Thermadiag · Thermadiag · Jun 18, 2026 · May 22, 2026 · Jun 11, 2026 · Jun 11, 2026
diff --git a/.github/workflows/cmake.yml b/.github/workflows/cmake.yml
@@ -2,9 +2,9 @@ name: build
 on: 
 
   push:
-    branches: [ "main", "v2dev" ]
+    branches: [ "main", "radix" ]
   pull_request:
-    branches: [ "main", "v2dev", "releases/**" ]
+    branches: [ "main",  "releases/**" ]
   workflow_dispatch:
   workflow_call:
 

diff --git a/CMakeLists.txt b/CMakeLists.txt
@@ -24,6 +24,7 @@ set(SEQ_CMAKEDIR "${CMAKE_INSTALL_LIBDIR}/cmake/seq")
 # create source file list
 file(GLOB SeqSources
      "seq/*.hpp"
+	"seq/internal/*.hpp"
 )
 
 add_library(seq INTERFACE ${SeqSources})
@@ -56,15 +57,16 @@ install(DIRECTORY "${CMAKE_CURRENT_SOURCE_DIR}/seq"
 
 
 # Configure and install seq.pc
-configure_file(seq.pc.in seq.pc @ONLY)
-install(FILES ${CMAKE_CURRENT_BINARY_DIR}/seq.pc
+configure_file(seq.pc.in ${CMAKE_BINARY_DIR}/seq.pc @ONLY)
+install(FILES ${CMAKE_BINARY_DIR}/seq.pc
 			DESTINATION ${SEQ_INSTALL_DATAROOTDIR}/pkgconfig)
 
 
 # Configure and install seq_config.hpp
-configure_file(seq_config.hpp.in seq_config.hpp @ONLY)
-install(FILES ${CMAKE_CURRENT_BINARY_DIR}/seq_config.hpp
-			DESTINATION "${CMAKE_INSTALL_PREFIX}/include/seq")
+configure_file(seq_config.hpp.in ${CMAKE_BINARY_DIR}/seq_config.hpp @ONLY)
+install(FILES ${CMAKE_BINARY_DIR}/seq_config.hpp
+			DESTINATION "include/seq")
+
 
 # Configure and install seqConfig.cmake and seqConfigVersion.cmake
 include(CMakePackageConfigHelpers)

diff --git a/LICENSES/MPMCQueue.txt b/LICENSES/MPMCQueue.txt
@@ -0,0 +1,22 @@
+The MIT License (MIT)
+
+Copyright (c) 2018 Erik Rigtorp <erik@rigtorp.se>
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+
diff --git a/LICENSES/modycamel.txt b/LICENSES/modycamel.txt
@@ -0,0 +1,62 @@
+This license file applies to everything in this repository except that which
+is explicitly annotated as being written by other authors, i.e. the Boost
+queue (included in the benchmarks for comparison), Intel's TBB library (ditto),
+dlib::pipe (ditto),
+the CDSChecker tool (used for verification), the Relacy model checker (ditto),
+and Jeff Preshing's semaphore implementation (used in the blocking queue) which
+has a zlib license (embedded in lightweightsempahore.h).
+
+---
+
+Simplified BSD License:
+
+Copyright (c) 2013-2016, Cameron Desrochers.
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without modification,
+are permitted provided that the following conditions are met:
+
+- Redistributions of source code must retain the above copyright notice, this list of
+conditions and the following disclaimer.
+- Redistributions in binary form must reproduce the above copyright notice, this list of
+conditions and the following disclaimer in the documentation and/or other materials
+provided with the distribution.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY
+EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL
+THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT
+OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR
+TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE,
+EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+---
+
+I have also chosen to dual-license under the Boost Software License as an alternative to
+the Simplified BSD license above:
+
+Boost Software License - Version 1.0 - August 17th, 2003
+
+Permission is hereby granted, free of charge, to any person or organization
+obtaining a copy of the software and accompanying documentation covered by
+this license (the "Software") to use, reproduce, display, distribute,
+execute, and transmit the Software, and to prepare derivative works of the
+Software, and to permit third-parties to whom the Software is furnished to
+do so, all subject to the following:
+
+The copyright notices in the Software and this entire statement, including
+the above license grant, this restriction and the following disclaimer,
+must be included in all copies of the Software, in whole or in part, and
+all derivative works of the Software, unless such copies or derivative
+works are solely in the form of machine-executable object code generated by
+a source language processor.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT
+SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE
+FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE,
+ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+DEALINGS IN THE SOFTWARE.
diff --git a/README.md b/README.md
@@ -4,17 +4,20 @@
 [![CTest](https://github.com/Thermadiag/seq/actions/workflows/build-macos.yml/badge.svg?branch=main)](https://github.com/Thermadiag/seq/actions/workflows/build-macos.yml)
 [![CTest](https://github.com/Thermadiag/seq/actions/workflows/build-windows.yml/badge.svg?branch=main)](https://github.com/Thermadiag/seq/actions/workflows/build-windows.yml)
 
-Transitioning to v2.0
+For what it's worth, this library and its documentation were created without the help of AI.
+
+Transitioning to v2.1
 ---------------------
 
-*seq* v2.0 introduced a lot of changes. See this [note](docs/v2.md) for more details and explanations.
+*seq* v2.1 introduced a lot of changes. See [changelog](docs/changelog.md) for more details and explanations.
 
 Purpose
 -------
 
 The *seq* library is a header-only collection of original C++17 STL-like containers and related tools.
 
-*seq* library does not try to reimplement already existing container classes present in other libraries like <a href="https://github.com/facebook/folly">folly</a>, <a href="https://abseil.io/">abseil</a>, <a href="https://www.boost.org/">boost</a> and (of course) std. Instead, it provides new features (or a combination of features) that are usually not present in other libraries. Some low level API like bits manipulation or hashing functions are not new, but must be defined to keep the seq library self dependent.
+*seq* library does not try to reimplement already existing container classes present in other libraries like <a href="https://github.com/facebook/folly">folly</a>, <a href="https://abseil.io/">abseil</a>, <a href="https://www.boost.org/">boost</a> and (of course) std.
+Instead, it provides new features (or a combination of features) that are usually not present in other libraries. Some low level API like bits manipulation or hashing functions are not new, but must be defined to keep the seq library self dependent.
 
 Among other things (see modules below), the *seq* library defines several container classes as alternatives to STL containers or providing features not present in the STL.
 These containers generally adhere to the properties of STL containers (in C++17 version), though there are often some associated API differences and/or implementation details which differ from the standard library.
@@ -23,7 +26,7 @@ The *seq* containers are not necessarly drop-in replacement for their STL counte
 
 Currently, the *containers* module provide 5 types of containers:
 -	Sequential random-access containers: 
-	-	[seq::devector](docs/devector.md): double ended vector that optimized for front and back operations. Similar interface to `std::deque`.
+	-	[seq::devector](docs/devector.md): double ended vector optimized for front AND back operations. Similar interface to `std::deque`.
 	-	[seq::tiered_vector](docs/tiered_vector.md): tiered vector implementation optimized for fast insertion and deletion in the middle. Similar interface to `std::deque`.
 -	Sequential stable non random-access container: `seq::sequence`, fast stable list-like container.
 -	Sorted containers: 
@@ -36,28 +39,26 @@ Currently, the *containers* module provide 5 types of containers:
 -	Hash tables: 
 	-	[seq::ordered_set](docs/ordered_set.md): Ordered robin-hood hash table with backward shift deletion. Drop-in replacement for `std::unordered_set` (except for the bucket and node interface) with iterator/reference stability, with performances close to 'flat' hash maps. `seq::ordered_set` preserves the insertion order.
 	-	`seq::ordered_map`: associative version of `seq::ordered_set`.
-	-	[seq::radix_hash_set](docs/radix_tree.md): radix based hash table with a similar interface to `std::unordered_set`. Uses incremental rehash (no memory peak) with a very small memory footprint.
+	-	[seq::radix_hash_set](docs/radix_tree.md): radix based hash table with a similar interface to `std::unordered_set`. Uses incremental rehash (no memory peak), ideal when low memory footprint and low latency are required.
 	-	`seq::radix_hash_map`: associative version of `seq::radix_hash_set`.
 	-	[seq::concurrent_map](docs/concurrent_map.md) and `seq::concurrent_set`: higly scalable concurrent hash tables with interfaces similar to `boost::concurrent_flat_set/map`.
 -	Strings:
 	-	[seq::tiny_string](docs/tiny_string.md): relocatable string-like class with configurable Small String Optimization and tiny memory footprint. Makes most string containers faster.
-
+-	Other:
+	-	[seq::concurrent_queue](docs/concurrent_queue.md): fast thread-safe queue designed for Multi-Producer Multi-Consumer (MPMC) scenarios.
 
 Content
 -------
 
-The library is divided in 7 small modules:
+The library is divided in 5 small modules:
 -	[bits](docs/bits.md): low-level bits manipulation utilities
 -	[hash](docs/hash.md): tiny hashing framework
--	[charconv](docs/charconv.md): fast arithmetic to/from string conversion
--	[format](docs/format.md): fast and type safe formatting tools
 -	[containers](docs/containers.md): main module, collection of original containers: double ended vector, tiered-vector, ordered hash map, flat map based on tiered-vector, compressed vector...
 -	[any](docs/any.md): type-erasing polymorphic object wrapper used to build heterogeneous containers, including hash tables and sorted containers.
 -	[algorithm](docs/algorithm.md): a (small) collection of algorithm include the `net_sort` stable sorting algorithm.
 
 A cmake project is provided for installation and compilation of tests/benchmarks.
 
-*seq* library was tested with gcc 10.1.0 and 13.2.0 (Windows and Linux), msvc 19.43 (Windows), ClangCL 12.0.0 (Windows).
 
 Design
 ------
@@ -83,6 +84,18 @@ Currently, the following options are provided:
 -	SEQ_BUILD_TESTS(OFF): build all tests
 -	SEQ_BUILD_BENCHS(OFF): build all benchmarks
 
+It is also possible to directly copy/paste the *seq* folder into your project, while slightly dirty :).
+
+
+Benchmarks
+----------
+
+The library provides several (small) benchmarks and more will be added:
+-	Benchmark of [concurrent hash tables](docs/concurrent_map.md) at the end of the page. Its goal is to compare `seq::concurrent_set/map` to other implementations.
+-	Benchmark on [sorted containers](docs/sorted_benchmark.md). Its goal is to compare `seq::flat_set/map` and `seq::radix_set/map` to other implementations.
+-	Very tiny benchmark on [concurrent queues](docs/concurrent_queue.md) to compare `seq::concurrent_queue` with other implementations.
+-	[Memory and latency benchmark](docs/latency_benchmark.md) on hash tables to compare `seq::radix_hash_set/map` with other implementations.
+
 
 Acknowledgements
 ----------------
@@ -94,11 +107,14 @@ The `net_sort` stable sorting algorithm uses several ideas originaly coming (I t
 Benchmarks (in `seq/benchs`) compare the performances of the *seq* library with other great libraries that I use in personnal or professional projects:
 -	<a href="https://plflib.org/">plf</a>: used for the plf::colony container,
 -	<a href="https://github.com/greg7mdp/gtl">gtl</a>: used for its gtl::btree_set and gtl::parallel_flat_hash_map,
--	<a href="https://www.boost.org/">boost</a>: used for boost::flat_set, boost::unordered_flat_set and boost::concurrent_flat_map,
+-	<a href="https://www.boost.org/">boost</a>: used for boost::flat_set, boost::unordered_flat_set, boost::concurrent_flat_map, boost::lockfree::queue,
 -	<a href="https://github.com/martinus/unordered_dense">unordered_dense</a>: used for ankerl::unordered_dense::set,
 -	<a href="https://github.com/oneapi-src/oneTBB">TBB</a>: used for tbb::concurrent_unordered_map and tbb::concurrent_hash_map.
+-	<a href="https://github.com/Tessil">TSL</a>: numerous Tessil's hash tables.
+-	<a href="https://github.com/cameron314/concurrentqueue">Modycamel</a>: used for its MPMC concurrent queue.
+-	<a href="https://github.com/rigtorp/MPMCQueue">MPMCQueue</a>: Rigtorp's MPMC concurrent queue.
 
 Some of these libraries are included in the `seq/benchs` folder.
 
 
-seq:: library and this page Copyright (c) 2025, Victor Moncada
+seq:: library and this page Copyright (c) 2026, Victor Moncada
diff --git a/benchs/CMakeLists.txt b/benchs/CMakeLists.txt
@@ -6,16 +6,17 @@ enable_testing()
 # create the testing file and list of tests
 create_test_sourcelist (Benchs
   seq_benchs.cpp
-  bench_format.cpp
   bench_hash.cpp
   bench_concurrent_hash.cpp
   bench_map.cpp
+  bench_map_csv.cpp
   bench_sequence.cpp
-  bench_text_stream.cpp
   bench_tiered_vector.cpp
   bench_tiny_string.cpp
-  bench_boost_unordered_benchmarks.cpp
+  #bench_boost_unordered_benchmarks.cpp
   bench_sort.cpp
+  bench_concurrent_queue.cpp
+bench_hash_latency.cpp
   )
 
 # add the executable
@@ -60,6 +61,12 @@ if(TEST_BOOST_INSERT_ERASE)
 	target_compile_definitions(seq_benchs PRIVATE -DTEST_BOOST_INSERT_ERASE)
 endif()
 
+
+find_package(OpenMP )
+if(${OpenMP_FOUND})
+target_link_libraries(seq_benchs PRIVATE OpenMP::OpenMP_CXX)
+endif()
+
 # msvc warnings generated by seq_benchs.cpp
 target_compile_definitions(seq_benchs PRIVATE -D_CRT_SECURE_NO_WARNINGS)