Merge branch 'python' into breaking_bonds

.clang-tidy

@@ -7,6 +7,8 @@ Checks: |
    -clang-analyzer-optin.mpi.MPI-Checker,
    -clang-analyzer-security.FloatLoopCounter,
    bugprone-*,
+   -bugprone-easily-swappable-parameters,
+   -bugprone-implicit-widening-of-multiplication-result,
    clang-analyzer-alpha.*,
    modernize-deprecated-headers,
    modernize-make-shared,
@@ -71,15 +73,15 @@ CheckOptions:
   - key:             modernize-make-shared.IgnoreMacros
     value:           '1'
   - key:             modernize-make-shared.IncludeStyle
-    value:           '0'
+    value:           'llvm'
   - key:             modernize-make-shared.MakeSmartPtrFunction
     value:           'std::make_shared'
   - key:             modernize-make-shared.MakeSmartPtrFunctionHeader
     value:           memory
   - key:             modernize-make-unique.IgnoreMacros
     value:           '1'
   - key:             modernize-make-unique.IncludeStyle
-    value:           '0'
+    value:           'llvm'
   - key:             modernize-make-unique.MakeSmartPtrFunction
     value:           'std::make_unique'
   - key:             modernize-make-unique.MakeSmartPtrFunctionHeader

.cppcheck

@@ -0,0 +1,9 @@
+// clang-format off
+constParameter
+unusedFunction
+missingIncludeSystem
+noConstructor
+noExplicitConstructor
+redundantAssignment
+uselessAssignmentPtrArg
+preprocessorErrorDirective

.github/actions/build_and_check/action.yml

@@ -22,7 +22,7 @@ runs:
        pip3 install numpy cython h5py scipy
       shell: bash
     - run: |
-        export myconfig=maxset with_cuda=false test_timeout=600 with_asan=${{ inputs.asan }} with_ubsan=${{ inputs.ubsan }} check_skip_long=${{ inputs.check_skip_long }}
+        export myconfig=maxset with_cuda=false test_timeout=800 with_asan=${{ inputs.asan }} with_ubsan=${{ inputs.ubsan }} check_skip_long=${{ inputs.check_skip_long }}
         bash maintainer/CI/build_cmake.sh
       shell: bash
       # This is a workaround for the unfortunate interaction of MacOS and OpenMPI 4

.gitlab-ci.yml

@@ -1,4 +1,4 @@
-image: docker.pkg.github.com/espressomd/docker/ubuntu-20.04:254edd4a9c6e4d7b557be73158e400f5794e4f99
+image: ghcr.io/espressomd/docker/ubuntu-20.04:254edd4a9c6e4d7b557be73158e400f5794e4f99
 
 stages:
   - prepare
@@ -127,7 +127,7 @@ no_rotation:
 ubuntu:wo-dependencies:
   <<: *global_job_definition
   stage: build
-  image: docker.pkg.github.com/espressomd/docker/ubuntu-wo-dependencies:9ef2166b82d4c0eb258d17f5ec29f7bc39991f5d
+  image: ghcr.io/espressomd/docker/ubuntu-wo-dependencies:9ef2166b82d4c0eb258d17f5ec29f7bc39991f5d
   variables:
      myconfig: 'maxset'
      with_cuda: 'false'
@@ -145,7 +145,7 @@ ubuntu:wo-dependencies:
 debian:10:
   <<: *global_job_definition
   stage: build
-  image: docker.pkg.github.com/espressomd/docker/debian:9ef2166b82d4c0eb258d17f5ec29f7bc39991f5d
+  image: ghcr.io/espressomd/docker/debian:9ef2166b82d4c0eb258d17f5ec29f7bc39991f5d
   variables:
      with_cuda: 'false'
      myconfig: 'maxset'
@@ -160,7 +160,7 @@ debian:10:
 fedora:34:
   <<: *global_job_definition
   stage: build
-  image: docker.pkg.github.com/espressomd/docker/fedora:9ef2166b82d4c0eb258d17f5ec29f7bc39991f5d
+  image: ghcr.io/espressomd/docker/fedora:9ef2166b82d4c0eb258d17f5ec29f7bc39991f5d
   variables:
      with_cuda: 'false'
      myconfig: 'maxset'
@@ -201,7 +201,7 @@ clang-sanitizer:
 fast_math:
   <<: *global_job_definition
   stage: build
-  image: docker.pkg.github.com/espressomd/docker/cuda:9ef2166b82d4c0eb258d17f5ec29f7bc39991f5d
+  image: ghcr.io/espressomd/docker/cuda:9ef2166b82d4c0eb258d17f5ec29f7bc39991f5d
   variables:
      CC: 'gcc-9'
      CXX: 'g++-9'
@@ -222,7 +222,7 @@ fast_math:
 cuda11-maxset:
   <<: *global_job_definition
   stage: build
-  image: docker.pkg.github.com/espressomd/docker/cuda:9ef2166b82d4c0eb258d17f5ec29f7bc39991f5d
+  image: ghcr.io/espressomd/docker/cuda:9ef2166b82d4c0eb258d17f5ec29f7bc39991f5d
   variables:
      CC: 'gcc-9'
      CXX: 'g++-9'
@@ -488,7 +488,7 @@ run_doxygen:
     - docker
     - linux
 
-check_cuda_maxset_no_gpu:
+maxset_no_gpu:
   <<: *global_job_definition
   stage: additional_checks
   when: on_success
@@ -502,7 +502,7 @@ check_cuda_maxset_no_gpu:
     - docker
     - linux
 
-check_with_odd_no_of_processors:
+maxset_3_cores:
   <<: *global_job_definition
   stage: additional_checks
   when: on_success

CMakeLists.txt

@@ -77,6 +77,7 @@ option(WITH_STOKESIAN_DYNAMICS "Build with Stokesian Dynamics" OFF)
 option(WITH_BENCHMARKS "Enable benchmarks" OFF)
 option(WITH_VALGRIND_INSTRUMENTATION
        "Build with valgrind instrumentation markers" OFF)
+option(WITH_CPPCHECK "Run Cppcheck during compilation" OFF)
 if(CMAKE_CXX_COMPILER_ID MATCHES "Clang")
   option(WITH_CLANG_TIDY "Run Clang-Tidy during compilation" OFF)
 endif()
@@ -97,6 +98,7 @@ endif()
 option(WARNINGS_ARE_ERRORS "Treat warnings as errors during compilation" OFF)
 option(WITH_CCACHE "Use ccache compiler invocation." OFF)
 option(WITH_PROFILER "Enable profiler annotations." OFF)
+option(INSIDE_DOCKER "Enable when running inside Docker." OFF)
 set(TEST_TIMEOUT "300" CACHE STRING
                              "Timeout in seconds for each testsuite test")
 
@@ -126,18 +128,22 @@ include(MyConfig)
 
 include(CheckCXXSourceCompiles)
 
-set(__PRETTYFUNC__ __func__)
-foreach(func_name __PRETTY_FUNCTION__ __FUNCTION__)
+# cross-platform macro to print the function name in error messages
+set(PRETTY_FUNCTION_EXTENSION __func__)
+
+# search for a supported compiler extension that prints the function name as
+# well as its list of arguments, return type and namespace
+foreach(func_name __PRETTY_FUNCTION__ __FUNCSIG__ __FUNCTION__)
   check_cxx_source_compiles(
     "
      #include <string>
      int main() { std::string(${func_name}); }
      " result${func_name})
   if(result${func_name})
-    set(__PRETTYFUNC__ ${func_name})
+    set(PRETTY_FUNCTION_EXTENSION ${func_name})
     break()
   endif(result${func_name})
-endforeach(func_name __PRETTY_FUNCTION__ __FUNCTION__)
+endforeach()
 
 #
 # Libraries
@@ -270,6 +276,32 @@ endif(WITH_VALGRIND_INSTRUMENTATION)
 #
 
 find_package(MPI 3.0 REQUIRED)
+find_package(MpiexecBackend)
+
+# OpenMPI checks the number of processes against the number of CPUs
+if("${MPIEXEC_BACKEND_NAME}" STREQUAL "OpenMPI" AND "${MPIEXEC_BACKEND_VERSION}"
+                                                    VERSION_GREATER_EQUAL 2.0.0)
+  set(MPIEXEC_OVERSUBSCRIBE "-oversubscribe")
+else()
+  set(MPIEXEC_OVERSUBSCRIBE "")
+endif()
+
+# OpenMPI cannot run two jobs in parallel in a Docker container, because the
+# same base folder is used to store the process ids of multiple jobs. Since the
+# base folder is deleted upon completion of a job, other jobs will fail when
+# attempting to create subdirectories in the base folder.
+# https://github.com/open-mpi/ompi/issues/8510
+if("${MPIEXEC_BACKEND_NAME}" STREQUAL "OpenMPI" AND INSIDE_DOCKER)
+  cmake_host_system_information(RESULT hostname QUERY HOSTNAME)
+  function(set_mpiexec_tmpdir)
+    set(MPIEXEC_TMPDIR --mca orte_tmpdir_base
+                       "/tmp/ompi.${hostname}.$ENV{USER}.${ARGV0}" PARENT_SCOPE)
+  endfunction()
+else()
+  function(set_mpiexec_tmpdir)
+    set(MPIEXEC_TMPDIR "" PARENT_SCOPE)
+  endfunction()
+endif()
 
 #
 # Boost
@@ -322,9 +354,7 @@ target_compile_options(
             # disable warnings from -Wextra
             -Wno-sign-compare
             -Wno-unused-function
-            -Wno-unused-variable
             -Wno-unused-parameter
-            -Wno-missing-braces
             $<$<CXX_COMPILER_ID:GNU>:-Wno-clobbered>
             $<$<CXX_COMPILER_ID:Intel>:-wd592>)
 
@@ -421,6 +451,19 @@ if(WITH_CLANG_TIDY)
   set(CMAKE_CXX_CLANG_TIDY "${CLANG_TIDY_EXE};--extra-arg=--cuda-host-only")
 endif()
 
+if(WITH_CPPCHECK)
+  find_program(CMAKE_CXX_CPPCHECK NAMES cppcheck)
+  if(NOT CMAKE_CXX_CPPCHECK)
+    message(FATAL_ERROR "Could not find the program cppcheck.")
+  endif()
+  list(APPEND CMAKE_CXX_CPPCHECK "--enable=all"
+       "--std=c++${CMAKE_CXX_STANDARD}" "--quiet" "--inline-suppr"
+       "--suppressions-list=${CMAKE_CURRENT_SOURCE_DIR}/.cppcheck")
+  if(WARNINGS_ARE_ERRORS)
+    list(APPEND CMAKE_CXX_CPPCHECK "--error-exitcode=2")
+  endif()
+endif()
+
 #
 # Testing
 #

cmake/FindMpiexecBackend.cmake

@@ -0,0 +1,48 @@
+#
+# Copyright (C) 2022 The ESPResSo project
+#
+# This file is part of ESPResSo.
+#
+# ESPResSo is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# ESPResSo is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+#
+
+# Find the MPI backend.
+#
+# This code sets the following variables:
+#
+#  MPIEXEC_BACKEND_NAME MPIEXEC_BACKEND_VERSION
+
+set(MPIEXEC_BACKEND_NAME "unknown")
+set(MPIEXEC_BACKEND_VERSION 0.0.0)
+
+execute_process(
+  COMMAND ${MPIEXEC} --version RESULT_VARIABLE mpi_version_result
+  OUTPUT_VARIABLE mpi_version_output ERROR_VARIABLE mpi_version_output)
+if(mpi_version_result EQUAL 0)
+  if(mpi_version_output MATCHES "Intel\\(R\\) MPI Library")
+    set(MPIEXEC_BACKEND_NAME "Intel")
+    string(REGEX REPLACE ".*Build ([0-9]+).*" "\\1" MPIEXEC_BACKEND_VERSION ${mpi_version_output})
+  endif()
+  if(mpi_version_output MATCHES "HYDRA")
+    set(MPIEXEC_BACKEND_NAME "MPICH")
+    string(REGEX REPLACE ".*Version: +([0-9\\.]+).*" "\\1" MPIEXEC_BACKEND_VERSION ${mpi_version_output})
+  endif()
+  if(mpi_version_output MATCHES "\\(Open(RTE| MPI)\\)")
+    set(MPIEXEC_BACKEND_NAME "OpenMPI")
+    string(REGEX REPLACE ".*\\(Open(RTE| MPI)\\) ([0-9\\.]+).*" "\\2" MPIEXEC_BACKEND_VERSION ${mpi_version_output})
+  endif()
+endif()
+
+include( FindPackageHandleStandardArgs )
+FIND_PACKAGE_HANDLE_STANDARD_ARGS(MpiexecBackend REQUIRED_VARS MPIEXEC)

cmake/FindPythonHeaders.cmake

@@ -19,14 +19,14 @@
 # find the Python C++ headers
 execute_process(
   COMMAND ${PYTHON_EXECUTABLE} -c
-          "import distutils.sysconfig as cg; print(cg.get_python_inc())"
+          "import sysconfig; print(sysconfig.get_path('include'))"
   OUTPUT_VARIABLE PYTHON_INCLUDE_DIRS OUTPUT_STRIP_TRAILING_WHITESPACE)
 # find Python installation directory
 if(NOT PYTHON_INSTDIR)
   execute_process(
     COMMAND
       ${PYTHON_EXECUTABLE} -c
-      "import distutils.sysconfig as cg; print(cg.get_python_lib(prefix='${CMAKE_INSTALL_PREFIX}', plat_specific=True, standard_lib=False).replace('${CMAKE_INSTALL_PREFIX}/', '', 1))"
+      "import sysconfig; print(sysconfig.get_path('purelib', vars={'base': ''}).lstrip('/'))"
     OUTPUT_VARIABLE PYTHON_INSTDIR OUTPUT_STRIP_TRAILING_WHITESPACE)
 endif(NOT PYTHON_INSTDIR)
 

cmake/cmake_config.cmakein

@@ -23,7 +23,9 @@
 
 #define PACKAGE_NAME "${PROJECT_NAME}"
 
-#define __PRETTYFUNC__ @__PRETTYFUNC__@
-
-
-
+/**
+ * @brief Compiler-specific macro containing the demangled name
+ * of the function in the current scope. When the current compiler
+ * doesn't provide such an extension, defaults to @c __func__.
+ */
+#define PRETTY_FUNCTION_EXTENSION @PRETTY_FUNCTION_EXTENSION@

cmake/unit_test.cmake

@@ -1,17 +1,3 @@
-if(EXISTS ${MPIEXEC})
-  # OpenMPI 2.0 and higher checks the number of processes against the number of
-  # CPUs
-  execute_process(
-    COMMAND ${MPIEXEC} --version RESULT_VARIABLE mpi_version_result
-    OUTPUT_VARIABLE mpi_version_output ERROR_VARIABLE mpi_version_output)
-  if(mpi_version_result EQUAL 0 AND mpi_version_output MATCHES
-                                    "\\(Open(RTE| MPI)\\) ([2-9]\\.|1[0-9])")
-    set(MPIEXEC_OVERSUBSCRIBE "-oversubscribe")
-  else()
-    set(MPIEXEC_OVERSUBSCRIBE "")
-  endif()
-endif()
-
 # unit_test function
 function(UNIT_TEST)
   cmake_parse_arguments(TEST "" "NAME;NUM_PROC" "SRC;DEPENDS" ${ARGN})
@@ -30,10 +16,11 @@ function(UNIT_TEST)
     if(${TEST_NUM_PROC} GREATER ${TEST_NP})
       set(TEST_NUM_PROC ${TEST_NP})
     endif()
-
+    set_mpiexec_tmpdir("${TEST_NAME}")
     add_test(${TEST_NAME} ${MPIEXEC} ${MPIEXEC_OVERSUBSCRIBE}
              ${MPIEXEC_NUMPROC_FLAG} ${TEST_NUM_PROC} ${MPIEXEC_PREFLAGS}
-             ${CMAKE_CURRENT_BINARY_DIR}/${TEST_NAME} ${MPIEXEC_POSTFLAGS})
+             ${MPIEXEC_TMPDIR} ${CMAKE_CURRENT_BINARY_DIR}/${TEST_NAME}
+             ${MPIEXEC_POSTFLAGS})
   else()
     add_test(${TEST_NAME} ${TEST_NAME})
   endif()

doc/sphinx/advanced_methods.rst

@@ -1707,7 +1707,7 @@ In the *forward* reaction, the appropriate number of reactants (given by
 :math:`\nu_i`) is removed from the system, and the concomitant number of
 products is inserted into the system. In the *backward* reaction,
 reactants and products exchange their roles. The acceptance probability
-:math:`P^{\xi}` for move from state :math:`o` to :math:`n` reaction
+:math:`P^{\xi}` for a move from state :math:`o` to :math:`n` in the reaction
 ensemble is given by the criterion :cite:`smith94c`
 
 .. math::
@@ -1718,10 +1718,11 @@ ensemble is given by the criterion :cite:`smith94c`
 
 where :math:`\Delta E=E_\mathrm{new}-E_\mathrm{old}` is the change in potential energy,
 :math:`V` is the simulation box volume,
-and :math:`\beta=1/k_\mathrm{B}T`.
-The extent of reaction, :math:`\xi=1` for the forward, and
+:math:`\beta=1/k_\mathrm{B}T` is the Boltzmann factor, and
+:math:`\xi` is the extent of reaction, with :math:`\xi=1` for the forward and
 :math:`\xi=-1` for the backward direction.
-The parameter :math:`\Gamma` proportional to the reaction constant. It is defined as
+
+:math:`\Gamma` is proportional to the reaction constant. It is defined as
 
 .. math::
 

doc/sphinx/inter_bonded.rst

@@ -231,13 +231,13 @@ A virtual bond can be instantiated via
 :class:`espressomd.interactions.Virtual`::
 
     import espressomd.interactions
-    tab = espressomd.interactions.Virtual()
+    vb = espressomd.interactions.Virtual()
 
 
 This creates a virtual bond type identifier for a pair bond
 without associated potential or force. It can be used to specify topologies
 and for some analysis that rely on bonds, or for bonds that should be
-displayed in the visualization.
+displayed in the visualizer.