
generates a temp file inside system folder, so it will not work
without elevated privileges and thus is useless).
These two drawbacks mean that, at least for now, the new capture is
opt-in. To opt-in, `CATCH_CONFIG_EXPERIMENTAL_REDIRECT` needs to be
defined in the implementation file.
Closes#1243
Catch2's documentation promises that listeners are called _before_
reporters, but because of the previous implementation, they were
called _after_ reporters. This commit fixes that.
Closes#1234
The old version would lead to error when Catch was installed
as a subproject. The file would be written to the subproject's
build directory and then would not be installed properly.
Create a namespaced Catch2::Catch target that is 'linkable' through
`target_link_libraries()` and export it so it is findable through
`find_package()`.
`find_package()` will find versions with the same major number and with
minor number >= requested.
This makes catch a lot easier to use in CMake-based projects. Whether it
is found using `find_package` or included in the client project as a
subdirectory, the client can include the catch headers per-target with
`target_include_directories(target PRIVATE Catch2::Catch).
Example usage:
cmake_minimum_required(VERSION 3.1)
# include Catch2 as subdirectory or installed package
# add_subdirectory(Catch2)
find_package(Catch2 VERSION 2.1.0 REQUIRED)
add_executable(tests tests/catch_main.cpp)
target_link_libraries(tests PRIVATE Catch2::Catch)
Also had to add new project to redirect CTest output, add
separate batch scripts for AppVeyor because it doesn't handle
multi-line batch scripts in yaml properly, and other helper
scripts.
* Every Linux build tracks coverage when running Debug mode
* OS X not supported yet (Future WIP)
* Our own unit tests, non-default reporters and Clara are ignored
- Added new compilers and OS X images
- Option to run SelfTest under Valgrind
- Merge "Debug" and "Release" configurations into one run
-- This saves apt setup and cmake download step per compiler, 60-90s
- Fix C++14 compilation under Clang 3.8 and up
to prevent inheritance of include directories that possibly lead to a clash.
A clash occurs when a folder is included, e.g. examples, that wants to use the single-include directory instead of the normal include directory as used by the SelfTest in the next higher level.
Also hides std::chrono, std::pair and std::chrono::* behind
new configuration macros, CATCH_CONFIG_ENABLE_*_STRINGMAKER
to avoid dragging in <utility>, <tuple> and <chrono> in common
path, unless requested.
Swept:
`-Wpadded` in some places (where it caused extra size, instead of just
saying "hey, we padded struct at the end to align, just as standard says")
`-Wweak-vtables` everywhere (Clang)
`-Wexit-time-destructors` everywhere (Clang)
`-Wmissing-noreturn` everywhere (Clang)
The last three are enabled for Clang compilation going forward.
Also enabled `-Wunreachable-code` for Clang and GCC
Note, this doesn't mean it will start passing, just that it will
run the approval tests properly
Some changes are needed before it passes, as the Windows output
somewhat differs.
- this file excluded from the CATCH_CONFIG_DISABLE_MATCHERS path.
- matchers are always compiled in to the impl file
- _THROWS_WITH macros are still available with matchers disabled - but only the ones that take a string
- tests that use matchers have #ifdefs, so the whole SelfTest project can compile with matchers disable.
Martin Hořeňovský