insert(position, value) and insert(position, n, value) shifted the
elements before reading value, so a value taken from the vector at or
after the insert position read the shifted element instead.
Fixes#1629
* Adopt the Library Fundamentals TS make_array design for the sequence factories
make_array, make_deque, make_vector, make_list, make_forward_list and
make_indirect_vector now follow the Library Fundamentals TS
std::experimental::make_array design: a defaulted leading template
parameter selects the element type explicitly, falling back to the
decayed common type of the arguments when it is not supplied. A shared
trait (etl::private_make::element_type) performs the selection lazily,
so common_type is not instantiated when the element type is explicit.
This fixes the explicit-element-type calling style for the four
common-type factories, where make_vector<char>(lvalue) never compiled:
the explicit argument pinned the first pack parameter to char, making
it a char&& that no lvalue can bind to. It also gives make_array and
make_deque the deduced form (e.g. make_array(1, 2L) is array<long, 2>)
for free. All arguments are converted with static_cast, as in the
make_array const-lvalue fix, so const-qualified lvalues and narrowing
initialisers work in both forms.
Intentional behaviour change: an explicit element type now converts
every argument instead of pinning only the first. make_vector<char>(0,
1, ...) previously produced vector<int> - the explicit char affected
only the first argument and the element type was still the common type
- and now produces vector<char>. test_make_vector documented the old
behaviour and has been updated.
The keyed set and map factories are unchanged: their key/mapped types
are separate mandatory template parameters ahead of a fully-deduced
pack, so the explicit-type style already works, and retrofitting
deduction there would interact with the defaulted comparator/hash
parameters.
Adds explicit-element-type tests for the four common-type factories
(const/non-const lvalues plus narrowing initialisers) and
deduced-element-type tests for make_array and make_deque, and updates
the four documented signatures.
* Address review comments on the make_xxx factory design
- Forward same-type arguments in the make_xxx factories instead of
casting them. static_cast produced a prvalue that, before C++17, had
to be materialised through the move constructor, rejecting
copyable-but-non-movable element types. The new
private_make::convert helper forwards arguments that already have
the element type and static_casts the rest.
- Drop ETL_NOEXCEPT from make_array: conversions can throw, and the
other factories do not declare it.
- Remove the redundant etl::decay in element_type<void, ...>;
etl::common_type already decays its arguments.
- Stop leaking the etl::private_make implementation namespace into the
documented factory signatures.
* Add a test for make_array with an explicit element type and no arguments
etl::make_array<int>() produces an empty etl::array<int, 0> and now has
a regression test. A call with neither an element type nor arguments
remains ill-formed, matching the Library Fundamentals TS: there is
nothing to deduce the element type from. The array documentation now
states this explicitly.
---------
Co-authored-by: John Wellbelove <jwellbelove@users.noreply.github.com>
Co-authored-by: Roland Reichwein <Roland.Reichwein@bmw.de>
* Fixed make_array rejecting const-qualified lvalues of the element type
make_array forwarded each argument as T (the element type) rather than
as its own deduced type. etl::forward<T> accepts only a non-const T& or
a T&&, and a T&& cannot bind to a reference-related const lvalue, so
any call passing a const-qualified lvalue of type T failed to compile:
static const char c = 42;
etl::make_array<char>(c, c); // error: would lose const qualifier
Arguments of *other* types slipped through via temporary
materialisation, which is why the existing conversion and move tests
never caught it.
Forward each argument as its own deduced type and convert explicitly
with static_cast<T>. The explicit conversion is required rather than
plain forwarding: function parameters are not constant expressions, so
narrowing initialisers such as make_array<char>(0, 1) - which the old
code accepted and the tests cover - would otherwise be rejected by the
braced initialiser list.
Adds regression tests for const/non-const lvalues of the element type
and for constexpr evaluation. Verified with clang 22 (C++14/17/20) and
GCC 14 (C++11/20); existing conversion and move-only patterns still
compile.
* Fixed make_deque rejecting const-qualified lvalues of the element type
make_deque had the same defect fixed for make_array in the previous
commit: it forwarded each argument as T (the element type) rather than
as its own deduced type. etl::forward<T> accepts only a non-const T& or
a T&&, and a T&& cannot bind to a reference-related const lvalue, so
any call passing a const-qualified lvalue of type T failed to compile:
static const char c = 42;
etl::make_deque<char>(c, c); // error: would lose const qualifier
An audit of all other make_xxx container factories found no further
instances - they forward each argument as its own deduced pack type,
which is the correct idiom.
Forward each argument as its own deduced type and convert explicitly
with static_cast<T>. The explicit conversion keeps narrowing
initialisers such as make_deque<char>(0, 1) - which the old code
accepted and test_make_deque covers - working, which plain forwarding
into the braced initialiser list would reject.
Adds a regression test for const/non-const lvalues of the element type.
No constexpr test, unlike make_array: etl::deque is not a literal type,
so make_deque cannot be evaluated at compile time. Verified with
clang 22 (C++14/17/20); the new test fails to compile before the fix
and passes after, and all 101 deque tests pass.
* Added narrowing-conversion regression tests for make_array and make_deque
Review feedback on the make_deque fix: the const-lvalue regression test
only passes arguments that already have the element type, so nothing
exercises the static_cast<T> conversion that keeps narrowing
initialisers working. If the implementation regressed to plain
forwarding into the braced initialiser list, narrowing calls such as
make_deque<char>(0, 1) would fail to compile.
make_array already covered narrowing from int literals via
test_make_array, but neither container covered conversion from
const-qualified lvalues of a different type (const int narrowed to
char), which combines both dimensions of the fix.
Adds a matching test to each container covering int literals plus
const/non-const int lvalues, all narrowed to char. Verified with
clang 22 (C++17): 102 deque and 53 array tests pass.
* Added const-lvalue regression tests for the remaining make_xxx factories
Review feedback on the make_array/make_deque fixes asked whether the
same defect - rejecting const-qualified lvalue arguments by forwarding
each argument as the element type rather than as its own deduced type -
exists in the other make_xxx container factories. The audit found that
it does not: the other factories forward each argument as its own
deduced pack type. These tests pin that down explicitly for every
remaining factory, mirroring the make_array and make_deque regression
tests:
- make_vector, make_forward_list, make_indirect_vector (element type
deduced via common_type): const/non-const char lvalues, element type
checked.
- make_set, make_multiset, make_flat_set, make_flat_multiset,
make_unordered_set, make_unordered_multiset: const/non-const lvalues
of the key type.
- make_map, make_multimap, make_flat_map, make_flat_multimap,
make_unordered_map, make_unordered_multimap: const/non-const lvalues
of the pair type.
make_reference_flat_* are excluded: they cannot currently be
instantiated at all (no initializer_list constructor exists in the
reference containers), which is reported separately.
Verified with clang 22 (C++17): all 15 test suites pass (996 tests).
* Added make_forward_list parity tests matching the make_list fixes
Checked whether the make_list defect - taking its argument pack by
value, costing an extra move per lvalue argument - also applies to
make_forward_list. It does not: make_forward_list already takes its
pack by forwarding reference, as make_vector does. No header change is
needed.
Adds the two tests that accompanied the make_list fix so that
forward_list has the same coverage and the same regression detector:
- test_make_forward_list_common_type_from_mixed_types, confirming the
element type deduction through common_type.
- test_make_forward_list_argument_copy_and_move_counts, an
instrumented type counting copy/move constructions. This is the test
that fails if the pack ever regresses to by-value (one extra move
per lvalue argument). The exact move counts are C++17-only, as
before C++17 the non-guaranteed return elision may add a move per
element.
Verified with clang 22 (C++17): all 67 forward_list tests pass.
---------
Co-authored-by: Roland Reichwein <Roland.Reichwein@bmw.de>
* Fix histogram test
Use different Start index so skipped start_index copy would not get
unnoticed in the test.
* Fix several test_vector_* for valid data
Instead of comparing the moved-from data which is invalid, compare with
the valid initial_data.
* Fix test_vector to check for the correct iterator position after erase
This was done before, but a recent change dropped it.
* Fix bit_stream: Guard nbits > 0 to prevent division by zero
* Cleanup
---------
Co-authored-by: John Wellbelove <jwellbelove@users.noreply.github.com>
commit c5dc331004b4d9611ab996935bedada203676c71
Author: John Wellbelove <john.wellbelove@etlcpp.com>
Date: Sat Jul 11 12:32:20 2026 +0100
Updated release notes
commit 4bf6ed6ce8d79857658d763e368ee56617542b2c
Author: vcoselev <145198867+vcoselev@users.noreply.github.com>
Date: Sat Jul 11 12:05:39 2026 +0100
Fix for 1405: etl::exchange doesn't work with unique_ptr (#1477)
* Separate bit order and endianness in bit_stream.h (#1495)
* Factor out ranges tests from test_algorithm.cpp to test_algorithm_ranges.cpp (#1497)
* Fix exchange to use forwarding reference (U&&) and move
- etl::exchange was using the copy constructor. Changed for etl::move.
- The assignment to the new object is now done with forward.
- Added a test with unique_ptr.
---------
Co-authored-by: Roland Reichwein <Roland.Reichwein@bmw.de>
Co-authored-by: John Wellbelove <jwellbelove@users.noreply.github.com>
commit 2ed9019df04d470530c41c1f27d35adc30b9a32d
Author: Moritz Pflanzer <moritz@pflanzer.eu>
Date: Fri Jul 10 11:42:32 2026 +0200
Update operator[] docs for map class (#1499)
* Separate bit order and endianness in bit_stream.h (#1495)
* Update operator[] docs for map class
No const variant is implemented for the operator[].
---------
Co-authored-by: Roland Reichwein <Roland.Reichwein@bmw.de>
commit 423e18d96dc645f1c733cb0d8ae6b675b9eb7e83
Author: Roland Reichwein <Roland.Reichwein@bmw.de>
Date: Fri Jul 10 11:14:08 2026 +0200
Factor out ranges tests from test_algorithm.cpp to test_algorithm_ranges.cpp (#1497)
commit 338bfdbb1f4df3c5c08a68a9038f13b8dae1c28b
Author: John Wellbelove <jwellbelove@users.noreply.github.com>
Date: Fri Jul 10 10:12:04 2026 +0100
Add coverage report to docs (#1498)
* Updated release notes
* Added link to coverage report
---------
Co-authored-by: John Wellbelove <john.wellbelove@etlcpp.com>
commit 269025424f4c6f61cebac090f4df82985fb3b239
Merge: c004cc9d faa4c15b
Author: John Wellbelove <john.wellbelove@etlcpp.com>
Date: Thu Jul 9 14:18:32 2026 +0100
Merge branch 'development' of https://github.com/ETLCPP/etl into development
commit c004cc9d508efcf398630758a47aae95b37a7eb8
Author: John Wellbelove <john.wellbelove@etlcpp.com>
Date: Thu Jul 9 10:06:17 2026 +0100
Updated release notes
commit faa4c15b49c8ad15475c83ae23e00f5a46f803ca
Author: John Wellbelove <john.wellbelove@etlcpp.com>
Date: Thu Jul 9 10:06:17 2026 +0100
Updated release notes
commit 1523e63e35a15cfd1fa72d230a3c79efa7b9c3e0
Author: Roland Reichwein <Roland.Reichwein@bmw.de>
Date: Wed Jul 8 20:48:33 2026 +0200
Support non-assignable alternatives in variant (#1496)
commit 878a0ba5a28f6b0dcae9f472a6a6a1f107de0a74
Author: John Wellbelove <john.wellbelove@etlcpp.com>
Date: Wed Jul 8 11:07:01 2026 +0100
Updated version and release notes
commit 5287fe4b3b9f58330881d2a62abf6ce8e1159b8d
Author: John Wellbelove <jwellbelove@users.noreply.github.com>
Date: Wed Jul 8 09:06:50 2026 +0100
Allow etl::vector to store pointers to functions and member functions (#1492)
* Refactored vector.h
Deleted ivectorpointer.h
* Minor changes
* Fixed code incompatible with C++03
* Clang-format changes
* Removed redundant double-qualified name
* Fixed is_is_object_pointer_v to is_object_pointer_v
* Added new test files to meson.build
* Added overflow checks to vector_ext copy and move constructors and assignments
Added assert throw tests.
* Added diagnostic pushes to eliminate GCC warning false positives
* Changed the GCC C++23 Github action to use GCC14
* Applied clang-format
* Added guard to pop_heap
* Added guard to adjust_heap
* Attempt to fix GCC diagnostic
* Attempt to fix GCC diagnostic
Re-enable ranges algorithm tests
* Attempt to fix GCC diagnostic
Re-enable ranges algorithm tests
* clang-format
* Attempt to fix GCC diagnostic
Re-enable ranges algorithm tests
* Attempt to fix GCC diagnostic
Re-enable ranges algorithm tests
* Attempt to fix GCC diagnostic
Re-enable ranges algorithm tests
* Missing newline at end of file
* Attempt to fix GCC diagnostic
Re-enable ranges algorithm tests
* Attempt to fix GCC diagnostic
Re-enable ranges algorithm tests
* Attempt to fix GCC diagnostic
Re-enable ranges algorithm tests
* Attempt to fix GCC diagnostic
Re-enable ranges algorithm tests
* Attempt to fix GCC diagnostic
* clang-format
* Changed gcc --version to gcc-14 --version
---------
Co-authored-by: John Wellbelove <john.wellbelove@etlcpp.com>
Co-authored-by: Roland Reichwein <Roland.Reichwein@bmw.de>
commit 40dddd7f9601985ae7350e60a404af9028bc9fe9
Author: Roland Reichwein <Roland.Reichwein@bmw.de>
Date: Tue Jul 7 23:41:17 2026 +0200
Add optional etl_profile label_flag for Bazel profile injection (#1491)
Co-authored-by: John Wellbelove <jwellbelove@users.noreply.github.com>
commit 84fca4cd62f72c382c062376ca934803116a20d5
Author: Roland Reichwein <Roland.Reichwein@bmw.de>
Date: Tue Jul 7 21:45:47 2026 +0200
Add missing memory functions, according the standard (#1490)
Adding:
- etl::align
- etl::assume_aligned
- etl::is_sufficiently_aligned
- etl::launder
- etl::pointer_traits
- etl::start_lifetime_as
Co-authored-by: John Wellbelove <jwellbelove@users.noreply.github.com>
commit dca6653b3759b2cdc79307fec7425589feea478d
Author: Roland Reichwein <Roland.Reichwein@bmw.de>
Date: Tue Jul 7 21:00:09 2026 +0200
Add missing interfaces to etl::chrono (#1489)
Co-authored-by: John Wellbelove <jwellbelove@users.noreply.github.com>
commit c52b2ac7b1f675e6b1d34052239423b7f3634033
Author: Roland Reichwein <Roland.Reichwein@bmw.de>
Date: Tue Jul 7 18:06:05 2026 +0200
Fix etl::mem_cast cross-size copy constructor and assignment (#1488)
The templated cross-size copy constructor and assignment operator were
broken in two ways:
- They accessed other.buffer / rhs.buffer, which is private in a
different mem_cast instantiation, so the cross-size overloads failed
to compile whenever they were actually instantiated.
- They copied Size_ (the destination size) bytes from a source buffer
that is only Other_Size bytes large, reading past the end of the
source when the destination was larger.
Use the public data() accessor and copy Other_Size bytes. The existing
static_assert(Size >= Other_Size) guarantees the destination is big
enough.
Add test_mem_cast_copy_and_assign_from_smaller regression test.
Co-authored-by: John Wellbelove <jwellbelove@users.noreply.github.com>
commit 165674cb23dda1fd4f7e5a1182f49554a796bfe8
Author: Roland Reichwein <Roland.Reichwein@bmw.de>
Date: Tue Jul 7 16:41:50 2026 +0200
Fix etl::pearson::reset() not resetting the first-byte flag (#1487)
reset() cleared the hash array but left the `first` flag false, so a
calculator reused after reset() took the XOR update path on its next
add() instead of the initialisation path. This produced an incorrect
hash for any pearson object that was reset and reused.
Reset `first` to true in reset() so a reset calculator behaves like a
freshly constructed one.
Add test_pearson_reset regression test.
Co-authored-by: John Wellbelove <jwellbelove@users.noreply.github.com>
commit 032ec4cc80f1c0f1baa0383bdbc5f9239b68f268
Author: Roland Reichwein <Roland.Reichwein@bmw.de>
Date: Tue Jul 7 13:40:04 2026 +0200
Fix etl::bitset single-element from_string shift on empty string (#1486)
The single-element from_string overloads (char, wchar_t, char16_t,
char32_t) computed element_type(1) << (string_length - 1U) before the
copy loop. For an empty string, or when active_bits is 0, string_length
is 0, so the shift count underflows to SIZE_MAX. Shifting by more than
the element width is undefined behaviour, which also breaks constexpr
evaluation.
Guard the mask with a zero-length check so the shift is only performed
when string_length > 0. The multi-element from_string overloads were
already safe.
Add test_construct_from_empty_string regression test.
Co-authored-by: John Wellbelove <jwellbelove@users.noreply.github.com>
commit 98cb365d4c28e7b72f8547b734a03a4068acd3f5
Author: Roland Reichwein <Roland.Reichwein@bmw.de>
Date: Tue Jul 7 12:32:56 2026 +0200
Fix etl::histogram copy/move not preserving start_index (#1485)
The run-time-offset specialization of etl::histogram stores a start_index
member used to map keys to bins (accumulator[key - start_index]). Its copy
constructor, move constructor, copy assignment, and move assignment copied
only the accumulator and left start_index uninitialized, so a copied or
moved histogram indexed the wrong bin in operator[]/add(), causing
out-of-bounds access (undefined behavior).
All four special member functions now also copy start_index. The
compile-time-offset specialization is unaffected, as its start index is a
template constant rather than a data member.
Add regression tests covering copy/move construction and assignment.
Co-authored-by: John Wellbelove <jwellbelove@users.noreply.github.com>
commit 5cd2f28cd54e4019f7be126bcf7394a6a8318eb9
Author: Drew Rife <darife@jlg.com>
Date: Tue Jul 7 03:14:48 2026 -0400
Refactor variant::emplace to perfect forward args instead of copy/move construct (#1494)
* refactor: emplace logic to use do_emplace for in-place construction with forwarded arguments #1493
* Add test for variant_variadic
---------
Co-authored-by: Roland Reichwein <Roland.Reichwein@bmw.de>
commit a47c0aae81b886259918961d5fcee20faa06fdd0
Author: Roland Reichwein <Roland.Reichwein@bmw.de>
Date: Mon Jul 6 17:32:26 2026 +0200
Fix etl::message_timer firing following timers early on unregister (#1484)
active_list.remove(id, has_expired) adjusts the next timer's delta only
when has_expired is false. When unregistering an active, non-expired timer,
unregister_timer() incorrectly passed true, skipping that adjustment, so
every timer after the removed one in the active list fired early by the
removed timer's delta.
Pass false instead, matching the etl::callback_timer family. tick() still
passes true, since the timer has genuinely expired there. The bug affected
all four variants: message_timer, message_timer_interrupt,
message_timer_atomic and message_timer_locked. Existing tests missed it
because they only ever unregistered the tail timer, where the delta
adjustment is a no-op.
Add a regression test that unregisters a non-tail active timer and checks
the following timer still fires at its original absolute time.
* Only enable the assign and insert member functions for class ivector<T*> & class ivector<const T*> when the value_type the iterator references is a pointer to T.
* Imporoved enable_if for assign and insert.
* clang-format
* std::void_t -> etl::void_t
* Made enable_if condition into a common class struct
---------
Co-authored-by: John Wellbelove <john.wellbelove@etlcpp.com>
* Fix compilation with -Wsign-conversion
For tests with GCC and Clang
Fixes https://github.com/ETLCPP/etl/issues/632
* Fix support of negative Id in type_lookup.h: type_from_id
* Fix element access arithmetic in atomic_gcc_sync.h fetch_add/fetch_sub
* Fix rounded_integral_division.h: divide_round_half_odd(): direction is always 1
* Print test names at test time (#1343)
* Mark uninitialized use from std library
Similar to other cases, adds compiler pragmas against warnings
caused by std library in optimized builds of tests
---------
Co-authored-by: John Wellbelove <jwellbelove@users.noreply.github.com>
* Add checks to vector and fix test failures
* Add tests to vector
* Add tests to external buffer and non trivial
* All tests should be implemented
* Add checks to vector_pointer and vector_pointer_ext
* Get bad iterators from a second vector
C++14 compiler compatibility
Updated test run scripts
Changed some ETL_ASSERT macros to ETL_ASSERT_OR_RETURN
Changed unit test macros for C++20 compaibility
Updated test run scripts
Updated CMake files to allow C++ standard selection
Replaced ETL_ASSERT_AND_RETURN with ETL_ASSERT_OR_RETURN
Updated C++14 & C++20 unit test compatibility
Changed native char8_t check
Added optional optimisation argument to bash script
* Add back_inserter implementation
- Mainly a copy of the STL implementation found in LLVM.
- Add test_back_insert_iterator unit test.
* Add documentation and use ETL_OR_STD macro
* Add MIT License in header
* Move back_insert_iterator into iterator.h
* Remove unused code
* Strictly follow the C++ STL naming
https://en.cppreference.com/w/cpp/iterator/back_insert_iterator
Strictly following the C++ STL would have container_ be container.
* Make the check for C++11 clearer
* Run the unit test only for C++11
* Add front_insert_iterator
- Make back_insert_iterator available to C++03
- Add ETL_CONSTEXPR17,ETL_NODISCARD and ETL_USING_CPP11
- Replace std:move usage by etl::move
- Update doc
* Use explicit namespace for adressof() and move().
