Tags

v2.13.7

--- Fixes ---
* Added missing `<iterator>` include in benchmarking. (#2231)
* Fixed noexcept build with benchmarking enabled (#2235)
* Fixed build for compilers with C++17 support but without C++17 library support (#2195)
* JUnit only uses 3 decimal places when reporting durations (#2221)
* `!mayfail` tagged tests are now marked as `skipped` in JUnit reporter output (#2116)

-- 2.13.6 --

--- Fixes ---
* Disabling all signal handlers no longer breaks compilation  (#2212, #2213)

--- Miscellaneous ---
* `catch_discover_tests` should handle escaped semicolon (`;`) better (#2214, #2215)

2.13.5

--- Improvements ---
* Detection of MAC and IPHONE platforms has been improved (#2140, #2157)
* Added workaround for bug in XLC 16.1.0.1 (#2155)
* Add detection for LCC when it is masquerading as GCC (#2199)
* Modified posix signal handling so it supports newer libcs (#2178)
  * `MINSIGSTKSZ` was no longer usable in constexpr context.

--- Fixes ---
* Fixed compilation of benchmarking when `min` and `max` macros are defined (#2159)
  * Including `windows.h` without `NOMINMAX` remains a really bad idea, don't do it

--- Miscellaneous ---
* `Catch2WithMain` target (static library) is no longer built by default (#2142)
  * Building it by default was at best unnecessary overhead for people not using it, and at worst it caused trouble with install paths
  * To have it built, set CMake option `CATCH_BUILD_STATIC_LIBRARY` to `ON`
* The check whether Catch2 is being built as a subproject is now more reliable (#2202, #2204)
  * The problem was that if the variable name used internally was defined the project including Catch2 as subproject, it would not be properly overwritten for Catch2's CMake.

v2.13.4

--- Improvements
* Improved the hashing algorithm used for shuffling test cases (#2070)
  * `TEST_CASE`s that differ only in the last character should be properly shuffled
  * Note that this means that v2.13.4 gives you a different order of test cases than 2.13.3, even given the same seed.

--- Miscellaneous
* Deprecated `ParseAndAddCatchTests` CMake integration (#2092)
  * It is impossible to implement it properly for all the different test case variants Catch2 provides, and there are better options provided.
  * Use `catch_discover_tests` instead, which uses runtime information about available tests.
* Fixed bug in `catch_discover_tests` that would cause it to fail when used in specific project structures (#2119)
* Added Bazel build file
* Added an experimental static library target to CMake

v2.13.3

--- Fixes
* Fixed possible infinite loop when combining generators with section filter (`-c` option) (#2025)

--- Miscellaneous
* Fixed `ParseAndAddCatchTests` not finding `TEST_CASE`s without tags (#2055, #2056)
* `ParseAndAddCatchTests` supports `CMP0110` policy for changing behaviour of `add_test` (#2057)
  * This was the shortlived change in CMake 3.18.0 that temporarily broke `ParseAndAddCatchTests`

v3.0.0-preview3

--- Changes compared to preview 2

* Incorporated v2 changes up to v2.13.2
* Some runtime performance improvements
* The static library compiles 20-30% faster
* The overhead of including catch_test_macros.hpp is 25-30% smaller
* Amalgamated, 1 .hpp + 1 .cpp, distribution is provided
* Reporter interface has slightly changed
* Listener base renamed
* Rewritten contributing documentation
  * Primary development is now on `devel` branch
  * v3 is now the primary development version

v2.13.2

=== Improvements ===
* Implemented workaround for AppleClang shadowing bug (#2030)
* Implemented workaround for NVCC ICE (#2005, #2027)

=== Fixes ===
* Fixed detection of `std::uncaught_exceptions` support under non-msvc platforms (#2021)
* Fixed the experimental stdout/stderr capture under Windows (#2013)

=== Miscellaneous ===
* `catch_discover_tests` has been improved significantly (#2023, #2039)
  * You can now specify which reporter should be used
  * You can now modify where the output will be written
  * `WORKING_DIRECTORY` setting is respected
* `ParseAndAddCatchTests` now supports `TEMPLATE_TEST_CASE` macros (#2031)
* Various documentation fixes and improvements (#2022, #2028, #2034)

v2.13.1

--- Improvements ---
* `ParseAndAddCatchTests` handles CMake v3.18.0 correctly (#1984)
* Improved autodetection of `std::byte` (#1992)
* Simplified implementation of templated test cases (#2007)
  * This should have a tiny positive effect on its compilation throughput

--- Fixes ---
* Automatic stringification of ranges handles sentinel ranges properly (#2004)

v2.13.0

--- Improvements
* `GENERATE` can now follow a `SECTION` at the same level of nesting (#1938)
  * The `SECTION`(s) before the `GENERATE` will not be run multiple times, the following ones will.
* Added `-D`/`--min-duration` command line flag (#1910)
  * If a test takes longer to finish than the provided value, its name and duration will be printed.
  * This flag is overriden by setting `-d`/`--duration`.

--- Fixes
* `TAPReporter` no longer skips successful assertions (#1983)

v2.12.4

--- Improvements
* Added support for MacOS on ARM (#1971)

v2.12.3

--- Fixes
* `GENERATE` nested in a for loop no longer creates multiple generators (#1913)
* Fixed copy paste error breaking `TEMPLATE_TEST_CASE_SIG` for 6 or more arguments (#1954)
* Fixed potential UB when handling non-ASCII characters in CLI args (#1943)

--- Improvements
* There can be multiple calls to `GENERATE` on a single line
* Improved `fno-except` support for platforms that do not provide shims for exception-related std functions (#1950)
  * E.g. the Green Hills C++ compiler
* XmlReporter now also reports test-case-level statistics (#1958)
  * This is done via a new element, `OverallResultsCases`

--- Miscellaneous
* Added `.clang-format` file to the repo (#1182, #1920)
* Rewrote contributing docs
  * They should explain the different levels of testing and so on much better

v2.12.2

--- Fixes
* Fixed compilation failure if `is_range` ADL found deleted function (#1929)
* Fixed potential UB in `CAPTURE` if the expression contained non-ASCII characters (#1925)

--- Improvements
* `std::result_of` is not used if `std::invoke_result` is available (#1934)
* JUnit reporter writes out `status` attribute for tests (#1899)
* Suppresed clang-tidy's `hicpp-vararg` warning (#1921)
  * Catch2 was already suppressing the `cppcoreguidelines-pro-type-vararg` alias of the warning

v3.0.0-preview2

**Catch2 now uses statically compiled library as its distribution model.
This also means that to get all of Catch2's functionality in a test file,
you have to include multiple headers.**

For quick'n'dirty migration, you can replace the old `-include <catch2/catch.hpp>`
with `-include <catch2/catch_all.hpp>`. This is a (one of) convenience
header(s) that brings in _all_ of headers in Catch2. By doing this,
you should be able to migrate instantly, but at the cost of (significantly)
increased compilation times. You should prefer piecemeal including
headers that are actually required by your test code.

The basic set of functionality (`TEST_CASE`, `SECTION`, `REQUIRE`) is in
`catch2/catch_test_macros.hpp`. Matchers are in `matchers` subfolder,
generators in `generators` subfolder, and so on.

Note that documentation has not yet been updated to account for the
new design.

--- FAQ

* Why is Catch2 moving to separate headers?
  * The short answer is future extensibility and scalability. The long answer is complex and can be found on my blog, but at the most basic level, it is that providing single-header distribution is at odds with providing variety of useful features. When Catch2 was distributed in a single header, adding a new Matcher would cause overhead for everyone, but was useful only to a subset of users. This meant that the barrier to entry for new Matchers/Generators/etc is high in single header model, but much smaller in the new model.
* Will Catch2 again distribute single-header version in the future?
  * No. But I intend to provide sqlite-style distribution option, with 1 header and 1 "unity" .cpp file. Do note that the header will have similar problem to the `catch_all.hpp` header.
* Why the big breaking change caused by replacing `catch.hpp` with `catch_all.hpp`?
  * The convenience header `catch_all.hpp` exists for two reasons. One of them is to provide a way for quick migration from Catch2, the second one is to provide a simple way to test things with Catch2. Using it for migration has one drawback in that it is **big**. This means that including it _will_ cause significant compile time drag, and so using it to migrate should be a concious decision by the user, not something they can just stumble into unknowingly.

--- (Potentially) Breaking changes
* **Catch2 now uses statically compiled library as its distribution model**
  * **Including `catch.hpp` no longer works**
* `ANON_TEST_CASE` has been removed, use `TEST_CASE` with no arguments instead (-1220)
* `--list*` commands no longer have non-zero return code (-1410)
* `--list-test-names-only` has been removed (-1190)
  * You should use verbosity-modifiers for `--list-tests` instead
* `--list*` commands are now piped through the reporters
  * The top-level reporter interface provides default implementation that works just as the old one
  * XmlReporter outputs a machine-parseable XML
* `TEST_CASE` description support has been removed
  * If the second argument has text outside tags, the text will be ignored.
* Hidden test cases are no longer included just because they don't match an exclusion tag
  * Previously, a `TEST_CASE("A", "[.foo]")` would be included by asking for `~[bar]`.
* `PredicateMatcher` is no longer type erased.
  * This means that the type of the provided predicate is part of the `PredicateMatcher`'s type
* `SectionInfo` no longer contains section description as a member (-1319)
  * You can still write `SECTION("ShortName", "Long and wordy description")`, but the description is thrown away
  * The description type now must be a `const char*` or be implicitly convertible to it
* The `[!hide]` tag has been removed.
  * Use `[.]` or `[.foo]` instead.
* Lvalues of composed matchers cannot be composed further
* Uses of `REGISTER_TEST_CASE` macro need to be followed by a semicolon
  * This does not change `TEST_CASE` and friends in any way
* `IStreamingReporter::IsMulti` member function was removed
  * This is _very_ unlikely to actually affect anyone, as it was default-implemented in the interface, and only used internally
* Various classes not designed for user-extension have been made final
  * `ListeningReporter` is now `final`
  * Concrete Matchers (e.g. `UnorderedEquals` vector matcher) are now `final`
  * All Generators are now `final`
* Matcher namespacing has been redone
  * Matcher types are no longer in deeply nested namespaces
  * Matcher factory functions are no longer brought into `Catch` namespace
  * This means that all public-facing matcher-related functionality is now in `Catch::Matchers` namespace
* Defining `CATCH_CONFIG_MAIN` will no longer create main in that TU.
  * Link with `libCatch2Main.a`, or the proper CMake/pkg-config target
  * If you want to write custom main, include `catch2/catch_session.hpp`
* `CATCH_CONFIG_EXTERNAL_INTERFACES` has been removed.
  * You should instead include the appropriate headers as needed.
* `CATCH_CONFIG_IMPL` has been removed.
  * The implementation is now compiled into a static library.

--- 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
  * Catch2 currently provides _some_ generic matchers, but there should be more before final release of v3
    * So far, `IsEmpty`, `SizeIs`, and `Contains` are provided.
    * At least `ElementsAre` and `UnorderedElementsAre` are planned.

--- Fixes
* The `INFO` macro no longer contains superfluous semicolon (-1456)
* The `--list*` family of command line flags now return 0 on success (-1410, -1146)

--- Other changes
* `CATCH_CONFIG_DISABLE_MATCHERS` no longer exists.
  * If you do not want to use Matchers in a TU, do not include their header.
* `CATCH_CONFIG_ENABLE_CHRONO_STRINGMAKER` no longer exists.
  * `StringMaker` specializations for <chrono> are always provided
* Catch2's CMake now provides 2 targets, `Catch2` and `Catch2WithMain`.
  * `Catch2` is the statically compiled implementation by itself
  * `Catch2WithMain` also links in the default main
* Catch2's pkg-config integration also provides 2 packages
  * `catch2` is the statically compiled implementation by itself
  * `catch2-with-main` also links in the default main

v2.12.1

--- Fixes ---
* Vector matchers now support initializer list literals better

--- Improvements ---
* Added support for `^` (bitwise xor) to `CHECK` and `REQUIRE`

v2.12.0

--- Improvements ---
* Running tests in random order (`--order rand`) has been reworked significantly (#1908)
  * Given same seed, all platforms now produce the same order
  * Given same seed, the relative order of tests does not change if you select only a subset of them
* Vector matchers support custom allocators (#1909)
* `|` and `&` (bitwise or and bitwise and) are now supported in `CHECK` and `REQUIRE`
  * The resulting type must be convertible to `bool`

--- Fixes ---
* Fixed computation of benchmarking column widths in ConsoleReporter (#1885, #1886)
* Suppressed clang-tidy's `cppcoreguidelines-pro-type-vararg` in assertions (#1901)
  * It was a false positive trigered by the new warning support workaround
* Fixed bug in test specification parser handling of OR'd patterns using escaping (#1905)

--- Miscellaneous ---
* Worked around IBM XL's codegen bug (#1907)
  * It would emit code for _destructors_ of temporaries in an unevaluated context
* Improved detection of stdlib's support for `std::uncaught_exceptions` (#1911)

v2.11.3

--- Fixes ---
* Fixed compilation error caused by lambdas in assertions under MSVC

v2.11.2

--- Improvements ---
* GCC and Clang now issue warnings for suspicious code in assertions (#1880)
  * E.g. `REQUIRE( int != unsigned int )` will now issue mixed signedness comparison warning
  * This has always worked on MSVC, but it now also works for GCC and current Clang versions
* Colorization of "Test filters" output should be more robust now
* `--wait-for-keypress` now also accepts `never` as an option (#1866)
* Reporters no longer round-off nanoseconds when reporting benchmarking results (#1876)
* Catch2's debug break now supports iOS while using Thumb instruction set (#1862)
* It is now possible to customize benchmark's warm-up time when running the test binary (#1844)
  * `--benchmark-warmup-time {ms}`
* User can now specify how Catch2 should break into debugger (#1846)

--- Fixes ---
* Fixes missing `<random>` include in benchmarking (#1831)
* Fixed missing `<iterator>` include in benchmarking (#1874)
* Hidden test cases are now also tagged with `[!hide]` as per documentation (#1847)
* Detection of whether libc provides `std::nextafter` has been improved (#1854)
* Detection of `wmain` no longer incorrectly looks for `WIN32` macro (#1849)
  * Now it just detects Windows platform
* Composing already-composed matchers no longer modifies the partially-composed matcher expression
  * This bug has been present for the last ~2 years and nobody reported it

v2.11.1

=== Improvements ===
* Breaking into debugger is supported on iOS (#1817)
* `google-build-using-namespace` clang-tidy warning is suppressed (#1799)

=== Fixes ===
* Clang on Windows is no longer assumed to implement MSVC's traditional preprocessor (#1806)
* `ObjectStorage` now behaves properly in `const` contexts (#1820)
* `GENERATE_COPY(a, b)` now compiles properly (#1809, #1815)
* Some more cleanups in the benchmarking support

2.11.0

=== Improvements ===
* JUnit reporter output now contains more details in case of failure (#1347, #1719)
* Added SonarQube Test Data reporter (#1738)
  * It is in a separate header, just like the TAP, Automake, and TeamCity reporters
* `range` generator now allows floating point numbers (#1776)
* Reworked part of internals to increase throughput

=== Fixes ===
* The single header version should contain full benchmarking support (#1800)
* `[.foo]` is now properly parsed as `[.][foo]` when used on the command line (#1798)
* Fixed compilation of benchmarking on platforms where `steady_clock::period` is not `std::nano` (#1794)