Merge branch 'develop' into patch-2

.github/release_steps.md

@@ -28,14 +28,14 @@ Create a 'next\_release' branch off 'develop' and make the following changes:
   ..versionadded:: or ..versionchanged:: are missing and need to be
   added
 
-Submit this pull request, rebase if needed.  This is the last pull
+Submit this pull request.  This is the last pull request merged for the
-request merged for the release and should not contain any other
+release and should not contain any other changes. (Exceptions: this
-changes. (Exceptions: this document, last minute trivial(!) changes).
+document, last minute trivial(!) changes).
 
 This PR shall not be merged before **all** pending tests have completed
 and cleared.  We currently use a mix of automated tests running on
 either Temple's Jenkins cluster or GitHub workflows.  Those include time
-consuming tests not run on pull requests.  If needed, a bugfix pull
+consuming tests not run on pull requests.  If needed, a bug-fix pull
 request should be created and merged to clear all tests.
 
 ### Create release on GitHub
@@ -56,7 +56,7 @@ git pull
 git checkout release
 git pull
 git merge --ff-only develop
-git tag -s -m "LAMMPS feature release 19 November 2024" patch_19Nov2024
+git tag -s -m "LAMMPS feature release 4 February 2025" patch_4Feb2025
 git push git@github.com:lammps/lammps.git --tags develop release
 ```
 
@@ -77,7 +77,7 @@ release page with a summary of all the changes included and references
 to the pull requests they were merged from or check the existing draft
 for any necessary changes from pull requests that were merged but are
 not listed.  Then select the applied tag for the release in the "Choose
-a tag" dropdown list. Go to the bottom of the list and select the "Set
+a tag" drop-down list. Go to the bottom of the list and select the "Set
 as pre-release" checkbox.  The "Set as the latest release" button is
 reserved for stable releases and updates to them.
 
@@ -87,12 +87,18 @@ you can return to edit the release page and publish it.
 
 ### Prepare pre-compiled packages, update packages to GitHub
 
-Build a fully static LAMMPS installation using a musl-libc
-cross-compiler, install into a lammps-static folder, and create a
-tarball called lammps-linux-x86_64-19Nov2024.tar.gz (or using a
-corresponding date with a future release) from the lammps-static folder.
 A suitable build environment is provided with the
-https://download.lammps.org/static/fedora37_musl.sif container image.
+https://download.lammps.org/static/fedora41_musl_mingw.sif container
+image.  The corresponding container build definition file is maintained
+in the tools/singularity folder of the LAMMPS source distribution.
+
+#### Fully portable static Linux x86_64 non-MPI binaries
+
+The following commands use the Fedora container to build a fully static
+LAMMPS installation using a musl-libc cross-compiler, install it into a
+`lammps-static` folder, and create a tarball called
+`lammps-linux-x86_64-4Feb2025.tar.gz` (or using a corresponding date
+with a future release) from the `lammps-static` folder.
 
 ``` sh
 rm -rf release-packages
@@ -105,49 +111,173 @@ cmake -S lammps-release/cmake -B build-release -G Ninja -D CMAKE_INSTALL_PREFIX=
 cmake --build build-release --target all
 cmake --build build-release --target install
 /usr/musl/bin/x86_64-linux-musl-strip lammps-static/bin/*
-tar -czvvf lammps-linux-x86_64-19Nov2024.tar.gz lammps-static
+tar -czvvf ../lammps-linux-x86_64-4Feb2025.tar.gz lammps-static
 exit # fedora 41 container
+cd ..
 ```
 
 The resulting tar archive can be uploaded to the GitHub release page with:
 
-```
+``` sh
-gh release upload patch_19Nov2024 lammps-linux-x86_64-19Nov2024.tar.gz
+gh release upload patch_4Feb2025 lammps-linux-x86_64-4Feb2025.tar.gz
 ```
 
+#### Linux x86_64 Flatpak bundle with GUI included
+
 Make sure you have the `flatpak` and `flatpak-builder` packages
-installed locally (they cannot be used from the container) and build a
+installed locally (they require binaries that run with elevated
+privileges and thus cannot be used from the container) and build a
 LAMMPS and LAMMPS-GUI flatpak bundle in the `release-packages` folder
 with:
 
 ``` sh
+cd release-packages
 flatpak --user remote-add --if-not-exists flathub https://dl.flathub.org/repo/flathub.flatpakrepo
 flatpak-builder  --force-clean --verbose --repo=$PWD/flatpak-repo --install-deps-from=flathub --state-dir=$PWD --user --ccache --default-branch=release flatpak-build lammps-release/tools/lammps-gui/org.lammps.lammps-gui.yml
-flatpak build-bundle --runtime-repo=https://flathub.org/repo/flathub.flatpakrepo --verbose $PWD/flatpak-repo LAMMPS-Linux-x86_64-GUI-19Nov2024.flatpak org.lammps.lammps-gui release
+flatpak build-bundle --runtime-repo=https://flathub.org/repo/flathub.flatpakrepo --verbose $PWD/flatpak-repo ../LAMMPS-Linux-x86_64-GUI-4Feb2025.flatpak org.lammps.lammps-gui release
+cd ..
 ```
 
 The resulting flatpak bundle file can be uploaded to the GitHub release page with:
 
-```
+``` sh
-gh release upload patch_19Nov2024 LAMMPS-Linux-x86_64-GUI-19Nov2024.flatpak
+gh release upload patch_4Feb2025 LAMMPS-Linux-x86_64-GUI-4Feb2025.flatpak
 ```
 
-Also build serial executable packages that also include LAMMPS-GUI for
+#### LAMMPS Source tarball
-Linux, macOS, and Windows, and upload them to the GitHub release.
 
-Clean up:
+The container for the static binary can also be used to prepare the source
+tarball including the HTML and PDF manual (this is currently done automatically
+when the releases is created and the tarball uploaded to https://download.lammps.org/tars/).
+The steps are as follows:
+
+``` sh
+cd release-packages
+apptainer shell fedora41_musl_mingw.sif
+cd lammps-release
+rm -f ../release.tar*
+git archive --output=../release.tar --prefix=lammps-4Feb2025/ HEAD
+cd doc
+make clean-all
+make html pdf
+tar -rf ../../release.tar --transform 's,^,lammps-4Feb2025/doc/,' html Manual.pdf
+gzip -9v ../../release.tar
+mv ../../release.tar.gz ../../lammps-src-4Feb2025.tar.gz
+exit # fedora41 container
+cd ..
+```
+
+The resulting source tarball can be uploaded to the GitHub release page with:
+
+``` sh
+gh release upload patch_4Feb2025 lammps-src-4Feb2025.tar.gz
+```
+
+#### Build Windows Installer Packages with MinGW Linux-to-Windows Cross-compiler
+
+The various Windows installer packages can also be built with
+apptainer container image.
+
+``` sh
+cd release-packages
+apptainer shell fedora41_musl_mingw.sif
+git clone --depth 10 https://github.com/lammps/lammps-packages.git lammps-packages
+cd lammps-packages/mingw-cross
+ln -sf ../../lammps-release lammps
+./buildall.sh release >& mk.log & less +F mk.log
+```
+
+The installer with the GUI included can be uploaded to the GitHub release page with:
+
+``` sh
+ln -sf LAMMPS-64bit-GUI-4Feb2025.exe LAMMPS-Win10-64bit-GUI-4Feb2025.exe
+gh release upload patch_4Feb2025 LAMMPS-Win10-64bit-GUI-4Feb2025.exe
+```
+
+The symbolic link is used to have a consistent naming scheme for the packages
+attached to the GitHub release page.
+
+#### Clean up:
 
 ``` sh
 cd ..
 rm -r release-packages
 ```
 
-TODO:
+#### Build Multi-arch App-bundle for macOS
-- add detailed commands for building GUI packages on Ubuntu 20.04LTS (move to 22.04LTS?),
-  macOS, and Windows cross-compiler and upload to GitHub
-- build all Windows cross-compiled installer packages using lammps-packages repo
 
-### Update download website
+Building app-bundles for macOS is not as easily automated and portable
+as some of the other steps.  It requires a machine actually running
+macOS.  In that machine the Xcode compiler package needs to be
+installed. This also includes tools for building and manipulating disk
+images.  This compiler supports building executables for both, the
+x86_64 and the arm64 architectures.  This requires building with CMake
+and using the CMake settings:
+
+``` sh
+-D CMAKE_OSX_ARCHITECTURES=arm64;x86_64
+-D CMAKE_OSX_DEPLOYMENT_TARGER=11.0
+```
+
+This will add the compiler flags `-arch arm64 -arch x86_64
+-mmacosx-version-min=11.0` and thus produce object for both
+architectures and support for macOS versions back to version 11 (aka Big
+Sur).  With these settings the following libraries should be compiled
+and installed (e.g. to `$HOME/.local`) as static libraries only:
+- libomp taken from the LLVM/Clang source distribution (to support OpenMP)
+- jpeg
+- zlib
+- png
+- Qt (for LAMMPS-GUI)
+
+When configuring LAMMPS the `cmake/presets/clang.cmake` should be used
+and as many packages as possible enabled. For LAMMPS-GUI, MPI should be
+disabled with `-D BUILD_MPI=OFF` and LAMMPS-GUI enabled with 
+`-D BUILD_LAMMPS_GUI=ON`.  If the CMake configuration is successful,
+settings for building a macOS app-bundle are enabled and with `cmake
+--build build --target dmg` extra steps will be executed that will build
+a macOS application installer image under the name
+`LAMMPS_GUI-macOS-multiarch-4Feb2025.dmg`
+
+The application image can be uploaded to the GitHub release page with:
+
+``` sh
+ln -sf LAMMPS_GUI-macOS-multiarch-4Feb2025.dmg LAMMPS-macOS-multiarch-GUI-4Feb2025.dmg
+gh release upload patch_4Feb2025 LAMMPS-macOS-multiarch-GUI-4Feb2025.dmg
+```
+
+The symbolic link is used to have a consistent naming scheme for the packages
+attached to the GitHub release page.
+
+We are currently building the application images on macOS 12 (aka Monterey).
+
+#### Build Linux x86_64 binary tarball on Ubuntu 20.04LTS
+
+While the flatpak Linux version uses portable runtime libraries provided
+by the flatpak environment, we also build regular Linux executables that
+use a wrapper script and matching shared libraries in a tarball.  To be
+compatible with many Linux distributions, one has to build this on a
+very old Linux distribution, since most Linux system libraries are
+usually backward compatible but not forward compatible.  This is
+currently done on an Ubuntu 20.04LTS system. Once LAMMPS moves to
+require CMake 3.20 and C++17, we will have to move to Ubuntu 22.04LTS.
+This installation (either on a real or a virtual machine) should have
+the packages installed that are indicated in
+`tools/singularity/ubuntu20.04.def` plus Qt version 5.x with development
+headers, so that LAMMPS-GUI can be compiled.
+
+Also the building of the binary tarball and setup of the bundled
+libraries and wrapper scripts is automated and can executed with `cmake
+--build build --target tgz`.  This should produce a file
+`LAMMPS_GUI-Linux-amd64-4Feb2025.tar.gz` which can be uploaded to the
+GitHub release page with:
+
+``` sh
+ln -sf LAMMPS_GUI-Linux-amd64-4Feb2025.tar.gz LAMMPS-Linux-x86_64-GUI-4Feb2025.tar.gz
+gh release upload patch_4Feb2025 LAMMPS-Linux-x86_64-GUI-4Feb2025.tar.gz
+```
+
+### Update download page on LAMMPS website
 
 Check out the LAMMPS website repo
 https://github.com/lammps/lammps-website.git and edit the file
@@ -156,7 +286,7 @@ html` and review `html/download.html` Then add and commit to git and
 push the changes to GitHub.  The Temple Jenkis cluster will
 automatically update https://www.lammps.org/download.html accordingly.
 
-Notify Steve of the release so he can update `src/bug.txt` on the
+Also notify Steve of the release so he can update `src/bug.txt` on the
 website from the available release notes.
 
 ## LAMMPS Stable Release

.github/workflows/check-vla.yml

@@ -77,7 +77,7 @@ jobs:
               -D PKG_MDI=on \
               -D PKG_MANIFOLD=on \
               -D PKG_ML-PACE=on \
-              -D PKG_ML-RANN=off \
+              -D PKG_ML-RANN=on \
               -D PKG_MOLFILE=on \
               -D PKG_RHEO=on \
               -D PKG_PTM=on \

README

@@ -23,17 +23,20 @@ more information about the code and its uses.
 The LAMMPS distribution includes the following files and directories:
 
 README                     this file
-LICENSE                    the GNU General Public License (GPL)
+LICENSE                    the GNU General Public License (GPLv2)
-bench                      benchmark problems
+CITATION.cff               Citation information for LAMMPS in CFF format
+bench                      benchmark inputs
 cmake                      CMake build files
 doc                        documentation
-examples                   simple test problems
+examples                   example inputs for many LAMMPS commands
-fortran                    Fortran wrapper for LAMMPS
+fortran                    Fortran 2003 module for LAMMPS
 lib                        additional provided or external libraries
 potentials                 interatomic potential files
-python                     Python wrappers for LAMMPS
+python                     Python module for LAMMPS
 src                        source files
 tools                      pre- and post-processing tools
+unittest                   test programs for use with CTest
+.github                    Git and GitHub related files and tools
 
 Point your browser at any of these files to get started:
 
@@ -42,6 +45,8 @@ https://docs.lammps.org/Intro.html          hi-level introduction
 https://docs.lammps.org/Build.html          how to build LAMMPS
 https://docs.lammps.org/Run_head.html       how to run LAMMPS
 https://docs.lammps.org/Commands_all.html   Table of available commands
+https://docs.lammps.org/Howto.html          Short tutorials and HowTo discussions
+https://docs.lammps.org/Errors.html         How to interpret and debug errors
 https://docs.lammps.org/Library.html        LAMMPS library interfaces
 https://docs.lammps.org/Modify.html         how to modify and extend LAMMPS
 https://docs.lammps.org/Developer.html      LAMMPS developer info

cmake/CMakeLists.txt

@@ -209,7 +209,7 @@ endif()
 ########################################################################
 # User input options                                                   #
 ########################################################################
-# backward compatibility with CMake before 3.12 and older LAMMPS documentation
+# backward compatibility with older LAMMPS documentation
 if (PYTHON_EXECUTABLE)
   set(Python_EXECUTABLE "${PYTHON_EXECUTABLE}")
 endif()
@@ -225,6 +225,12 @@ if(DEFINED ENV{VIRTUAL_ENV} AND NOT Python_EXECUTABLE)
     "   Setting Python interpreter to: ${Python_EXECUTABLE}")
 endif()
 
+find_package(Python COMPONENTS Interpreter QUIET)
+# NOTE: RHEL 8.0 and Ubuntu 18.04LTS ship with Python 3.6, Python 3.8 was EOL in 2024
+if(Python_VERSION VERSION_LESS 3.6)
+  message(FATAL_ERROR "LAMMPS requires Python 3.6 or later")
+endif()
+
 set(LAMMPS_MACHINE "" CACHE STRING "Suffix to append to lmp binary (WON'T enable any features automatically")
 mark_as_advanced(LAMMPS_MACHINE)
 if(LAMMPS_MACHINE)
@@ -930,7 +936,7 @@ endif()
 include(Testing)
 include(CodeCoverage)
 include(CodingStandard)
-find_package(ClangFormat 11.0)
+find_package(ClangFormat 11.0 QUIET)
 
 if(ClangFormat_FOUND)
   add_custom_target(format-src

cmake/Modules/CodeCoverage.cmake

@@ -7,76 +7,76 @@
 # For Python coverage the coverage package needs to be installed
 ###############################################################################
 if(ENABLE_COVERAGE)
-    find_program(GCOVR_BINARY gcovr)
+  find_program(GCOVR_BINARY gcovr)
-    find_package_handle_standard_args(GCOVR DEFAULT_MSG GCOVR_BINARY)
+  find_package_handle_standard_args(GCOVR DEFAULT_MSG GCOVR_BINARY)
 
-    find_program(COVERAGE_BINARY coverage)
+  find_program(COVERAGE_BINARY coverage)
-    find_package_handle_standard_args(COVERAGE DEFAULT_MSG COVERAGE_BINARY)
+  find_package_handle_standard_args(COVERAGE DEFAULT_MSG COVERAGE_BINARY)
 
-    if(GCOVR_FOUND)
+  if(GCOVR_FOUND)
-        get_filename_component(ABSOLUTE_LAMMPS_SOURCE_DIR ${LAMMPS_SOURCE_DIR} ABSOLUTE)
+    get_filename_component(ABSOLUTE_LAMMPS_SOURCE_DIR ${LAMMPS_SOURCE_DIR} ABSOLUTE)
 
-        add_custom_target(
+    add_custom_target(
-            gen_coverage_xml
+        gen_coverage_xml
-            COMMAND ${GCOVR_BINARY} -s -x -r ${ABSOLUTE_LAMMPS_SOURCE_DIR} --object-directory=${CMAKE_BINARY_DIR} -o coverage.xml
+        COMMAND ${GCOVR_BINARY} -s -x -r ${ABSOLUTE_LAMMPS_SOURCE_DIR} --object-directory=${CMAKE_BINARY_DIR} -o coverage.xml
-            WORKING_DIRECTORY ${CMAKE_BINARY_DIR}
+        WORKING_DIRECTORY ${CMAKE_BINARY_DIR}
-            COMMENT "Generating XML coverage report..."
+        COMMENT "Generating XML coverage report..."
-        )
+    )
 
-        set(COVERAGE_HTML_DIR ${CMAKE_BINARY_DIR}/coverage_html)
+    set(COVERAGE_HTML_DIR ${CMAKE_BINARY_DIR}/coverage_html)
 
-        add_custom_target(coverage_html_folder
+    add_custom_target(coverage_html_folder
-            COMMAND ${CMAKE_COMMAND} -E make_directory ${COVERAGE_HTML_DIR})
+        COMMAND ${CMAKE_COMMAND} -E make_directory ${COVERAGE_HTML_DIR})
 
-        add_custom_target(
+    add_custom_target(
-            gen_coverage_html
+        gen_coverage_html
-            COMMAND ${GCOVR_BINARY} -s  --html --html-details -r ${ABSOLUTE_LAMMPS_SOURCE_DIR} --object-directory=${CMAKE_BINARY_DIR} -o ${COVERAGE_HTML_DIR}/index.html
+        COMMAND ${GCOVR_BINARY} -s  --html --html-details -r ${ABSOLUTE_LAMMPS_SOURCE_DIR} --object-directory=${CMAKE_BINARY_DIR} -o ${COVERAGE_HTML_DIR}/index.html
-            WORKING_DIRECTORY ${CMAKE_BINARY_DIR}
+        WORKING_DIRECTORY ${CMAKE_BINARY_DIR}
-            COMMENT "Generating HTML coverage report..."
+        COMMENT "Generating HTML coverage report..."
-        )
+    )
-        add_dependencies(gen_coverage_html coverage_html_folder)
+    add_dependencies(gen_coverage_html coverage_html_folder)
 
-        add_custom_target(clean_coverage_html
+    add_custom_target(clean_coverage_html
-            ${CMAKE_COMMAND} -E remove_directory ${COVERAGE_HTML_DIR}
+        ${CMAKE_COMMAND} -E remove_directory ${COVERAGE_HTML_DIR}
-            COMMENT "Deleting HTML coverage report..."
+        COMMENT "Deleting HTML coverage report..."
-        )
+    )
 
-        add_custom_target(reset_coverage
+    add_custom_target(reset_coverage
-            ${CMAKE_COMMAND} -E remove -f */*.gcda */*/*.gcda */*/*/*.gcda
+        ${CMAKE_COMMAND} -E remove -f */*.gcda */*/*.gcda */*/*/*.gcda
-                              */*/*/*/*.gcda */*/*/*/*/*.gcda */*/*/*/*/*/*.gcda
+                          */*/*/*/*.gcda */*/*/*/*/*.gcda */*/*/*/*/*/*.gcda
-                              */*/*/*/*/*/*/*.gcda */*/*/*/*/*/*/*/*.gcda
+                          */*/*/*/*/*/*/*.gcda */*/*/*/*/*/*/*/*.gcda
-                              */*/*/*/*/*/*/*/*/*.gcda */*/*/*/*/*/*/*/*/*/*.gcda
+                          */*/*/*/*/*/*/*/*/*.gcda */*/*/*/*/*/*/*/*/*/*.gcda
-            WORKIND_DIRECTORY ${CMAKE_BINARY_DIR}
+        WORKIND_DIRECTORY ${CMAKE_BINARY_DIR}
-            COMMENT "Deleting coverage data files..."
+        COMMENT "Deleting coverage data files..."
-        )
+    )
-        add_dependencies(reset_coverage clean_coverage_html)
+    add_dependencies(reset_coverage clean_coverage_html)
-    endif()
+  endif()
 
-    if(COVERAGE_FOUND)
+  if(COVERAGE_FOUND)
-        set(PYTHON_COVERAGE_HTML_DIR ${CMAKE_BINARY_DIR}/python_coverage_html)
+    set(PYTHON_COVERAGE_HTML_DIR ${CMAKE_BINARY_DIR}/python_coverage_html)
-        configure_file(.coveragerc.in ${CMAKE_BINARY_DIR}/.coveragerc @ONLY)
+    configure_file(.coveragerc.in ${CMAKE_BINARY_DIR}/.coveragerc @ONLY)
 
-        add_custom_command(
+    add_custom_command(
-            OUTPUT ${CMAKE_BINARY_DIR}/unittest/python/.coverage
+        OUTPUT ${CMAKE_BINARY_DIR}/unittest/python/.coverage
-            COMMAND ${COVERAGE_BINARY} combine
+        COMMAND ${COVERAGE_BINARY} combine
-            WORKING_DIRECTORY ${CMAKE_BINARY_DIR}/unittest/python
+        WORKING_DIRECTORY ${CMAKE_BINARY_DIR}/unittest/python
-            COMMENT "Combine Python coverage files..."
+        COMMENT "Combine Python coverage files..."
-        )
+    )
 
-        add_custom_target(
+    add_custom_target(
-            gen_python_coverage_html
+        gen_python_coverage_html
-            COMMAND ${COVERAGE_BINARY} html --rcfile=${CMAKE_BINARY_DIR}/.coveragerc -d ${PYTHON_COVERAGE_HTML_DIR}
+        COMMAND ${COVERAGE_BINARY} html --rcfile=${CMAKE_BINARY_DIR}/.coveragerc -d ${PYTHON_COVERAGE_HTML_DIR}
-            DEPENDS ${CMAKE_BINARY_DIR}/unittest/python/.coverage ${CMAKE_BINARY_DIR}/.coveragerc
+        DEPENDS ${CMAKE_BINARY_DIR}/unittest/python/.coverage ${CMAKE_BINARY_DIR}/.coveragerc
-            WORKING_DIRECTORY ${CMAKE_BINARY_DIR}/unittest/python
+        WORKING_DIRECTORY ${CMAKE_BINARY_DIR}/unittest/python
-            COMMENT "Generating HTML Python coverage report..."
+        COMMENT "Generating HTML Python coverage report..."
-        )
+    )
 
-        add_custom_target(
+    add_custom_target(
-            gen_python_coverage_xml
+        gen_python_coverage_xml
-            COMMAND ${COVERAGE_BINARY} xml --rcfile=${CMAKE_BINARY_DIR}/.coveragerc -o ${CMAKE_BINARY_DIR}/python_coverage.xml
+        COMMAND ${COVERAGE_BINARY} xml --rcfile=${CMAKE_BINARY_DIR}/.coveragerc -o ${CMAKE_BINARY_DIR}/python_coverage.xml
-            DEPENDS ${CMAKE_BINARY_DIR}/unittest/python/.coverage ${CMAKE_BINARY_DIR}/.coveragerc
+        DEPENDS ${CMAKE_BINARY_DIR}/unittest/python/.coverage ${CMAKE_BINARY_DIR}/.coveragerc
-            WORKING_DIRECTORY ${CMAKE_BINARY_DIR}/unittest/python
+        WORKING_DIRECTORY ${CMAKE_BINARY_DIR}/unittest/python
-            COMMENT "Generating XML Python coverage report..."
+        COMMENT "Generating XML Python coverage report..."
-        )
+    )
-    endif()
+  endif()
 endif()

cmake/Modules/CodingStandard.cmake

@@ -1,40 +1,39 @@
-# use default (or custom) Python executable, if version is sufficient
+# use default (or custom) Python executable.
-if(Python_VERSION VERSION_GREATER_EQUAL 3.6)
+# Python version check is in main CMakeLists.txt file
+if(Python_EXECUTABLE)
   set(Python3_EXECUTABLE ${Python_EXECUTABLE})
 endif()
 find_package(Python3 COMPONENTS Interpreter)
 
 if(Python3_EXECUTABLE)
-    if(Python3_VERSION VERSION_GREATER_EQUAL 3.6)
+  add_custom_target(
-        add_custom_target(
+    check-whitespace
-          check-whitespace
+    ${Python3_EXECUTABLE} ${LAMMPS_TOOLS_DIR}/coding_standard/whitespace.py .
-          ${Python3_EXECUTABLE} ${LAMMPS_TOOLS_DIR}/coding_standard/whitespace.py .
+    WORKING_DIRECTORY  ${LAMMPS_DIR}
-          WORKING_DIRECTORY  ${LAMMPS_DIR}
+    COMMENT "Check for whitespace errors")
-          COMMENT "Check for whitespace errors")
+  add_custom_target(
-        add_custom_target(
+    check-homepage
-          check-homepage
+    ${Python3_EXECUTABLE} ${LAMMPS_TOOLS_DIR}/coding_standard/homepage.py .
-          ${Python3_EXECUTABLE} ${LAMMPS_TOOLS_DIR}/coding_standard/homepage.py .
+    WORKING_DIRECTORY  ${LAMMPS_DIR}
-          WORKING_DIRECTORY  ${LAMMPS_DIR}
+    COMMENT "Check for homepage URL errors")
-          COMMENT "Check for homepage URL errors")
+  add_custom_target(
-        add_custom_target(
+    check-permissions
-          check-permissions
+    ${Python3_EXECUTABLE} ${LAMMPS_TOOLS_DIR}/coding_standard/permissions.py .
-          ${Python3_EXECUTABLE} ${LAMMPS_TOOLS_DIR}/coding_standard/permissions.py .
+    WORKING_DIRECTORY  ${LAMMPS_DIR}
-          WORKING_DIRECTORY  ${LAMMPS_DIR}
+    COMMENT "Check for permission errors")
-          COMMENT "Check for permission errors")
+  add_custom_target(
-        add_custom_target(
+    fix-whitespace
-          fix-whitespace
+    ${Python3_EXECUTABLE} ${LAMMPS_TOOLS_DIR}/coding_standard/whitespace.py -f .
-          ${Python3_EXECUTABLE} ${LAMMPS_TOOLS_DIR}/coding_standard/whitespace.py -f .
+    WORKING_DIRECTORY  ${LAMMPS_DIR}
-          WORKING_DIRECTORY  ${LAMMPS_DIR}
+    COMMENT "Fix whitespace errors")
-          COMMENT "Fix whitespace errors")
+  add_custom_target(
-        add_custom_target(
+    fix-homepage
-          fix-homepage
+    ${Python3_EXECUTABLE} ${LAMMPS_TOOLS_DIR}/coding_standard/homepage.py -f .
-          ${Python3_EXECUTABLE} ${LAMMPS_TOOLS_DIR}/coding_standard/homepage.py -f .
+    WORKING_DIRECTORY  ${LAMMPS_DIR}
-          WORKING_DIRECTORY  ${LAMMPS_DIR}
+    COMMENT "Fix homepage URL errors")
-          COMMENT "Fix homepage URL errors")
+  add_custom_target(
-        add_custom_target(
+    fix-permissions
-          fix-permissions
+    ${Python3_EXECUTABLE} ${LAMMPS_TOOLS_DIR}/coding_standard/permissions.py -f .
-          ${Python3_EXECUTABLE} ${LAMMPS_TOOLS_DIR}/coding_standard/permissions.py -f .
+    WORKING_DIRECTORY  ${LAMMPS_DIR}
-          WORKING_DIRECTORY  ${LAMMPS_DIR}
+    COMMENT "Fix permission errors")
-          COMMENT "Fix permission errors")
-    endif()
 endif()

cmake/Modules/Documentation.cmake

@@ -13,7 +13,7 @@ if(BUILD_DOC)
   endif()
   find_package(Python3 REQUIRED COMPONENTS Interpreter)
   if(Python3_VERSION VERSION_LESS 3.8)
-    message(FATAL_ERROR "Python 3.8 and up is required to build the HTML documentation")
+    message(FATAL_ERROR "Python 3.8 and up is required to build the LAMMPS HTML documentation")
   endif()
   set(VIRTUALENV ${Python3_EXECUTABLE} -m venv)
 
@@ -65,8 +65,8 @@ if(BUILD_DOC)
     find_package(Sphinx)
   endif()
 
-  set(MATHJAX_URL "https://github.com/mathjax/MathJax/archive/3.1.3.tar.gz" CACHE STRING "URL for MathJax tarball")
+  set(MATHJAX_URL "https://github.com/mathjax/MathJax/archive/3.2.2.tar.gz" CACHE STRING "URL for MathJax tarball")
-  set(MATHJAX_MD5 "b81661c6e6ba06278e6ae37b30b0c492" CACHE STRING "MD5 checksum of MathJax tarball")
+  set(MATHJAX_MD5 "08dd6ef33ca08870220d9aade2a62845" CACHE STRING "MD5 checksum of MathJax tarball")
   mark_as_advanced(MATHJAX_URL)
   GetFallbackURL(MATHJAX_URL MATHJAX_FALLBACK)
 

cmake/Modules/LAMMPSInterfacePlugin.cmake

@@ -34,8 +34,26 @@ if(MSVC)
   add_compile_definitions(_CRT_SECURE_NO_WARNINGS)
 endif()
 
-# C++11 is required
+if(NOT CMAKE_CXX_STANDARD)
-set(CMAKE_CXX_STANDARD 11)
+  if(cxx_std_17 IN_LIST CMAKE_CXX_COMPILE_FEATURES)
+    set(CMAKE_CXX_STANDARD 17)
+  else()
+    set(CMAKE_CXX_STANDARD 11)
+  endif()
+endif()
+if(CMAKE_CXX_STANDARD LESS 11)
+  message(FATAL_ERROR "C++ standard must be set to at least 11")
+endif()
+if(CMAKE_CXX_STANDARD LESS 17)
+  message(WARNING "Selecting C++17 standard is preferred over C++${CMAKE_CXX_STANDARD}")
+endif()
+if(PKG_KOKKOS AND (CMAKE_CXX_STANDARD LESS 17))
+  set(CMAKE_CXX_STANDARD 17)
+endif()
+# turn off C++17 check in lmptype.h
+if(LAMMPS_CXX11)
+  add_compile_definitions(LAMMPS_CXX11)
+endif()
 set(CMAKE_CXX_STANDARD_REQUIRED ON)
 
 # Need -restrict with Intel compilers

cmake/Modules/Packages/ML-IAP.cmake

@@ -24,9 +24,7 @@ if(MLIAP_ENABLE_PYTHON)
   if(NOT PKG_PYTHON)
     message(FATAL_ERROR "Must enable PYTHON package for including Python support in ML-IAP")
   endif()
-  if(Python_VERSION VERSION_LESS 3.6)
+  # Python version check is in main CMakeLists.txt file
-    message(FATAL_ERROR "Python support in ML-IAP requires Python 3.6 or later")
-  endif()
 
   set(MLIAP_BINARY_DIR ${CMAKE_BINARY_DIR}/cython)
   file(GLOB MLIAP_CYTHON_SRC CONFIGURE_DEPENDS ${LAMMPS_SOURCE_DIR}/ML-IAP/*.pyx)

cmake/Modules/Packages/ML-QUIP.cmake

@@ -37,7 +37,7 @@ if(DOWNLOAD_QUIP)
   endforeach()
   # Fix cmake crashing when MATH_LINKOPTS not set, required for e.g. recent Cray Programming Environment
   set(temp "${temp} -L/_DUMMY_PATH_\n")
-  set(temp "${temp}PYTHON=python\nPIP=pip\nEXTRA_LINKOPTS=\n")
+  set(temp "${temp}PYTHON=${Python_EXECUTABLE}\nPIP=pip\nEXTRA_LINKOPTS=\n")
   set(temp "${temp}HAVE_CP2K=0\nHAVE_VASP=0\nHAVE_TB=0\nHAVE_PRECON=1\nHAVE_LOTF=0\nHAVE_ONIOM=0\n")
   set(temp "${temp}HAVE_LOCAL_E_MIX=0\nHAVE_QC=0\nHAVE_GAP=1\nHAVE_DESCRIPTORS_NONCOMMERCIAL=1\n")
   set(temp "${temp}HAVE_TURBOGAP=0\nHAVE_QR=1\nHAVE_THIRDPARTY=0\nHAVE_FX=0\nHAVE_SCME=0\nHAVE_MTP=0\n")

cmake/Modules/Packages/PLUMED.cmake

@@ -40,6 +40,13 @@ mark_as_advanced(PLUMED_URL)
 mark_as_advanced(PLUMED_MD5)
 GetFallbackURL(PLUMED_URL PLUMED_FALLBACK)
 
+# adjust C++ standard support for self-compiled Plumed2
+if(CMAKE_CXX_STANDARD GREATER 11)
+  set(PLUMED_CXX_STANDARD 14)
+else()
+  set(PLUMED_CXX_STANDARD 11)
+endif()
+
 if((CMAKE_SYSTEM_NAME STREQUAL "Windows") AND (CMAKE_CROSSCOMPILING))
   if(CMAKE_SYSTEM_PROCESSOR STREQUAL "x86_64")
     set(CROSS_CONFIGURE mingw64-configure)
@@ -55,7 +62,7 @@ if((CMAKE_SYSTEM_NAME STREQUAL "Windows") AND (CMAKE_CROSSCOMPILING))
     URL_MD5 ${PLUMED_MD5}
     BUILD_IN_SOURCE 1
     CONFIGURE_COMMAND ${CROSS_CONFIGURE} --disable-shared --disable-bsymbolic
-                                         --disable-python --enable-cxx=11
+                                         --disable-python --enable-cxx=${PLUMED_CXX_STANDARD}
                                          --enable-modules=-adjmat:+crystallization:-dimred:+drr:+eds:-fisst:+funnel:+logmfd:+manyrestraints:+maze:+opes:+multicolvar:-pamm:-piv:+s2cm:-sasa:-ves
                                          ${PLUMED_CONFIG_OMP}
                                          ${PLUMED_CONFIG_MPI}
@@ -142,7 +149,7 @@ else()
       CONFIGURE_COMMAND <SOURCE_DIR>/configure --prefix=<INSTALL_DIR>
                                              ${CONFIGURE_REQUEST_PIC}
                                              --enable-modules=all
-                                             --enable-cxx=11
+                                             --enable-cxx=${PLUMED_CXX_STANDARD}
                                              --disable-python
                                              ${PLUMED_CONFIG_MPI}
                                              ${PLUMED_CONFIG_OMP}

cmake/Modules/Packages/PYTHON.cmake

@@ -1,6 +1,6 @@
 
 if(NOT Python_INTERPRETER)
-  # backward compatibility with CMake before 3.12 and older LAMMPS documentation
+  # backward compatibility with older LAMMPS documentation
   if(PYTHON_EXECUTABLE)
     set(Python_EXECUTABLE ${PYTHON_EXECUTABLE})
   endif()

doc/README

@@ -22,12 +22,12 @@ doxygen-warn.log  logfile with warnings from running doxygen
 and:
 
 github-development-workflow.md   notes on the LAMMPS development workflow
-include-file-conventions.md      notes on LAMMPS' include file conventions
 documentation_conventions.md     notes on writing documentation for LAMMPS
 
 If you downloaded a LAMMPS tarball from www.lammps.org, then the html
 folder and the PDF manual should be included. If you downloaded LAMMPS
-from GitHub then you either need to build them.
+using GitHub then you either need to build them yourself or read the
+online version at https://docs.lammps.org/
 
 You can build the HTML and PDF files yourself, by typing "make html"
 or by "make pdf", respectively.  This requires various tools and files.
@@ -39,10 +39,10 @@ environment and local folders.
 
 Installing prerequisites for the documentation build
 
-To run the HTML documention build toolchain, python 3.x, doxygen, git,
+To run the HTML documention build toolchain, python 3.8 or later,
-and the venv python module have to be installed if not already available.
+doxygen 1.8.10 or later, git, and the venv python module have to be
-Also internet access is initially required to download external files
+installed if not already available.  Also internet access is initially
-and tools.
+required to download external files and tools.
 
 Building the PDF format manual requires in addition a compatible LaTeX
 installation with support for PDFLaTeX and several add-on LaTeX packages
@@ -52,16 +52,24 @@ installed.  This includes:
 - babel
 - capt-of
 - cmap
+- dvipng
+- ellipse
 - fncychap
+- fontawesom
 - framed
 - geometry
+- gyre
 - hyperref
 - hypcap
 - needspace
+- pict2e
 - times
 - tabulary
+- titlesec
 - upquote
 - wrapfig
+- xindy
+
 Also the latexmk script is required to run PDFLaTeX and related tools.
 the required number of times to have self-consistent output and include
 updated bibliography and indices.

doc/src/Build.rst

@@ -14,6 +14,29 @@ As an alternative, you can download a package with pre-built executables
 or automated build trees, as described in the :doc:`Install <Install>`
 section of the manual.
 
+Prerequisites
+-------------
+
+Which software you need to compile and use LAMMPS strongly depends on
+which :doc:`features and settings <Build_settings>` and which
+:doc:`optional packages <Packages_list>` you are trying to include.
+Common to all is that you need a C++ and C compiler, where the C++
+compiler has to support at least the C++11 standard (note that some
+compilers require command-line flag to activate C++11 support).
+Furthermore, if you are building with CMake, you need at least CMake
+version 3.20 and a compatible build tool (make or ninja-build); if you
+are building the the legacy GNU make based build system you need GNU
+make (other make variants are not going to work since the build system
+uses features unique to GNU make) and a Unix-like build environment with
+a Bourne shell, and shell tools like "sed", "grep", "touch", "test",
+"tr", "cp", "mv", "rm", "ln", "diff" and so on. Parts of LAMMPS
+interface with or use Python version 3.6 or later.
+
+The LAMMPS developers aim to keep LAMMPS very portable and usable -
+at least in parts - on most operating systems commonly used for
+running MD simulations.  Please see the :doc:`section on portablility
+<Intro_portability>` for more details.
+
 .. toctree::
    :maxdepth: 1
 

doc/src/Build_cmake.rst

@@ -52,9 +52,9 @@ software or for people that want to modify or extend LAMMPS.
   compilers can be configured and built concurrently from the same
   source tree.
 - Simplified packaging of LAMMPS for Linux distributions, environment
-  modules, or automated build tools like `Homebrew <https://brew.sh/>`_.
+  modules, or automated build tools like `Spack <https://spack.io>`_
-- Integration of automated unit and regression testing (the LAMMPS side
+  or `Homebrew <https://brew.sh/>`_.
-  of this is still under active development).
+- Integration of automated unit and regression testing.
 
 .. _cmake_build:
 

doc/src/Build_extras.rst

@@ -1139,11 +1139,10 @@ POEMS package
 PYTHON package
 ---------------------------
 
-Building with the PYTHON package requires you have a the Python development
+Building with the PYTHON package requires you have a the Python
-headers and library available on your system, which needs to be a Python 2.7
+development headers and library available on your system, which
-version or a Python 3.x version.  Since support for Python 2.x has ended,
+needs to be Python version 3.6 or later.  See ``lib/python/README``
-using Python 3.x is strongly recommended. See ``lib/python/README`` for
+for additional details.
-additional details.
 
 .. tabs::
 
@@ -1159,7 +1158,7 @@ additional details.
       set the Python_EXECUTABLE variable to specify which Python
       interpreter should be used.  Note note that you will also need to
       have the development headers installed for this version,
-      e.g. python2-devel.
+      e.g. python3-devel.
 
    .. tab:: Traditional make
 

doc/src/Build_make.rst

@@ -30,9 +30,9 @@ additional tools to be available and functioning.
   * A Bourne shell compatible "Unix" shell program (frequently this is ``bash``)
   * A few shell utilities: ``ls``, ``mv``, ``ln``, ``rm``, ``grep``, ``sed``, ``tr``, ``cat``, ``touch``, ``diff``, ``dirname``
   * Python (optional, required for ``make lib-<pkg>`` in the ``src``
-    folder).  Python scripts are currently tested with python 2.7 and
+    folder).  Python scripts are currently tested with 3.6 to 3.11.
-    3.6 to 3.11. The procedure for :doc:`building the documentation
+    The procedure for :doc:`building the documentation <Build_manual>`
-    <Build_manual>` *requires* Python 3.5 or later.
+    *requires* Python 3.8 or later.
 
 Getting started
 ^^^^^^^^^^^^^^^

doc/src/Build_manual.rst

@@ -116,9 +116,9 @@ environment variable.
 Prerequisites for HTML
 ----------------------
 
-To run the HTML documentation build toolchain, python 3, git, doxygen,
+To run the HTML documentation build toolchain, Python 3.8 or later, git,
-and virtualenv have to be installed locally.  Here are instructions for
+doxygen, and virtualenv have to be installed locally.  Here are
-common setups:
+instructions for common setups:
 
 .. tabs::
 
@@ -128,13 +128,7 @@ common setups:
 
          sudo apt-get install git doxygen
 
-   .. tab:: RHEL or CentOS (Version 7.x)
+   .. tab:: Fedora or RHEL/AlmaLinux/RockyLinux (8.x or later)
-
-      .. code-block:: bash
-
-         sudo yum install git doxygen
-
-   .. tab:: Fedora or RHEL/CentOS (8.x or later)
 
       .. code-block:: bash
 
@@ -154,7 +148,36 @@ Prerequisites for PDF
 
 In addition to the tools needed for building the HTML format manual,
 a working LaTeX installation with support for PDFLaTeX and a selection
-of LaTeX styles/packages are required.  To run the PDFLaTeX translation
+of LaTeX styles/packages are required.  Apart from LaTeX packages that
+are usually installed by default, the following packages are required:
+
+.. table_from_list::
+   :columns: 11
+
+   - amsmath
+   - anysize
+   - babel
+   - capt-of
+   - cmap
+   - dvipng
+   - ellipse
+   - fncychap
+   - fontawesome
+   - framed
+   - geometry
+   - gyre
+   - hyperref
+   - hypcap
+   - needspace
+   - pict2e
+   - times
+   - tabulary
+   - titlesec
+   - upquote
+   - wrapfig
+   - xindy
+
+To run the PDFLaTeX translation
 the ``latexmk`` script needs to be installed as well.
 
 Prerequisites for ePUB and MOBI

doc/src/Build_settings.rst

@@ -8,7 +8,7 @@ Optional build settings
 LAMMPS can be built with several optional settings.  Each subsection
 explains how to do this for building both with CMake and make.
 
-* `C++11 standard compliance`_ when building all of LAMMPS
+* `C++11 and C++17 standard compliance`_ when building all of LAMMPS
 * `FFT library`_ for use with the :doc:`kspace_style pppm <kspace_style>` command
 * `Size of LAMMPS integer types and size limits`_
 * `Read or write compressed files`_
@@ -23,14 +23,15 @@ explains how to do this for building both with CMake and make.
 
 .. _cxx11:
 
-C++11 standard compliance
+C++11 and C++17 standard compliance
--------------------------
+-----------------------------------
 
-A C++11 standard compatible compiler is a requirement for compiling LAMMPS.
+A C++11 standard compatible compiler is currently the minimum
-LAMMPS version 3 March 2020 is the last version compatible with the previous
+requirement for compiling LAMMPS.  LAMMPS version 3 March 2020 is the
-C++98 standard for the core code and most packages. Most currently used
+last version compatible with the previous C++98 standard for the core
-C++ compilers are compatible with C++11, but some older ones may need extra
+code and most packages. Most currently used C++ compilers are compatible
-flags to enable C++11 compliance.  Example for GNU c++ 4.8.x:
+with C++11, but some older ones may need extra flags to enable C++11
+compliance.  Example for GNU c++ 4.8.x:
 
 .. code-block:: make
 
@@ -40,6 +41,17 @@ Individual packages may require compliance with a later C++ standard
 like C++14 or C++17.  These requirements will be documented with the
 :doc:`individual packages <Packages_details>`.
 
+.. versionchanged:: 4Feb2025
+
+Starting with LAMMPS version 4 February 2025 we are starting a
+transition to require the C++17 standard.  Most current compilers are
+compatible and if the C++17 standard is available by default, LAMMPS
+will enable C++17 and will compile normally.  If the chosen compiler is
+not compatible with C++17, but only supports C++11, then the define
+-DLAMMPS_CXX11 is required to fall back to compiling with a C++11
+compiler.  After the next stable release of LAMMPS in summer 2025, the
+LAMMPS development branch and future releases will require C++17.
+
 ----------
 
 .. _fft:

doc/src/Commands_all.rst

@@ -140,6 +140,7 @@ additional letter in parenthesis: k = KOKKOS.
    * :doc:`plugin <plugin>`
    * :doc:`prd <prd>`
    * :doc:`python <python>`
+   * :doc:`region2vmd <region2vmd>`
    * :doc:`tad <tad>`
    * :doc:`temper <temper>`
    * :doc:`temper/grem <temper_grem>`

doc/src/Intro_portability.rst

@@ -13,10 +13,14 @@ Programming language standards
 
 Most of the C++ code currently requires a compiler compatible with the
 C++11 standard, the KOKKOS package currently requires C++17.  Most of
-the Python code is written to be compatible with Python 3.5 or later or
+the Python code is written to be compatible with Python 3.6 or later.
-Python 2.7.  Some Python scripts *require* Python 3 and a few others
-still need to be ported from Python 2 to Python 3.
 
+.. deprecated:: TBD
+
+Python 2.x is no longer supported and trying to use it, e.g. for the
+LAMMPS Python module should result in an error.  If you come across
+some part of the LAMMPS distribution that is not (yet) compatible with
+Python 3, please notify the LAMMPS developers.
 
 Build systems
 ^^^^^^^^^^^^^
@@ -24,8 +28,8 @@ Build systems
 LAMMPS can be compiled from source code using a (traditional) build
 system based on shell scripts, a few shell utilities (grep, sed, cat,
 tr) and the GNU make program. This requires running within a Bourne
-shell (``/bin/sh``).  Alternatively, a build system with different back ends
+shell (``/bin/sh``).  Alternatively, a build system with different back
-can be created using CMake.  CMake must be at least version 3.16.
+ends can be created using CMake.  CMake must be at least version 3.16.
 
 Operating systems
 ^^^^^^^^^^^^^^^^^

doc/src/Modify_requirements.rst

@@ -189,10 +189,8 @@ of the contribution.  As of January 2023, all previously included
 Fortran code for the LAMMPS executable has been replaced by equivalent
 C++ code.
 
-Python code must be compatible with Python 3.5 and later.  Large parts
+Python code currently must be compatible with Python 3.6.  If a later
-of LAMMPS (including the :ref:`PYTHON package <PKG-PYTHON>`) are also
+version or Python is required, it needs to be documented.
-compatible with Python 2.7.  Compatibility with Python 2.7 is desirable,
-but compatibility with Python 3.5 is **required**.
 
 Compatibility with older programming language standards is very
 important to maintain portability and availability of LAMMPS on many

doc/src/Packages_details.rst

@@ -2428,7 +2428,7 @@ ways to use LAMMPS and Python together.
 
    Building with the PYTHON package assumes you have a Python development
    environment (headers and libraries) available on your system, which needs
-   to be either Python version 2.7 or Python 3.5 and later.
+   to be Python version 3.6 or later.
 
 **Install:**
 

doc/src/Python_install.rst

@@ -7,6 +7,10 @@ LAMMPS shared library through the Python `ctypes <ctypes_>`_
 module.  Because of the dynamic loading, it is required that LAMMPS is
 compiled in :ref:`"shared" mode <exe>`.
 
+.. versionchanged:: TBD
+
+LAMMPS currently only supports Python version 3.6 or later.
+
 Two components are necessary for Python to be able to invoke LAMMPS code:
 
 * The LAMMPS Python Package (``lammps``) from the ``python`` folder
@@ -136,11 +140,6 @@ folder that the dynamic loader searches or inside of the installed
          # create virtual environment in folder $HOME/myenv
          python3 -m venv $HOME/myenv
 
-      For Python versions prior 3.3 you can use `virtualenv
-      <https://packaging.python.org/en/latest/key_projects/#virtualenv>`_
-      command instead of "python3 -m venv".  This step has to be done
-      only once.
-
       To activate the virtual environment type:
 
       .. code-block:: bash
@@ -245,14 +244,14 @@ make MPI calls directly from Python in your script, if you desire.
 We have tested this with `MPI for Python <https://mpi4py.readthedocs.io/>`_
 (aka mpi4py) and you will find installation instruction for it below.
 
-Installation of mpi4py (version 3.0.3 as of Sep 2020) can be done as
+Installation of mpi4py (version 4.0.1 as of Feb 2025) can be done as
 follows:
 
 - Via ``pip`` into a local user folder with:
 
   .. code-block:: bash
 
-     pip install --user mpi4py
+     python3 -m pip install --user mpi4py
 
 - Via ``dnf`` into a system folder for RedHat/Fedora systems:
 
@@ -261,20 +260,20 @@ follows:
      # for use with OpenMPI
      sudo dnf install python3-mpi4py-openmpi
      # for use with MPICH
-     sudo dnf install python3-mpi4py-openmpi
+     sudo dnf install python3-mpi4py-mpich
 
 - Via ``pip`` into a virtual environment (see above):
 
   .. code-block:: console
 
      $ source $HOME/myenv/activate
-     (myenv)$ pip install mpi4py
+     (myenv)$ python -m pip install mpi4py
 
 - Via ``pip`` into a system folder (not recommended):
 
   .. code-block:: bash
 
-     sudo pip install mpi4py
+     sudo python3 -m pip install mpi4py
 
 For more detailed installation instructions and additional options,
 please see the `mpi4py installation <https://mpi4py.readthedocs.io/en/stable/install.html>`_ page.

doc/src/Python_overview.rst

@@ -44,15 +44,11 @@ Below is an example output for Python version 3.8.5.
 .. warning::
 
    The options described in this section of the manual for using Python
-   with LAMMPS currently support either Python 2 or 3.  Specifically
+   with LAMMPS support only Python 3.6 or later.  For use with Python
-   version 2.7 or later and 3.6 or later.  Since the Python community no
+   2.x you will need to use an older LAMMPS version like 29 Aug 2024
-   longer maintains Python 2 (see `this notice
+   or older.  If you notice Python code in the LAMMPS distribution that
-   <https://www.python.org/doc/sunset-python-2/>`_), we recommend use of
+   is not compatible with Python 3, please contact the LAMMPS developers
-   Python 3 with LAMMPS.  While Python 2 code should continue to work,
+   or submit `and issue on GitHub <https://github.com/lammps/lammps/issues>`_
-   that is not something we can guarantee long-term.  If you notice
-   Python code in the LAMMPS distribution that is not compatible with
-   Python 3, please contact the LAMMPS developers or submit `and issue
-   on GitHub <https://github.com/lammps/lammps/issues>`_
 
 ---------
 

doc/src/commands_list.rst

@@ -82,6 +82,7 @@ Commands
    read_dump
    read_restart
    region
+   region2vmd
    replicate
    rerun
    reset_atoms

doc/src/region2vmd.rst

@@ -0,0 +1,179 @@
+.. index:: region2vmd
+
+region2vmd command
+==================
+
+Syntax
+""""""
+
+.. code-block:: LAMMPS
+
+   region2vmd file keyword arg ...
+
+* filename = name of file to write VMD script commands to
+* zero or more keyword/arg pairs may be appended
+* keyword = *region* or *color* or *material* or *command*
+
+  .. parsed-literal::
+
+     *region* region-ID = name of region to translate to VMD graphics
+     *color* color-name = set color for following visualized objects
+     *material* material-name = set material for following visualized objects
+     *command* string = string with custom VMD script command (in quotes)
+
+Examples
+""""""""
+
+.. code-block:: LAMMPS
+
+   region2vmd regions.vmd material Opaque color red region c1 color green region c2
+   region2vmd vizbox.vmd command "mol new system.lammpstrj waitfor all" region box
+   region2vmd regdefs.vmd region upper region lower region hole
+
+Description
+"""""""""""
+
+.. versionadded:: TBD
+
+Write a `VMD <https:://ks.uiuc.edu/Research/vmd/>`_ Tcl script file with
+commands that aim to create a visualization of :doc:`regions <region>`.
+There may be multiple region visualizations stored in a single file.
+
+The visualization is implemented by creating a new (and empty) "VMD
+molecule" and then assigning a sequence of VMD graphics primitives to
+represent the region in VMD.  Each region will be stored in a separate
+"VMD molecule" with the name "LAMMPS region <region ID>".
+
+The *region2vmd* command is following by the filename for the resulting
+VMD script and an arbitrary number of keyword argument pairs to either
+write out a new *region* visualization, change the *color* or *material*
+setting, or to insert arbitrary VMD script *command*\ s.  The keywords
+and arguments are processed in sequence.
+
+The *region* keyword must be followed by a previously defined LAMMPS
+:doc:`region <region>`.  Only a limited set region styles and region
+settings are currently supported.  See **Restrictions** below.
+Unsupported region styles or regions with unsupported settings will be
+skipped and a corresponding message is printed.
+
+The *color* keyword must be followed by a color name that is defined in
+VMD.  This color will be used by all following region visualizations.
+The default setting is 'silver'. VMD has the following colors
+pre-defined:
+
+.. table_from_list::
+   :columns: 11
+
+   * blue
+   * red
+   * gray
+   * orange
+   * yellow
+   * tan
+   * silver
+   * green
+   * white
+   * pink
+   * cyan
+   * purple
+   * lime
+   * mauve
+   * ochre
+   * iceblue
+   * black
+   * yellow2
+   * yellow3
+   * green2
+   * green3
+   * cyan2
+   * cyan3
+   * blue2
+   * blue3
+   * violet
+   * violet2
+   * magenta
+   * magenta2
+   * red2
+   * red3
+   * orange2
+   * orange3
+
+The *material* keyword must be followed by a material name that is defined in
+VMD.  This material will be used by all following visualizations.  The
+default setting is 'Transparent'.  VMD has the following materials
+pre-defined:
+
+.. table_from_list::
+   :columns: 8
+
+   * Opaque
+   * Transparent
+   * BrushedMetal
+   * Diffuse
+   * Ghost
+   * Glass1
+   * Glass2
+   * Glass3
+   * Glossy
+   * HardPlastic
+   * MetallicPastel
+   * Steel
+   * Translucent
+   * Edgy
+   * EdgyShiny
+   * EdgyGlass
+   * Goodsell
+   * AOShiny
+   * AOChalky
+   * AOEdgy
+   * BlownGlass
+   * GlassBubble
+   * RTChrome
+
+The *command* keyword must be followed by a VMD script command as a
+single string in quotes.  This VMD command will be directly inserted
+into the created VMD script.
+
+The created file can be loaded into VMD either from the command line
+with the '-e' flag, or from the command prompt with 'play <script
+file>', or from the File menu via "Load VMD visualization state".
+
+.. admonition:: Setting the "top" molecule in VMD
+   :class: note
+
+   It is usually desirable to have the "molecule" with the LAMMPS
+   trajectory set at "top" molecule in VMD and not one of the "region
+   molecules".  The VMD script generated by this region2vmd assumes that
+   this molecule is already loaded and set as the current "top"
+   molecule.  Thus at the beginning of the script the index of the top
+   molecule is stored in the VMD variable 'oldtop' and at the end of the
+   script, that "top" molecule is restored.  If no molecule is loaded,
+   this can be inserted into the script with a custom command. The
+   molecule index to this new molecules should be assigned to the oldtop
+   variable.  This can be done with e.g.  ``set oldtop [mol new
+   {regions.vmd} waitfor all]``
+
+----------
+
+Restrictions
+""""""""""""
+
+This command is part of the EXTRA-COMMAND package.  It is only enabled
+if LAMMPS was built with that package.  See the :doc:`Build package
+<Build_package>` page for more info.
+
+Only the following region styles are currently supported: *block*,
+*cone*, *cylinder*, *ellipsoid*, *prism*, and *sphere*. Regions formed
+from unions or intersections of other regions are not supported.
+
+Rotating regions are currently not supported.
+
+Related commands
+""""""""""""""""
+
+:doc:`region <region>`
+
+Defaults
+""""""""
+
+*color* = silver, *material* = Transparent

doc/utils/sphinx-config/false_positives.txt

@@ -99,6 +99,7 @@ AMD
 amino
 Amirjalayer
 Amit
+amsmath
 amu
 Amzallag
 analytical
@@ -128,6 +129,7 @@ Antisymmetrized
 antisymmetry
 anton
 Antonelli
+anysize
 api
 apolar
 Apoorva
@@ -613,6 +615,7 @@ Courant
 covalent
 covalently
 covariance
+cp
 cpp
 cpu
 cradius
@@ -905,6 +908,7 @@ dUs
 Duval
 dV
 dvector
+dvipng
 dVx
 dW
 dx
@@ -1215,12 +1219,14 @@ fmz
 fN
 Fn
 fname
+fncychap
 fno
 Fnudge
 foces
 Fock
 Fogarty
 Foiles
+fontawesome
 fopenmp
 forceclear
 forestgreen
@@ -1353,6 +1359,7 @@ Goga
 Goldfarb
 Gompper
 Gonzalez-Melchor
+Goodsell
 googlemail
 googletest
 Gordan
@@ -1406,6 +1413,7 @@ Gunsteren
 Gunzenmuller
 Guo
 gw
+gyre
 gyromagnetic
 gz
 gzip
@@ -1531,10 +1539,12 @@ hydrostatically
 hydroxyl
 Hynninen
 Hyoungki
+hypcap
 hyperdynamics
 hyperparameters
 hyperplane
 hyperradius
+hyperref
 hyperspherical
 hysteretic
 hz
@@ -1544,6 +1554,7 @@ Ibanez
 ibar
 ibm
 icc
+iceblue
 ico
 icosahedral
 idealgas
@@ -2512,6 +2523,7 @@ Ndof
 Ndouble
 ndx
 neb
+needspace
 neel
 Neel
 Neelov
@@ -2724,6 +2736,7 @@ nylo
 nz
 Nz
 nzlo
+ochre
 ocl
 octahedral
 octants
@@ -2738,6 +2751,7 @@ Okazaki
 O'Keefe
 OKeefe
 oldlace
+oldtop
 olecular
 Oleinik
 Olfason
@@ -2914,6 +2928,7 @@ picogram
 picograms
 picosecond
 picoseconds
+pict
 pid
 piecewise
 Pieniazek
@@ -3685,6 +3700,7 @@ Sz
 Tabbernor
 tabinner
 tabstyle
+tabulary
 Tadmor
 Tafipolsky
 tagID
@@ -3782,6 +3798,7 @@ timestamps
 timestep
 timestepping
 timesteps
+titlesec
 TiN
 TiO
 Tirado
@@ -3958,6 +3975,7 @@ untilted
 Unwin
 uparrow
 upenn
+upquote
 upto
 Urbakh
 Urbana
@@ -4144,6 +4162,7 @@ Workum
 Worley
 wormlike
 wpbe
+wrapfig
 Wriggers
 writedata
 Wuppertal
@@ -4174,6 +4193,7 @@ Xia
 Xiaohu
 Xiaowang
 Xie
+xindy
 xk
 xlat
 xlattice

lib/python/Makefile.lammps.python2

@@ -1,7 +0,0 @@
-# Settings that the LAMMPS build will import when this package library is used
-# See the README file for more explanation
-
-python_SYSINC = $(shell which python2-config > /dev/null 2>&1 && python2-config --includes || (which python-config > /dev/null 2>&1 && python-config --includes || :))
-python_SYSLIB = $(shell which python2-config > /dev/null 2>&1 && python2-config --ldflags || (which python-config > /dev/null 2>&1 && python-config --ldflags || :))
-python_SYSPATH =
-PYTHON=$(shell which python2 > /dev/null 2>&1 && echo python2 || echo python)

lib/python/README

@@ -9,30 +9,29 @@ installation. If needed, you can copy one of the other provided
 Makefile.lammps.* files to to Makefile.lammps before building
 LAMMPS itself.
 
-The files Makefile.lammps.python2 and Makefile.lammps.python3 are
+The file Makefile.lammps.python3 is similar to the default file, but
-similar to the default file, but meant for the case that both,
+meant for the case that both, python 2 and python 3, are installed
-python 2 and python 3, are installed simultaneously and you want
+simultaneously.  LAMMPS only supports python 3.  If neither of these
-to prefer one over the other. If neither of these files work, you
+files work, you may have to create a custom Makefile.lammps file
-may have to create a custom Makefile.lammps file suitable for
+suitable for the version of Python on your system.  To illustrate, these
-the version of Python on your system.  To illustrate, these are
+are example settings from the Makefile.lammps.python3.13 file:
-example settings from the Makefile.lammps.python2.7 file:
 
-python_SYSINC = -I/usr/local/include/python2.7
+python_SYSINC = -I/usr/local/include/python3.13
-python_SYSLIB = -lpython2.7 -lnsl -ldl -lreadline -ltermcap -lpthread -lutil -lm
+python_SYSLIB = -lpython3.13 -ldl  -lm
-python_SYSPATH =
+python_SYSPATH = -L/usr/lib64
-PYTHON=python2.7
+PYTHON=python3.13
 
 python_SYSINC refers to the directory where Python's Python.h file is
 found.  LAMMPS includes this file.
 
 python_SYSLIB refers to the libraries needed to link to from an
 application (LAMMPS in this case) to "embed" Python in the
-application.  The Python library itself is listed (-lpython2.7) are
+application.  The Python library itself is listed (-lpython3.13) are
 are several system libraries needed by Python.
 
 python_SYSPATH refers to the path (e.g. -L/usr/local/lib) where the
 Python library can be found.  You may not need this setting if the
-path is already included in your LD_LIBRARY_PATH environment variable.
+path is already included in your LIBRARY_PATH environment variable.
 
 PYTHON is the name of the python interpreter. It is used for
 installing the LAMMPS python module with "make install-python"
@@ -45,7 +44,7 @@ run in embedded mode on your machine.
 
 Here is what this Python doc page says about it:
 
-https://docs.python.org/2/extending/embedding.html#compiling-and-linking-under-unix-like-systems
+https://docs.python.org/3/extending/embedding.html#compiling-and-linking-under-unix-like-systems
 
 "It is not necessarily trivial to find the right flags to pass to your
 compiler (and linker) in order to embed the Python interpreter into

python/README

@@ -1,20 +1,27 @@
-This directory contains Python code which wraps LAMMPS as a library
+This directory contains the LAMMPS Python module that allows the LAMMPS
-and allows the LAMMPS library interface to be invoked from Python,
+C library interface to be invoked from Python, either from a Python
-either from a Python script or using Python interactively.
+script or using Python interactively.
 
 Details on the Python interface to LAMMPS and how to build LAMMPS as a
 shared library, for use with Python, are given in
-doc/Section_python.html and in doc/Section_start.html#start_5.
+https://docs.lammps.org/Build_basics.html#exe
+and
+https://docs.lammps.org/Python_install.html
 
-Basically you need to follow these steps in the src directory:
+Basically you need to use the flag -D BUILD_SHARED_LIBS=ON when
+configuring LAMMPS with CMake and then in the build folder use the
+command % cmake --build . --target install-python
+
+or for the legacy GNU build system execute these steps in the src folder:
 
 % make g++ mode=shlib           # build for whatever machine target you wish
 % make install-python           # install into site-packages folder
 
 You can replace the last step by a one-time setting of environment
-variables in your shell script.  Or you can run the python/install.py
+variables in your shell environment.  Or you can run the
-script directly to give you more control over where the two relevant
+python/install.py script directly to give you more control over where
-files are installed.  See doc/Python_install.html for details.
+the two relevant files are installed.  See
+https://docs.lammps.org/Python_install.html for details.
 
 You should then be able to launch Python and instantiate an instance
 of LAMMPS:
@@ -24,39 +31,41 @@ of LAMMPS:
 >>> lmp = lammps()
 
 If that gives no errors, you have successfully wrapped LAMMPS with
-Python.  See doc/Section_python.html#py_7 for tests you can then use
+Python.  See https://docs.lammps.org/Python_launch.html for examples how
-to run LAMMPS both in serial or parallel thru Python.
+to run LAMMPS both in serial or parallel from Python.
 
 Note that you can also invoke Python code from within a LAMMPS input
-script, using the "python" command.  See the doc/python.html doc page
+script, using the "python" command.  See the
-for details.  The Python code you invoke can also call back to LAMMPS
+https://docs.lammps.org/python.html doc page for details.  The Python
-using the same interface described here for wrapping LAMMPS.
+code you invoke can also call back to LAMMPS using the same interface
+described here for wrapping LAMMPS.
 
 -------------------------------------------------------------------
 
 Once you have successfully wrapped LAMMPS, you can run the Python
 scripts in the examples sub-directory:
 
-trivial.py          read/run a LAMMPS input script thru Python
+trivial.py            read/run a LAMMPS input script thru Python
-demo.py             invoke various LAMMPS library interface routines
+demo.py               invoke various LAMMPS library interface routines
-simple.py           parallel example, mimicing examples/COUPLE/simple/simple.cpp
+simple.py             parallel example, mimicing examples/COUPLE/simple/simple.cpp
-split.py            parallel example
+split.py              parallel example
-mc.py               Monte Carlo energy relaxation wrapper on LAMMPS
+mc.py                 Monte Carlo energy relaxation wrapper on LAMMPS
-gui.py              GUI go/stop/temperature-slider to control LAMMPS
+gui.py                GUI go/stop/temperature-slider to control LAMMPS
-plot.py             real-time temperature plot with GnuPlot via Pizza.py
+plot.py               real-time temperature plot with GnuPlot via Pizza.py
-matplotlib_plot.py  real-time temperature plot with Matplotlib via Pizza.py
+matplotlib_plot.py    real-time temperature plot with Matplotlib via Pizza.py
-viz_tool.py         real-time viz via some viz package
+viz_<tool>.py         real-time viz via some viz package
-vizplotgui_tool.py  combination of viz.py and plot.py and gui.py
+vizplotgui_<tool>.py  combination of viz.py and plot.py and gui.py
 
-For the viz_tool.py and vizplotgui_tool.py commands, replace "tool"
+For the viz_<tool>.py and vizplotgui_<tool>.py commands, replace
-with "gl" or "atomeye" or "pymol", depending on what visualization
+"<tool>" with "gl" or "atomeye" or "pymol", depending on what
-package you have installed.  We hope to add a VMD option soon.
+visualization package you have installed.  We hope to add a VMD option
+soon.
 
 Note that for GL, you need to be able to run the Pizza.py GL tool,
 which is included in the pizza sub-directory.  See the Pizza.py doc
 pages for more info:
 
-http://www.sandia.gov/~sjplimp/pizza.html
+https://lammps.github.io/pizza/
 
 Note that for AtomEye, you need version 3, and their is a line in the
 scripts that specifies the path and name of the executable.  See
@@ -115,17 +124,14 @@ one-processor run, where both Python and LAMMPS will run on single
 processors.  Each running job will read the same input file, and write
 to same log.lammps file, which isn't too useful.
 
-However, if you have the mpi4py Python package installed and uncomment mpi4py
+However, if you have the mpi4py Python package installed and uncomment
-code in simple.py, then the above commands will invoke 1 instance of a
+mpi4py code in simple.py, then the above commands will invoke 1 instance
-P-processor run.  Both Python and LAMMPS will run on P processors.  The job will
+of a P-processor run.  Both Python and LAMMPS will run on P processors.
-read the input file and write a single log.lammps file.
+The job will read the input file and write a single log.lammps file.
-
-The split.py script can also be run in parallel.  It uses mpi4py
-version 2.0.0 (or later), which makes it possible to pass a
-communicator when creating the LAMMPS object and thus run multiple
-instances of LAMMPS at the same time, each on a different subset of
-MPI ranks.  Or run LAMMPS on one subset and some other program on the
-rest of the MPI ranks, concurrently.  See comments in the split.py
-script for more details.
-
 
+The split.py script can also be run in parallel.  It uses mpi4py version
+2.0.0 (or later), which makes it possible to pass a communicator when
+creating the LAMMPS object and thus run multiple instances of LAMMPS at
+the same time, each on a different subset of MPI ranks.  Or run LAMMPS
+on one subset and some other program on the rest of the MPI ranks,
+concurrently.  See comments in the split.py script for more details.

python/lammps/__init__.py

@@ -23,6 +23,9 @@ def get_version_number():
     if __file__.find(join('python', 'lammps', '__init__.py')) > 0:
         return 0
 
+    if version_info.major < 3 or (version_info.major == 3 and version_info.minor < 6):
+        raise SystemError('LAMMPS only supports Python version 3.6 or later')
+
     vstring = None
     if version_info.major == 3 and version_info.minor >= 8:
         from importlib.metadata import version, PackageNotFoundError

python/lammps/core.py

@@ -83,8 +83,6 @@ class command_wrapper(object):
 
   def _wrap_args(self, x):
       if callable(x):
-          if sys.version_info <  (3,):
-            raise Exception("Passing functions or lambdas directly as arguments is only supported in Python 3 or newer")
           import hashlib
           import __main__
           sha = hashlib.sha256()
@@ -105,11 +103,6 @@ class command_wrapper(object):
     all the arguments, concatinates them to a single string, and executes it using
     :py:meth:`lammps.command`.
 
-    Starting with Python 3.6 it also supports keyword arguments. key=value is
-    transformed into 'key value'. Note, since these have come last in the
-    parameter list, only a subset of LAMMPS commands can be used with this
-    syntax.
-
     LAMMPS commands that accept callback functions (such as fix python/invoke)
     can be passed functions and lambdas directly. The first argument of such
     callbacks will be an lammps object constructed from the passed LAMMPS
@@ -121,9 +114,6 @@ class command_wrapper(object):
     def handler(*args, **kwargs):
       cmd_args = [name] + [str(self._wrap_args(x)) for x in args]
 
-      if len(kwargs) > 0 and sys.version_info < (3,6):
-         raise Exception("Keyword arguments are only supported in Python 3.6 or newer")
-
       # Python 3.6+ maintains ordering of kwarg keys
       for k in kwargs.keys():
           cmd_args.append(k)
@@ -530,16 +520,10 @@ class lammps(object):
 
     else:
       # magic to convert ptr to ctypes ptr
-      if sys.version_info >= (3, 0):
+      # Python 3 (uses PyCapsule API)
-        # Python 3 (uses PyCapsule API)
+      pythonapi.PyCapsule_GetPointer.restype = c_void_p
-        pythonapi.PyCapsule_GetPointer.restype = c_void_p
+      pythonapi.PyCapsule_GetPointer.argtypes = [py_object, c_char_p]
-        pythonapi.PyCapsule_GetPointer.argtypes = [py_object, c_char_p]
+      self.lmp = c_void_p(pythonapi.PyCapsule_GetPointer(ptr, None))
-        self.lmp = c_void_p(pythonapi.PyCapsule_GetPointer(ptr, None))
-      else:
-        # Python 2 (uses PyCObject API)
-        pythonapi.PyCObject_AsVoidPtr.restype = c_void_p
-        pythonapi.PyCObject_AsVoidPtr.argtypes = [py_object]
-        self.lmp = c_void_p(pythonapi.PyCObject_AsVoidPtr(ptr))
 
     # check if library initilialization failed
     if not self.lmp:

src/.gitignore

@@ -1549,6 +1549,8 @@
 /reaxff_inline.h
 /reaxff_omp.h
 /reaxff_types.h
+/region2vmd.cpp
+/region2vmd.h
 /remap.cpp
 /remap.h
 /remap_wrap.cpp

src/EXTRA-COMMAND/region2vmd.cpp

@@ -0,0 +1,893 @@
+/* -*- c++ -*--------------------------------------------------------------
+   LAMMPS - Large-scale Atomic/Molecular Massively Parallel Simulator
+   https://www.lammps.org/, Sandia National Laboratories
+   LAMMPS development team: developers@lammps.org
+
+   Copyright (2003) Sandia Corporation.  Under the terms of Contract
+   DE-AC04-94AL85000 with Sandia Corporation, the U.S. Government retains
+   certain rights in this software.  This software is distributed under
+   the GNU General Public License.
+
+   See the README file in the top-level LAMMPS directory.
+------------------------------------------------------------------------- */
+/* ----------------------------------------------------------------------
+   Contributing author:  Axel Kohlmeyer (Temple U)
+------------------------------------------------------------------------- */
+
+#include "region2vmd.h"
+
+#include "atom.h"
+#include "comm.h"
+#include "domain.h"
+#include "error.h"
+#include "math_const.h"
+#include "math_extra.h"
+#include "region.h"
+#include "safe_pointers.h"
+
+#include "region_block.h"
+#include "region_cone.h"
+#include "region_cylinder.h"
+#include "region_ellipsoid.h"
+#include "region_plane.h"
+#include "region_prism.h"
+#include "region_sphere.h"
+
+#include <cmath>
+#include <cstring>
+#include <unordered_set>
+
+using namespace LAMMPS_NS;
+
+using MathConst::MY_2PI;
+static constexpr double SMALL = 1.0e-10;
+static constexpr double DELTA = 1.0e-5;
+static constexpr int RESOLUTION = 20;
+static constexpr double RADINC = MY_2PI / RESOLUTION;
+
+static const std::unordered_set<std::string> vmdcolors{
+    "blue",    "red",      "gray",   "orange", "yellow",  "tan",    "silver",  "green",  "white",
+    "pink",    "cyan",     "purple", "lime",   "mauve",   "ochre",  "iceblue", "black",  "yellow2",
+    "yellow3", "green2",   "green3", "cyan2",  "cyan3",   "blue2",  "blue3",   "violet", "violet2",
+    "magenta", "magenta2", "red2",   "red3",   "orange2", "orange3"};
+
+static const std::unordered_set<std::string> vmdmaterials{
+    "Opaque",      "Transparent", "BrushedMetal", "Diffuse",     "Ghost",          "Glass1",
+    "Glass2",      "Glass3",      "Glossy",       "HardPlastic", "MetallicPastel", "Steel",
+    "Translucent", "Edgy",        "EdgyShiny",    "EdgyGlass",   "Goodsell",       "AOShiny",
+    "AOChalky",    "AOEdgy",      "BlownGlass",   "GlassBubble", "RTChrome"};
+
+static constexpr char draw_ellipsoid_function[] =
+    "\n# VMD script code to emulate ellipsoids with trinorm graphics objects\n"
+    "proc vmd_draw_ellipsoid {mol level {center {0.0 0.0 0.0}} {radius {1.0 1.0 1.0}}} {\n"
+    "   set orient {1.0 0.0 0.0 0.0}\n"
+    "   set gid {}\n\n"
+    "   # vertices of an octahedron inscribed\n"
+    "   # in a sphere with radius 1.0.\n"
+    "   set vec1 {-1.0  0.0  0.0}\n"
+    "   set vec2 { 1.0  0.0  0.0}\n"
+    "   set vec3 { 0.0 -1.0  0.0}\n"
+    "   set vec4 { 0.0  1.0  0.0}\n"
+    "   set vec5 { 0.0  0.0 -1.0}\n"
+    "   set vec6 { 0.0  0.0  1.0}\n\n"
+    "   # build list of triangles representing\n"
+    "   # the octahedron. the points of the \n"
+    "   # triangles have to be given clockwise from\n"
+    "   # looking at the surface from the outside.\n"
+    "   set trilist [list \\\n"
+    "                [list $vec1 $vec4 $vec5] \\\n"
+    "                [list $vec5 $vec4 $vec2] \\\n"
+    "                [list $vec2 $vec4 $vec6] \\\n"
+    "                [list $vec6 $vec4 $vec1] \\\n"
+    "                [list $vec5 $vec3 $vec1] \\\n"
+    "                [list $vec2 $vec3 $vec5] \\\n"
+    "                [list $vec6 $vec3 $vec2] \\\n"
+    "                [list $vec1 $vec3 $vec6] ]\n\n"
+    "   # refinement iterations to approximate a sphere:\n"
+    "   # each triangle is split into 4 subtriangles\n"
+    "   #        p2\n"
+    "   #        /\\\n"
+    "   #       /  \\\n"
+    "   #    pb/____\\pc\n"
+    "   #     /\\    /\\\n"
+    "   #    /  \\  /  \\\n"
+    "   #   /____\\/____\\\n"
+    "   # p1     pa    p3\n\n"
+    "   # we construct vectors to the three midpoints and rescale\n"
+    "   # them to unit length. then build new triangles enumerating\n"
+    "   # the points in clockwise order again.\n"
+    "   for {set i 0} {$i < $level} {incr i} {\n"
+    "       set newlist {}\n"
+    "       foreach tri $trilist {\n"
+    "           foreach {p1 p2 p3} $tri {}\n"
+    "           set pa [vecnorm [vecadd $p1 $p3]]\n"
+    "           set pb [vecnorm [vecadd $p1 $p2]]\n"
+    "           set pc [vecnorm [vecadd $p2 $p3]]\n"
+    "           lappend newlist [list $p1 $pb $pa]\n"
+    "           lappend newlist [list $pb $p2 $pc]\n"
+    "           lappend newlist [list $pa $pb $pc]\n"
+    "           lappend newlist [list $pa $pc $p3]\n"
+    "       }\n"
+    "       set trilist $newlist\n"
+    "   }\n\n"
+    "   # compute vertex scaling factor to deform a sphere to an ellipsoid\n"
+    "   proc radscale {radius vector} {\n"
+    "       foreach {a b c} $radius {}\n"
+    "       foreach {x y z} $vector {}\n"
+    "       return [expr {sqrt(1.0/($x/$a*$x/$a+$y/$b*$y/$b+$z/$c*$z/$c))}]\n"
+    "   }\n\n"
+    "   # convert quaternion to rotation matrix\n"
+    "   proc quattorot {quat} {\n"
+    "       foreach {w x y z} $quat {}\n"
+    "       set norm [expr {$w*$w + $x*$x + $y*$y +$z*$z}]\n"
+    "       set sc 0.0\n"
+    "       if {$norm > 0.0} { set s [expr {2.0/$norm}] }\n"
+    "       set X [expr {$x*$s}]\n"
+    "       set Y [expr {$y*$s}]\n"
+    "       set Z [expr {$z*$s}]\n"
+    "       return [list \\\n"
+    "           [list [expr {1.0-($y*$y*$s + $z*$z*$s)}] \\\n"
+    "                 [expr {$x*$y*$s - $w*$z*$s}] \\\n"
+    "                 [expr {$x*$z*$s + $w*$y*$s}] ] \\\n"
+    "           [list [expr {$x*$y*$s + $w*$z*$s}] \\\n"
+    "                 [expr {1.0-($x*$x*$s + $z*$z*$s)}] \\\n"
+    "                 [expr {$y*$z*$s - $w*$x*$s}] ] \\\n"
+    "           [list [expr {$x*$z*$s - $w*$y*$s}] \\\n"
+    "                 [expr {$y*$z*$s + $w*$x*$s}] \\\n"
+    "                 [expr {1.0-($x*$x*$s + $y*$y*$s)}] ] ]\n"
+    "   }\n\n"
+    "   # apply rotation matrix to vector\n"
+    "   proc rotvec {mat vec} {\n"
+    "       set new {}\n"
+    "       foreach c $mat {\n"
+    "           lappend new [vecdot $c $vec]\n"
+    "       }\n"
+    "       return $new\n"
+    "   }\n\n"
+    "   foreach tri $trilist {\n"
+    "       foreach {vec1 vec2 vec3} $tri {}\n"
+    "       # rescale to desired radius\n"
+    "       set rad1 [radscale $radius $vec1]\n"
+    "       set rad2 [radscale $radius $vec2]\n"
+    "       set rad3 [radscale $radius $vec3]\n"
+    "       # get and apply rotation matrix\n"
+    "       set mat [quattorot $orient]\n"
+    "       set vec1 [rotvec $mat $vec1]\n"
+    "       set vec2 [rotvec $mat $vec2]\n"
+    "       set vec3 [rotvec $mat $vec3]\n"
+    "       # deform sphereoid and translate\n"
+    "       set pos1 [vecadd [vecscale $rad1 $vec1] $center]\n"
+    "       set pos2 [vecadd [vecscale $rad2 $vec2] $center]\n"
+    "       set pos3 [vecadd [vecscale $rad3 $vec3] $center]\n"
+    "       # since the original vectors to the vertices are those of\n"
+    "       # a unit sphere, we can use them directly as surface normals.\n"
+    "       lappend gid [graphics $mol trinorm $pos1 $pos2 $pos3 $vec1 $vec2 $vec3]\n"
+    "   }\n"
+    "   return $gid\n"
+    "}\n\n";
+
+/* ---------------------------------------------------------------------- */
+
+void Region2VMD::command(int narg, char **arg)
+{
+  if (narg < 3) utils::missing_cmd_args(FLERR, "region2vmd", error);
+
+  // automatically close file when it goes out of scope
+  SafeFilePtr fp;
+  if (comm->me == 0) {
+    fp = fopen(arg[0], "w");
+    if (fp == nullptr) {
+      error->one(FLERR, Error::ARGZERO, "Cannot open file {} for writing: {}", arg[0],
+                 utils::getsyserror());
+    } else {
+      utils::logmesg(lmp, "Writing region visualizations to VMD Tcl script file {}:\n", arg[0]);
+      fputs("# save old top molecule index\nset oldtop [molinfo top]\n", fp);
+    }
+  }
+
+  // defaults
+  std::string color = "silver";
+  std::string material = "Transparent";
+  bool def_ellipsoid_func = false;
+
+  int iarg = 1;
+  std::string thisarg = arg[iarg];
+  while (iarg < narg) {
+    if (iarg + 2 > narg) utils::missing_cmd_args(FLERR, "region2vmd", error);
+    thisarg = arg[iarg];
+    ++iarg;
+
+    if (thisarg == "color") {
+      color = arg[iarg];
+      if (const auto &search = vmdcolors.find(color); search == vmdcolors.end())
+        error->all(FLERR, iarg, "Color {} is not a known VMD color", color);
+
+    } else if (thisarg == "material") {
+      material = arg[iarg];
+      if (const auto &search = vmdmaterials.find(material); search == vmdmaterials.end())
+        error->all(FLERR, iarg, "Material {} is not a known VMD material", material);
+
+    } else if (thisarg == "command") {
+      if (fp) {
+        fputs("\n# custom command\n", fp);
+        fputs(arg[iarg], fp);
+        fputs("\n", fp);
+      }
+
+    } else if (thisarg == "region") {
+      auto *region = domain->get_region_by_id(arg[iarg]);
+      if (!region) {
+        error->all(FLERR, iarg, "Region {} does not exist", arg[iarg]);
+      } else {
+        if (fp) {
+          // add VMD / Tcl function to draw ellipsoids only once
+          if (!def_ellipsoid_func && (strcmp(region->style, "ellipsoid") == 0)) {
+            fputs(draw_ellipsoid_function, fp);
+            def_ellipsoid_func = true;
+          }
+          utils::logmesg(lmp, " writing region {} ...", region->id);
+          utils::print(fp, "\n# region {} of style {}\n", region->id, region->style);
+
+          fputs("# create new empty VMD molecule to store the graphics primitives\n"
+                "set gfxmol [mol new]\nmol top $gfxmol\n",
+                fp);
+          utils::print(fp, "mol rename $gfxmol {{LAMMPS region {}}}\n", region->id);
+          fputs("# set color and material\n", fp);
+          utils::print(fp, "graphics $gfxmol color {}\n", color);
+          utils::print(fp, "graphics $gfxmol material {}\n", material);
+          write_region(fp, region);
+        }
+      }
+
+    } else {
+      error->all(FLERR, iarg - 1, "Unknown region2vmd keyword {}", thisarg);
+    }
+    ++iarg;
+  }
+
+  // done. file will be close automatically
+  if (fp) {
+    // reset views and restore previous top molecule
+    fputs("after idle {if {$oldtop >= 0} {mol top $oldtop}; display resetview}\n", fp);
+  }
+}
+
+/* ----------------------------------------------------------------------
+  write out one region using VMD graphics primitives
+  fp is non-NULL only on MPI rank 0
+   ---------------------------------------------------------------------- */
+
+void Region2VMD::write_region(FILE *fp, Region *region)
+{
+  if (!fp || !region) return;
+
+  if (region->rotateflag) {
+    utils::logmesg(lmp, "Cannot (yet) handle rotating region {}. Skipping... ", region->id);
+    return;
+  }
+
+  // update internal variables
+  region->prematch();
+
+  // compute position offset for moving regions
+
+  double dx = 0.0;
+  double dy = 0.0;
+  double dz = 0.0;
+  if (region->moveflag) {
+    dx = region->dx;
+    dy = region->dy;
+    dz = region->dz;
+  }
+
+  // translate compatible regions to VMD graphics primitives, skip others.
+
+  const std::string regstyle = region->style;
+  if (regstyle == "block") {
+    const auto block = dynamic_cast<RegBlock *>(region);
+    if (!block) {
+      error->one(FLERR, Error::NOLASTLINE, "Region {} is not of style 'block'", region->id);
+    } else {
+      // a block is represented by 12 triangles
+      // comment out VMD command when side is open
+      utils::print(fp,
+                   "{6}draw triangle {{{0} {2} {4}}} {{{0} {2} {5}}} {{{0} {3} {4}}}\n"
+                   "{6}draw triangle {{{0} {3} {4}}} {{{0} {3} {5}}} {{{0} {2} {5}}}\n"
+                   "{8}draw triangle {{{0} {2} {4}}} {{{0} {2} {5}}} {{{1} {2} {4}}}\n"
+                   "{8}draw triangle {{{1} {2} {4}}} {{{1} {2} {5}}} {{{0} {2} {5}}}\n"
+                   "{7}draw triangle {{{1} {2} {4}}} {{{1} {2} {5}}} {{{1} {3} {4}}}\n"
+                   "{7}draw triangle {{{1} {3} {4}}} {{{1} {3} {5}}} {{{1} {2} {5}}}\n"
+                   "{10}draw triangle {{{0} {2} {4}}} {{{1} {2} {4}}} {{{1} {3} {4}}}\n"
+                   "{10}draw triangle {{{0} {2} {4}}} {{{0} {3} {4}}} {{{1} {3} {4}}}\n"
+                   "{11}draw triangle {{{0} {2} {5}}} {{{1} {2} {5}}} {{{1} {3} {5}}}\n"
+                   "{11}draw triangle {{{0} {2} {5}}} {{{0} {3} {5}}} {{{1} {3} {5}}}\n"
+                   "{9}draw triangle {{{0} {3} {4}}} {{{0} {3} {5}}} {{{1} {3} {5}}}\n"
+                   "{9}draw triangle {{{0} {3} {4}}} {{{1} {3} {4}}} {{{1} {3} {5}}}\n",
+                   block->xlo + dx, block->xhi + dx, block->ylo + dy, block->yhi + dy,
+                   block->zlo + dz, block->zhi + dz, block->open_faces[0] ? "# " : "",
+                   block->open_faces[1] ? "# " : "", block->open_faces[2] ? "# " : "",
+                   block->open_faces[3] ? "# " : "", block->open_faces[4] ? "# " : "",
+                   block->open_faces[5] ? "# " : "");
+    }
+
+  } else if (regstyle == "cone") {
+    const auto cone = dynamic_cast<RegCone *>(region);
+    if (!cone) {
+      error->one(FLERR, Error::NOLASTLINE, "Region {} is not of style 'cone'", region->id);
+    } else {
+      // draw cone
+      if (!cone->open_faces[2]) {
+
+        // both radii too small
+        if ((cone->radiuslo < SMALL) && (cone->radiushi < SMALL)) {
+          ;    // nothing to draw
+
+          // lo end has a tip
+        } else if (cone->radiuslo < SMALL) {
+          double v1[3], v2[3], v3[3], v4[3], v5[3];
+
+          if (cone->axis == 'x') {
+            // set tip coordinate
+            v1[0] = cone->lo + dx;
+            v1[1] = cone->c1 + dy;
+            v1[2] = cone->c2 + dz;
+
+            // loop around radius
+            for (int i = 0; i < RESOLUTION; ++i) {
+              v2[0] = cone->hi + dx;
+              v2[1] = cone->c1 + cone->radiushi * sin(RADINC * i) + dy;
+              v2[2] = cone->c2 + cone->radiushi * cos(RADINC * i) + dz;
+              v3[0] = cone->hi + dx;
+              v3[1] = cone->c1 + cone->radiushi * sin(RADINC * (i + 1.0)) + dy;
+              v3[2] = cone->c2 + cone->radiushi * cos(RADINC * (i + 1.0)) + dz;
+              v4[0] = 0.0;
+              v4[1] = sin(RADINC * i);
+              v4[2] = cos(RADINC * i);
+              v5[0] = 0.0;
+              v5[1] = sin(RADINC * (i + 1.0));
+              v5[2] = cos(RADINC * (i + 1.0));
+
+              utils::print(fp,
+                           "draw trinorm {{{} {} {}}} {{{} {} {}}} {{{} {} {}}} "
+                           "{{{} {} {}}} {{{} {} {}}} {{{} {} {}}}\n",
+                           v1[0], v1[1], v1[2], v2[0], v2[1], v2[2], v3[0], v3[1], v3[2], 1.0, 0.0,
+                           0.0, v4[0], v4[1], v4[2], v5[0], v5[1], v5[2]);
+            }
+
+          } else if (cone->axis == 'y') {
+            // set tip coordinate
+            v1[0] = cone->c1 + dx;
+            v1[1] = cone->lo + dy;
+            v1[2] = cone->c2 + dz;
+
+            // loop around radius
+            for (int i = 0; i < RESOLUTION; ++i) {
+              v2[0] = cone->c1 + cone->radiushi * sin(RADINC * i) + dx;
+              v2[1] = cone->hi + dy;
+              v2[2] = cone->c2 + cone->radiushi * cos(RADINC * i) + dz;
+              v3[0] = cone->c1 + cone->radiushi * sin(RADINC * (i + 1.0)) + dx;
+              v3[1] = cone->hi + dy;
+              v3[2] = cone->c2 + cone->radiushi * cos(RADINC * (i + 1.0)) + dz;
+              v4[0] = sin(RADINC * i);
+              v4[1] = 0.0;
+              v4[2] = cos(RADINC * i);
+              v5[0] = sin(RADINC * (i + 1.0));
+              v5[1] = 0.0;
+              v5[2] = cos(RADINC * (i + 1.0));
+
+              utils::print(fp,
+                           "draw trinorm {{{} {} {}}} {{{} {} {}}} {{{} {} {}}} "
+                           "{{{} {} {}}} {{{} {} {}}} {{{} {} {}}}\n",
+                           v1[0], v1[1], v1[2], v2[0], v2[1], v2[2], v3[0], v3[1], v3[2], 0.0, 1.0,
+                           0.0, v4[0], v4[1], v4[2], v5[0], v5[1], v5[2]);
+            }
+
+          } else if (cone->axis == 'z') {
+            // set tip coordinate
+            v1[0] = cone->c1 + dx;
+            v1[1] = cone->c2 + dy;
+            v1[2] = cone->lo + dz;
+
+            // loop around radius
+            for (int i = 0; i < RESOLUTION; ++i) {
+              v2[0] = cone->c1 + cone->radiushi * sin(RADINC * i) + dx;
+              v2[1] = cone->c2 + cone->radiushi * cos(RADINC * i) + dy;
+              v2[2] = cone->hi + dz;
+              v3[0] = cone->c1 + cone->radiushi * sin(RADINC * (i + 1.0)) + dx;
+              v3[1] = cone->c2 + cone->radiushi * cos(RADINC * (i + 1.0)) + dy;
+              v3[2] = cone->hi + dz;
+              v4[0] = sin(RADINC * i);
+              v4[1] = cos(RADINC * i);
+              v4[2] = 0.0;
+              v5[0] = sin(RADINC * (i + 1.0));
+              v5[1] = cos(RADINC * (i + 1.0));
+              v5[2] = 0.0;
+              utils::print(fp,
+                           "draw trinorm {{{} {} {}}} {{{} {} {}}} {{{} {} {}}} "
+                           "{{{} {} {}}} {{{} {} {}}} {{{} {} {}}}\n",
+                           v1[0], v1[1], v1[2], v2[0], v2[1], v2[2], v3[0], v3[1], v3[2], 0.0, 0.0,
+                           1.0, v4[0], v4[1], v4[2], v5[0], v5[1], v5[2]);
+            }
+          }
+
+          // hi end has a tip
+        } else if (cone->radiushi < SMALL) {
+          double v1[3], v2[3], v3[3], v4[3], v5[3];
+
+          if (cone->axis == 'x') {
+            // set tip coordinate
+            v1[0] = cone->hi + dx;
+            v1[1] = cone->c1 + dy;
+            v1[2] = cone->c2 + dz;
+
+            // loop around radius
+            for (int i = 0; i < RESOLUTION; ++i) {
+              v2[0] = cone->lo + dx;
+              v2[1] = cone->c1 + cone->radiuslo * sin(RADINC * i) + dy;
+              v2[2] = cone->c2 + cone->radiuslo * cos(RADINC * i) + dz;
+              v3[0] = cone->lo + dx;
+              v3[1] = cone->c1 + cone->radiuslo * sin(RADINC * (i + 1.0)) + dy;
+              v3[2] = cone->c2 + cone->radiuslo * cos(RADINC * (i + 1.0)) + dz;
+              v4[0] = 0.0;
+              v4[1] = sin(RADINC * i);
+              v4[2] = cos(RADINC * i);
+              v5[0] = 0.0;
+              v5[1] = sin(RADINC * (i + 1.0));
+              v5[2] = cos(RADINC * (i + 1.0));
+
+              utils::print(fp,
+                           "draw trinorm {{{} {} {}}} {{{} {} {}}} {{{} {} {}}} "
+                           "{{{} {} {}}} {{{} {} {}}} {{{} {} {}}}\n",
+                           v1[0], v1[1], v1[2], v2[0], v2[1], v2[2], v3[0], v3[1], v3[2], 1.0, 0.0,
+                           0.0, v4[0], v4[1], v4[2], v5[0], v5[1], v5[2]);
+            }
+
+          } else if (cone->axis == 'y') {
+            // set tip coordinate
+            v1[0] = cone->c1 + dx;
+            v1[1] = cone->hi + dy;
+            v1[2] = cone->c2 + dz;
+
+            // loop around radius
+            for (int i = 0; i < RESOLUTION; ++i) {
+              v2[0] = cone->c1 + cone->radiuslo * sin(RADINC * i) + dx;
+              v2[1] = cone->lo + dy;
+              v2[2] = cone->c2 + cone->radiuslo * cos(RADINC * i) + dz;
+              v3[0] = cone->c1 + cone->radiuslo * sin(RADINC * (i + 1.0)) + dx;
+              v3[1] = cone->lo + dy;
+              v3[2] = cone->c2 + cone->radiuslo * cos(RADINC * (i + 1.0)) + dz;
+              v4[0] = sin(RADINC * i);
+              v4[1] = 0.0;
+              v4[2] = cos(RADINC * i);
+              v5[0] = sin(RADINC * (i + 1.0));
+              v5[1] = 0.0;
+              v5[2] = cos(RADINC * (i + 1.0));
+
+              utils::print(fp,
+                           "draw trinorm {{{} {} {}}} {{{} {} {}}} {{{} {} {}}} "
+                           "{{{} {} {}}} {{{} {} {}}} {{{} {} {}}}\n",
+                           v1[0], v1[1], v1[2], v2[0], v2[1], v2[2], v3[0], v3[1], v3[2], 0.0, 1.0,
+                           0.0, v4[0], v4[1], v4[2], v5[0], v5[1], v5[2]);
+            }
+
+          } else if (cone->axis == 'z') {
+            // set tip coordinate
+            v1[0] = cone->c1 + dx;
+            v1[1] = cone->c2 + dy;
+            v1[2] = cone->hi + dz;
+
+            // loop around radius
+            for (int i = 0; i < RESOLUTION; ++i) {
+              v2[0] = cone->c1 + cone->radiuslo * sin(RADINC * i) + dx;
+              v2[1] = cone->c2 + cone->radiuslo * cos(RADINC * i) + dy;
+              v2[2] = cone->lo + dz;
+              v3[0] = cone->c1 + cone->radiuslo * sin(RADINC * (i + 1.0)) + dx;
+              v3[1] = cone->c2 + cone->radiuslo * cos(RADINC * (i + 1.0)) + dy;
+              v3[2] = cone->lo + dz;
+              v4[0] = sin(RADINC * i);
+              v4[1] = cos(RADINC * i);
+              v4[2] = 0.0;
+              v5[0] = sin(RADINC * (i + 1.0));
+              v5[1] = cos(RADINC * (i + 1.0));
+              v5[2] = 0.0;
+              utils::print(fp,
+                           "draw trinorm {{{} {} {}}} {{{} {} {}}} {{{} {} {}}} "
+                           "{{{} {} {}}} {{{} {} {}}} {{{} {} {}}}\n",
+                           v1[0], v1[1], v1[2], v2[0], v2[1], v2[2], v3[0], v3[1], v3[2], 0.0, 0.0,
+                           1.0, v4[0], v4[1], v4[2], v5[0], v5[1], v5[2]);
+            }
+          }
+
+          // truncated cone
+        } else if (!cone->open_faces[2]) {
+          double v1[3], v2[3], v3[3], v4[3], v5[3], v6[3], v7[3], v8[3];
+
+          if (cone->axis == 'x') {
+
+            // loop around radii
+            for (int i = 0; i < RESOLUTION; ++i) {
+              v1[0] = cone->hi + dx;
+              v1[1] = cone->c1 + cone->radiushi * sin(RADINC * (i - 0.5)) + dy;
+              v1[2] = cone->c2 + cone->radiushi * cos(RADINC * (i - 0.5)) + dz;
+              v2[0] = cone->lo + dx;
+              v2[1] = cone->c1 + cone->radiuslo * sin(RADINC * i) + dy;
+              v2[2] = cone->c2 + cone->radiuslo * cos(RADINC * i) + dz;
+              v3[0] = cone->hi + dx;
+              v3[1] = cone->c1 + cone->radiushi * sin(RADINC * (i + 0.5)) + dy;
+              v3[2] = cone->c2 + cone->radiushi * cos(RADINC * (i + 0.5)) + dz;
+              v4[0] = cone->lo + dx;
+              v4[1] = cone->c1 + cone->radiuslo * sin(RADINC * (i + 1.0)) + dy;
+              v4[2] = cone->c2 + cone->radiuslo * cos(RADINC * (i + 1.0)) + dz;
+              v5[0] = 0.0;
+              v5[1] = sin(RADINC * (i - 0.5));
+              v5[2] = cos(RADINC * (i - 0.5));
+              v6[0] = 0.0;
+              v6[1] = sin(RADINC * i);
+              v6[2] = cos(RADINC * i);
+              v7[0] = 0.0;
+              v7[1] = sin(RADINC * (i + 0.5));
+              v7[2] = cos(RADINC * (i + 0.5));
+              v8[0] = 0.0;
+              v8[1] = sin(RADINC * (i + 1.0));
+              v8[2] = cos(RADINC * (i + 1.0));
+
+              utils::print(fp,
+                           "draw trinorm {{{} {} {}}} {{{} {} {}}} {{{} {} {}}} "
+                           "{{{} {} {}}} {{{} {} {}}} {{{} {} {}}}\n",
+                           v1[0], v1[1], v1[2], v2[0], v2[1], v2[2], v3[0], v3[1], v3[2], v5[0],
+                           v5[1], v5[2], v6[0], v6[1], v6[2], v7[0], v7[1], v7[2]);
+              utils::print(fp,
+                           "draw trinorm {{{} {} {}}} {{{} {} {}}} {{{} {} {}}} "
+                           "{{{} {} {}}} {{{} {} {}}} {{{} {} {}}}\n",
+                           v2[0], v2[1], v2[2], v4[0], v4[1], v4[2], v3[0], v3[1], v3[2], v6[0],
+                           v6[1], v6[2], v8[0], v8[1], v8[2], v7[0], v7[1], v7[2]);
+            }
+
+          } else if (cone->axis == 'y') {
+
+            // loop around radii
+            for (int i = 0; i < RESOLUTION; ++i) {
+              v1[0] = cone->c1 + cone->radiushi * sin(RADINC * (i - 0.5)) + dx;
+              v1[1] = cone->hi + dy;
+              v1[2] = cone->c2 + cone->radiushi * cos(RADINC * (i - 0.5)) + dz;
+              v2[0] = cone->c1 + cone->radiuslo * sin(RADINC * i) + dx;
+              v2[1] = cone->lo + dy;
+              v2[2] = cone->c2 + cone->radiuslo * cos(RADINC * i) + dz;
+              v3[0] = cone->c1 + cone->radiushi * sin(RADINC * (i + 0.5)) + dx;
+              v3[1] = cone->hi + dy;
+              v3[2] = cone->c2 + cone->radiushi * cos(RADINC * (i + 0.5)) + dz;
+              v4[0] = cone->c1 + cone->radiuslo * sin(RADINC * (i + 1.0)) + dx;
+              v4[1] = cone->lo + dy;
+              v4[2] = cone->c2 + cone->radiuslo * cos(RADINC * (i + 1.0)) + dz;
+              v5[0] = sin(RADINC * (i - 0.5));
+              v5[1] = 0.0;
+              v5[2] = cos(RADINC * (i - 0.5));
+              v6[0] = sin(RADINC * i);
+              v6[1] = 0.0;
+              v6[2] = cos(RADINC * i);
+              v7[0] = sin(RADINC * (i + 0.5));
+              v7[1] = 0.0;
+              v7[2] = cos(RADINC * (i + 0.5));
+              v8[0] = sin(RADINC * (i + 1.0));
+              v8[1] = 0.0;
+              v8[2] = cos(RADINC * (i + 1.0));
+
+              utils::print(fp,
+                           "draw trinorm {{{} {} {}}} {{{} {} {}}} {{{} {} {}}} "
+                           "{{{} {} {}}} {{{} {} {}}} {{{} {} {}}}\n",
+                           v1[0], v1[1], v1[2], v2[0], v2[1], v2[2], v3[0], v3[1], v3[2], v5[0],
+                           v5[1], v5[2], v6[0], v6[1], v6[2], v7[0], v7[1], v7[2]);
+              utils::print(fp,
+                           "draw trinorm {{{} {} {}}} {{{} {} {}}} {{{} {} {}}} "
+                           "{{{} {} {}}} {{{} {} {}}} {{{} {} {}}}\n",
+                           v2[0], v2[1], v2[2], v4[0], v4[1], v4[2], v3[0], v3[1], v3[2], v6[0],
+                           v6[1], v6[2], v8[0], v8[1], v8[2], v7[0], v7[1], v7[2]);
+            }
+
+          } else if (cone->axis == 'z') {
+
+            // loop around radii
+            for (int i = 0; i < RESOLUTION; ++i) {
+              v1[0] = cone->c1 + cone->radiushi * sin(RADINC * (i - 0.5)) + dx;
+              v1[1] = cone->c2 + cone->radiushi * cos(RADINC * (i - 0.5)) + dy;
+              v1[2] = cone->hi + dz;
+              v2[0] = cone->c1 + cone->radiuslo * sin(RADINC * i) + dx;
+              v2[1] = cone->c2 + cone->radiuslo * cos(RADINC * i) + dy;
+              v2[2] = cone->lo + dz;
+              v3[0] = cone->c1 + cone->radiushi * sin(RADINC * (i + 0.5)) + dx;
+              v3[1] = cone->c2 + cone->radiushi * cos(RADINC * (i + 0.5)) + dy;
+              v3[2] = cone->hi + dz;
+              v4[0] = cone->c1 + cone->radiuslo * sin(RADINC * (i + 1.0)) + dx;
+              v4[1] = cone->c2 + cone->radiuslo * cos(RADINC * (i + 1.0)) + dy;
+              v4[2] = cone->lo + dz;
+              v5[0] = sin(RADINC * (i - 0.5));
+              v5[1] = cos(RADINC * (i - 0.5));
+              v5[2] = 0.0;
+              v6[0] = sin(RADINC * i);
+              v6[1] = cos(RADINC * i);
+              v6[2] = 0.0;
+              v7[0] = sin(RADINC * (i + 0.5));
+              v7[1] = cos(RADINC * (i + 0.5));
+              v7[2] = 0.0;
+              v8[0] = sin(RADINC * (i + 1.0));
+              v8[1] = cos(RADINC * (i + 1.0));
+              v8[2] = 0.0;
+
+              utils::print(fp,
+                           "draw trinorm {{{} {} {}}} {{{} {} {}}} {{{} {} {}}} "
+                           "{{{} {} {}}} {{{} {} {}}} {{{} {} {}}}\n",
+                           v1[0], v1[1], v1[2], v2[0], v2[1], v2[2], v3[0], v3[1], v3[2], v5[0],
+                           v5[1], v5[2], v6[0], v6[1], v6[2], v7[0], v7[1], v7[2]);
+              utils::print(fp,
+                           "draw trinorm {{{} {} {}}} {{{} {} {}}} {{{} {} {}}} "
+                           "{{{} {} {}}} {{{} {} {}}} {{{} {} {}}}\n",
+                           v2[0], v2[1], v2[2], v4[0], v4[1], v4[2], v3[0], v3[1], v3[2], v6[0],
+                           v6[1], v6[2], v8[0], v8[1], v8[2], v7[0], v7[1], v7[2]);
+            }
+          }
+        }
+      }
+
+      // draw lids
+      if ((cone->radiuslo > SMALL) && !cone->open_faces[0]) {
+        double lid = cone->lo;
+        if (cone->axis == 'x') {
+          lid += dx;
+          utils::print(fp,
+                       "draw cylinder {{{0} {2} {3}}} {{{1:.15} {2} {3}}} radius {4} "
+                       "resolution {5} filled yes\n",
+                       lid, lid + DELTA, cone->c1 + dy, cone->c2 + dz, cone->radiuslo, RESOLUTION);
+        } else if (cone->axis == 'y') {
+          lid += dy;
+          utils::print(fp,
+                       "draw cylinder {{{2} {0} {3}}} {{{2} {1:.15} {3}}} radius {4} "
+                       "resolution {5} filled yes\n",
+                       lid, lid + DELTA, cone->c1 + dx, cone->c2 + dz, cone->radiuslo, RESOLUTION);
+        } else if (cone->axis == 'z') {
+          lid += dz;
+          utils::print(fp,
+                       "draw cylinder {{{2} {3} {0}}} {{{2} {3} {1:.15}}} radius {4} "
+                       "resolution {5} filled yes\n",
+                       lid, lid + DELTA, cone->c1 + dx, cone->c2 + dy, cone->radiuslo, RESOLUTION);
+        }
+      }
+      if ((cone->radiushi > SMALL) && !cone->open_faces[1]) {
+        double lid = cone->hi;
+        if (cone->axis == 'x') {
+          lid += dx;
+          utils::print(fp,
+                       "draw cylinder {{{0} {2} {3}}} {{{1:.15} {2} {3}}} radius {4} "
+                       "resolution {5} filled yes\n",
+                       lid, lid + DELTA, cone->c1 + dy, cone->c2 + dz, cone->radiushi, RESOLUTION);
+        } else if (cone->axis == 'y') {
+          lid += dy;
+          utils::print(fp,
+                       "draw cylinder {{{2} {0} {3}}} {{{2} {1:.15} {3}}} radius {4} "
+                       "resolution {5} filled yes\n",
+                       lid, lid + DELTA, cone->c1 + dx, cone->c2 + dz, cone->radiushi, RESOLUTION);
+        } else if (cone->axis == 'z') {
+          lid += dz;
+          utils::print(fp,
+                       "draw cylinder {{{2} {3} {0}}} {{{2} {3} {1:.15}}} radius {4} "
+                       "resolution {5} filled yes\n",
+                       lid, lid + DELTA, cone->c1 + dx, cone->c2 + dy, cone->radiushi, RESOLUTION);
+        }
+      }
+    }
+
+  } else if (regstyle == "cylinder") {
+    const auto cyl = dynamic_cast<RegCylinder *>(region);
+    if (!cyl) {
+      error->one(FLERR, Error::NOLASTLINE, "Region {} is not of style 'cylinder'", region->id);
+    } else {
+      // first draw the cylinder. filled only when *all* faces are closed.
+      // with any open face we draw each part separately
+      std::string filled = "filled no";
+      if (!cyl->open_faces[0] && !cyl->open_faces[1] && !cyl->open_faces[2]) {
+        filled = "filled yes";
+      }
+
+      // the cylinder uses a single cylinder primitive
+      if (!cyl->open_faces[2]) {
+        if (cyl->axis == 'x') {
+          utils::print(
+              fp, "draw cylinder {{{0} {2} {3}}} {{{1} {2} {3}}} radius {4} resolution {5} {6}\n",
+              cyl->lo + dx, cyl->hi + dx, cyl->c1 + dy, cyl->c2 + dz, cyl->radius, RESOLUTION,
+              filled);
+        } else if (cyl->axis == 'y') {
+          utils::print(
+              fp, "draw cylinder {{{2} {0} {3}}} {{{2} {1} {3}}} radius {4} resolution {5} {6}\n",
+              cyl->lo + dy, cyl->hi + dy, cyl->c1 + dx, cyl->c2 + dz, cyl->radius, RESOLUTION,
+              filled);
+        } else if (cyl->axis == 'z') {
+          utils::print(
+              fp, "draw cylinder {{{2} {3} {0}}} {{{2} {3} {1}}} radius {4} resolution {5} {6}\n",
+              cyl->lo + dz, cyl->hi + dz, cyl->c1 + dx, cyl->c2 + dy, cyl->radius, RESOLUTION,
+              filled);
+        }
+      }
+
+      // draw lids
+      if ((filled == "filled no") && !cyl->open_faces[0]) {
+        double lid = cyl->lo;
+        if (cyl->axis == 'x') {
+          lid += dx;
+          utils::print(fp,
+                       "draw cylinder {{{0} {2} {3}}} {{{1:.15} {2} {3}}} radius {4} "
+                       "resolution {5} filled yes\n",
+                       lid, lid + DELTA, cyl->c1 + dy, cyl->c2 + dz, cyl->radius, RESOLUTION);
+        } else if (cyl->axis == 'y') {
+          lid += dy;
+          utils::print(fp,
+                       "draw cylinder {{{2} {0} {3}}} {{{2} {1:.15} {3}}} radius {4} "
+                       "resolution {5} filled yes\n",
+                       lid, lid + DELTA, cyl->c1 + dx, cyl->c2 + dz, cyl->radius, RESOLUTION);
+        } else if (cyl->axis == 'z') {
+          lid += dz;
+          utils::print(fp,
+                       "draw cylinder {{{2} {3} {0}}} {{{2} {3} {1:.15}}} radius {4} "
+                       "resolution {5} filled yes\n",
+                       lid, lid + DELTA, cyl->c1 + dx, cyl->c2 + dy, cyl->radius, RESOLUTION);
+        }
+      }
+      if ((filled == "filled no") && !cyl->open_faces[1]) {
+        double lid = cyl->hi;
+        if (cyl->axis == 'x') {
+          lid += dx;
+          utils::print(fp,
+                       "draw cylinder {{{0} {2} {3}}} {{{1:.15} {2} {3}}} radius {4} "
+                       "resolution {5} filled yes\n",
+                       lid, lid + DELTA, cyl->c1 + dy, cyl->c2 + dz, cyl->radius, RESOLUTION);
+        } else if (cyl->axis == 'y') {
+          lid += dy;
+          utils::print(fp,
+                       "draw cylinder {{{2} {0} {3}}} {{{2} {1:.15} {3}}} radius {4} "
+                       "resolution {5} filled yes\n",
+                       lid, lid + DELTA, cyl->c1 + dx, cyl->c2 + dz, cyl->radius, RESOLUTION);
+        } else if (cyl->axis == 'z') {
+          lid += dz;
+          utils::print(fp,
+                       "draw cylinder {{{2} {3} {0}}} {{{2} {3} {1:.15}}} radius {4} "
+                       "resolution {5} filled yes\n",
+                       lid, lid + DELTA, cyl->c1 + dx, cyl->c2 + dy, cyl->radius, RESOLUTION);
+        }
+      }
+    }
+
+  } else if (regstyle == "ellipsoid") {
+    const auto ellipsoid = dynamic_cast<RegEllipsoid *>(region);
+    if (!ellipsoid) {
+      error->one(FLERR, Error::NOLASTLINE, "Region {} is not of style 'ellipsoid'", region->id);
+    } else {
+      // for ellipsoid we use a custom VMD function that emulates it using trinorm primitives
+      utils::print(fp, "draw ellipsoid 3 {{{} {} {}}} {{{} {} {}}}\n", ellipsoid->xc + dx,
+                   ellipsoid->yc + dy, ellipsoid->zc + dz, ellipsoid->a, ellipsoid->b,
+                   ellipsoid->c);
+    }
+
+  } else if (regstyle == "plane") {
+    const auto plane = dynamic_cast<RegPlane *>(region);
+    if (!plane) {
+      error->one(FLERR, Error::NOLASTLINE, "Region {} is not of style 'plane'", region->id);
+    } else {
+      double v1[3], v2[3], v3[3], v4[3], v5[3], v6[3];
+      MathExtra::copy3(plane->normal, v1);
+      MathExtra::norm3(v1);
+      v2[0] = plane->xp + dx;
+      v2[1] = plane->yp + dy;
+      v2[2] = plane->zp + dz;
+      // determine the largest component of the plane normal and use that to determine the triangle edges
+      int dim1, dim2, dim3;
+      if (fabs(v1[0]) > fabs(v1[1])) {      // x > y
+        if (fabs(v1[0]) > fabs(v1[2])) {    // x > z => x is largest
+          dim3 = 0;
+          dim1 = 1;
+          dim2 = 2;
+        } else {    // z is largest
+          dim3 = 2;
+          dim1 = 0;
+          dim2 = 1;
+        }
+      } else {                              // x < y
+        if (fabs(v1[1]) > fabs(v1[2])) {    // y > z => y is largest
+          dim3 = 1;
+          dim1 = 0;
+          dim2 = 2;
+        } else {    // z is largest
+          dim3 = 2;
+          dim1 = 0;
+          dim2 = 1;
+        }
+      }
+
+      MathExtra::scale3(2.0, domain->boxlo, v3);
+      v3[dim3] = 0.0;
+      MathExtra::cross3(v1, v3, v4);
+      MathExtra::add3(v4, v2, v5);
+      v3[dim1] = domain->boxhi[dim1];
+      v3[dim2] = domain->boxlo[dim2];
+      v3[dim3] = 0.0;
+      MathExtra::scale3(2.0, v3);
+      MathExtra::cross3(v1, v3, v4);
+      MathExtra::add3(v4, v2, v6);
+      utils::print(fp, "draw triangle {{{} {} {}}} {{{} {} {}}} {{{} {} {}}}\n", v2[dim1], v2[dim2],
+                   v2[dim3], v5[dim1], v5[dim2], v5[dim3], v6[dim1], v6[dim2], v6[dim3]);
+      v3[dim1] = domain->boxlo[dim1];
+      v3[dim2] = domain->boxhi[dim2];
+      v3[dim3] = 0.0;
+      MathExtra::scale3(2.0, v3);
+      MathExtra::cross3(v1, v3, v4);
+      MathExtra::add3(v4, v2, v6);
+      utils::print(fp, "draw triangle {{{} {} {}}} {{{} {} {}}} {{{} {} {}}}\n", v2[dim1], v2[dim2],
+                   v2[dim3], v5[dim1], v5[dim2], v5[dim3], v6[dim1], v6[dim2], v6[dim3]);
+      v3[dim1] = domain->boxhi[dim1];
+      v3[dim2] = domain->boxhi[dim2];
+      v3[dim3] = 0.0;
+      MathExtra::scale3(2.0, v3);
+      MathExtra::cross3(v1, v3, v4);
+      MathExtra::add3(v4, v2, v5);
+      v3[dim1] = domain->boxhi[dim1];
+      v3[dim2] = domain->boxlo[dim2];
+      v3[dim3] = 0.0;
+      MathExtra::scale3(2.0, v3);
+      MathExtra::cross3(v1, v3, v4);
+      MathExtra::add3(v4, v2, v6);
+      utils::print(fp, "draw triangle {{{} {} {}}} {{{} {} {}}} {{{} {} {}}}\n", v2[dim1], v2[dim2],
+                   v2[dim3], v5[dim1], v5[dim2], v5[dim3], v6[dim1], v6[dim2], v6[dim3]);
+      v3[dim1] = domain->boxlo[dim1];
+      v3[dim2] = domain->boxhi[dim2];
+      v3[dim3] = 0.0;
+      MathExtra::scale3(2.0, v3);
+      MathExtra::cross3(v1, v3, v4);
+      MathExtra::add3(v4, v2, v6);
+      utils::print(fp, "draw triangle {{{} {} {}}} {{{} {} {}}} {{{} {} {}}}\n", v2[dim1], v2[dim2],
+                   v2[dim3], v5[dim1], v5[dim2], v5[dim3], v6[dim1], v6[dim2], v6[dim3]);
+    }
+
+  } else if (regstyle == "prism") {
+    const auto prism = dynamic_cast<RegPrism *>(region);
+    if (!prism) {
+      error->one(FLERR, Error::NOLASTLINE, "Region {} is not of style 'prism'", region->id);
+    } else {
+      // a prism is represented by 12 triangles
+      // comment out VMD command when side is open
+      utils::print(fp,
+                   "{14}draw triangle {{{0} {2} {4}}} {{{1} {2} {4}}} {{{7} {3} {4}}}\n"
+                   "{14}draw triangle {{{0} {2} {4}}} {{{6} {3} {4}}} {{{7} {3} {4}}}\n"
+                   "{16}draw triangle {{{0} {2} {4}}} {{{1} {2} {4}}} {{{8} {9} {5}}}\n"
+                   "{16}draw triangle {{{0} {2} {4}}} {{{10} {9} {5}}} {{{8} {9} {5}}}\n"
+                   "{19}draw triangle {{{1} {2} {4}}} {{{8} {9} {5}}} {{{7} {3} {4}}}\n"
+                   "{19}draw triangle {{{11} {12} {5}}} {{{8} {9} {5}}} {{{7} {3} {4}}}\n"
+                   "{18}draw triangle {{{0} {2} {4}}} {{{6} {3} {4}}} {{{13} {12} {5}}}\n"
+                   "{18}draw triangle {{{0} {2} {4}}} {{{10} {9} {5}}} {{{13} {12} {5}}}\n"
+                   "{15}draw triangle {{{10} {9} {5}}} {{{8} {9} {5}}} {{{11} {12} {5}}}\n"
+                   "{15}draw triangle {{{10} {9} {5}}} {{{13} {12} {5}}} {{{11} {12} {5}}}\n"
+                   "{17}draw triangle {{{6} {3} {4}}} {{{7} {3} {4}}} {{{11} {12} {5}}}\n"
+                   "{17}draw triangle {{{6} {3} {4}}} {{{13} {12} {5}}} {{{11} {12} {5}}}\n",
+                   prism->xlo, prism->xhi, prism->ylo, prism->yhi, prism->zlo, prism->zhi,
+                   prism->xlo + prism->xy, prism->xhi + prism->xy, prism->xhi + prism->xz,
+                   prism->ylo + prism->yz, prism->xlo + prism->xz,
+                   prism->xhi + prism->xy + prism->xz, prism->yhi + prism->yz,
+                   prism->xlo + prism->xy + prism->xz, prism->open_faces[0] ? "# " : "",
+                   prism->open_faces[1] ? "# " : "", prism->open_faces[2] ? "# " : "",
+                   prism->open_faces[3] ? "# " : "", prism->open_faces[4] ? "# " : "",
+                   prism->open_faces[5] ? "# " : "");
+    }
+
+  } else if (regstyle == "sphere") {
+    const auto sphere = dynamic_cast<RegSphere *>(region);
+    if (!sphere) {
+      error->one(FLERR, Error::NOLASTLINE, "Region {} is not of style 'sphere'", region->id);
+    } else {
+      // a sphere uses a single sphere primitive
+      utils::print(fp, "draw sphere {{{} {} {}}} radius {} resolution {}\n", sphere->xc, sphere->yc,
+                   sphere->zc, sphere->radius, RESOLUTION);
+    }
+
+  } else {
+    utils::logmesg(lmp,
+                   "Cannot (yet) translate region {} of style {} to VMD graphics. Skipping... ",
+                   region->id, region->style);
+  }
+  utils::logmesg(lmp, " done\n");
+}

src/EXTRA-COMMAND/region2vmd.h

@@ -0,0 +1,41 @@
+/* -*- c++ -*-----------------------------------------------------------
+   LAMMPS - Large-scale Atomic/Molecular Massively Parallel Simulator
+   https://www.lammps.org/, Sandia National Laboratories
+   LAMMPS development team: developers@lammps.org
+
+   Copyright (2003) Sandia Corporation.  Under the terms of Contract
+   DE-AC04-94AL85000 with Sandia Corporation, the U.S. Government retains
+   certain rights in this software.  This software is distributed under
+   the GNU General Public License.
+
+   See the README file in the top-level LAMMPS directory.
+------------------------------------------------------------------------- */
+
+#ifdef COMMAND_CLASS
+// clang-format off
+CommandStyle(region2vmd,Region2VMD);
+// clang-format on
+#else
+
+#ifndef LMP_REGION2VMD_H
+#define LMP_REGION2VMD_H
+
+#include "command.h"
+
+namespace LAMMPS_NS {
+
+class Region;
+
+class Region2VMD : public Command {
+ public:
+  Region2VMD(class LAMMPS *lmp) : Command(lmp) {};
+  void command(int, char **) override;
+
+ protected:
+  void write_region(FILE *, Region *);
+};
+
+}    // namespace LAMMPS_NS
+
+#endif
+#endif

src/KOKKOS/comm_kokkos.cpp

@@ -793,7 +793,10 @@ void CommKokkos::exchange()
   }
 
   atomKK->sync(Host,ALL_MASK);
+  int prev_auto_sync = lmp->kokkos->auto_sync;
+  lmp->kokkos->auto_sync = 1;
   CommBrick::exchange();
+  lmp->kokkos->auto_sync = prev_auto_sync;
   atomKK->modified(Host,ALL_MASK);
 }
 

src/KOKKOS/mliap_model_python_kokkos.cpp

@@ -68,12 +68,12 @@ MLIAPModelPythonKokkos<DeviceType>::MLIAPModelPythonKokkos(LAMMPS *lmp, char *co
   // Recipe from lammps/src/pair_python.cpp :
   // add current directory to PYTHONPATH
   PyObject *py_path = PySys_GetObject((char *) "path");
-  PyList_Append(py_path, PY_STRING_FROM_STRING("."));
+  PyList_Append(py_path, PyUnicode_FromString("."));
 
   // if LAMMPS_POTENTIALS environment variable is set, add it to PYTHONPATH as well
   const char *potentials_path = getenv("LAMMPS_POTENTIALS");
   if (potentials_path != nullptr) {
-    PyList_Append(py_path, PY_STRING_FROM_STRING(potentials_path));
+    PyList_Append(py_path, PyUnicode_FromString(potentials_path));
   }
   PyGILState_Release(gstate);
   if (coefffilename) read_coeffs(coefffilename);

src/ML-IAP/mliap_model_python.cpp

@@ -62,12 +62,12 @@ MLIAPModelPython::MLIAPModelPython(LAMMPS *lmp, char *coefffilename, bool is_chi
   // Recipe from lammps/src/pair_python.cpp :
   // add current directory to PYTHONPATH
   PyObject *py_path = PySys_GetObject((char *) "path");
-  PyList_Append(py_path, PY_STRING_FROM_STRING("."));
+  PyList_Append(py_path, PyUnicode_FromString("."));
 
   // if LAMMPS_POTENTIALS environment variable is set, add it to PYTHONPATH as well
   const char *potentials_path = getenv("LAMMPS_POTENTIALS");
   if (potentials_path != nullptr) {
-    PyList_Append(py_path, PY_STRING_FROM_STRING(potentials_path));
+    PyList_Append(py_path, PyUnicode_FromString(potentials_path));
   }
   PyGILState_Release(gstate);
   if (coefffilename) read_coeffs(coefffilename);

src/ML-RANN/rann_fingerprint_bond.cpp

@@ -36,6 +36,8 @@ DISTRIBUTION A. Approved for public release; distribution unlimited. OPSEC#4918
 
 using namespace LAMMPS_NS::RANN;
 
+static constexpr double SMALL = 1.0e-12;
+
 Fingerprint_bond::Fingerprint_bond(PairRANN *_pair) : Fingerprint(_pair)
 {
   n_body_type = 3;
@@ -317,7 +319,8 @@ void Fingerprint_bond::do3bodyfeatureset_singleneighborloop(double * features,do
   i = ilist[ii];
   itype = pair->map[type[i]];
   int f = pair->net[itype].dimensions[0];
-  double expr[jnum][kmax+12];
+  std::vector<double> row(kmax+12, 0.0);
+  std::vector<std::vector<double>> expr(jnum, row);
   int p = kmax;
   int countmb=((mlength)*(mlength+1))>>1;
   // calculate interpolation expr, rinvs and dfc, for each neighbor
@@ -354,13 +357,13 @@ void Fingerprint_bond::do3bodyfeatureset_singleneighborloop(double * features,do
     expr[jj][p+1]=dely*rinvs;
     expr[jj][p+2]=delz*rinvs;
     //Hack to avoid nan when x y or z component of radial vector is exactly 0. Shouldn't affect accuracy.
-    if (expr[jj][p]*expr[jj][p]<0.000000000001) {
+    if (expr[jj][p]*expr[jj][p] < SMALL) {
       expr[jj][p] = 0.000001;
     }
-    if (expr[jj][p+1]*expr[jj][p+1]<0.000000000001) {
+    if (expr[jj][p+1]*expr[jj][p+1] < SMALL) {
       expr[jj][p+1] = 0.000001;
     }
-    if (expr[jj][p+2]*expr[jj][p+2]<0.000000000001) {
+    if (expr[jj][p+2]*expr[jj][p+2] < SMALL) {
       expr[jj][p+2] = 0.000001;
     }
     expr[jj][p+3] = -dfc*expr[jj][p];
@@ -377,7 +380,7 @@ void Fingerprint_bond::do3bodyfeatureset_singleneighborloop(double * features,do
   int kb = kmax;
   int mb = mlength;
   count = startingneuron;
-  double Bb[mb];
+  std::vector<double> Bb(mb, 0.0);
   double dBbx;
   double dBby;
   double dBbz;
@@ -397,7 +400,7 @@ void Fingerprint_bond::do3bodyfeatureset_singleneighborloop(double * features,do
     }
     for (n=0;n<kb;n++) {
       for (a1=0;a1<mb;a1++) {
-        Bb[a1]=0;
+        Bb[a1]=0.0;
       }
       ai = n;
       double y1 = alpha_k[ai]/re;
@@ -409,7 +412,7 @@ void Fingerprint_bond::do3bodyfeatureset_singleneighborloop(double * features,do
           continue;
         }
         double yprod = expr[jj][ai];
-        double *y4 = &expr[jj][p];
+        double *y4 = expr[jj].data() + p;
         for (a2=0;a2<a;a2++) {
           yprod *= y4[M[a2+1]];
         }
@@ -419,10 +422,7 @@ void Fingerprint_bond::do3bodyfeatureset_singleneighborloop(double * features,do
         }
       }
       if (atomtypes[1]!=atomtypes[2]) {//Bb!=Bg
-        double Bg[mb];
+        std::vector<double> Bg(mb, 0.0);
-        for (a1=0;a1<mb;a1++) {
-          Bg[a1]=0;
-        }
         ai = n;
         double y1 = alpha_k[ai]/re;
         //loop over ktype to get Bg
@@ -433,7 +433,7 @@ void Fingerprint_bond::do3bodyfeatureset_singleneighborloop(double * features,do
             continue;
           }
           double yprod = expr[jj][ai];
-          double *y4 = &expr[jj][p];
+          double *y4 = expr[jj].data() + p;
           for (a2=0;a2<a;a2++) {
             yprod *= y4[M[a2+1]];
           }
@@ -450,8 +450,8 @@ void Fingerprint_bond::do3bodyfeatureset_singleneighborloop(double * features,do
           if (atomtypes[2] != nelements && atomtypes[2] != jtype) {
             continue;
           }
-          double *y3 = &expr[jj][p+3];
+          double *y3 = expr[jj].data() + p + 3;
-          double *y4 = &expr[jj][p];
+          double *y4 = expr[jj].data() + p;
           ai = n;
           yprod = expr[jj][ai];
           for (a2=0;a2<a;a2++) {
@@ -477,8 +477,8 @@ void Fingerprint_bond::do3bodyfeatureset_singleneighborloop(double * features,do
           if (atomtypes[1] != nelements && atomtypes[1] != jtype) {
             continue;
           }
-          double *y3 = &expr[jj][p+3];
+          double *y3 = expr[jj].data() + p + 3;
-          double *y4 = &expr[jj][p];
+          double *y4 = expr[jj].data() + p;
           ai = n;
           yprod = expr[jj][ai];
           for (a2=0;a2<a;a2++) {
@@ -512,8 +512,8 @@ void Fingerprint_bond::do3bodyfeatureset_singleneighborloop(double * features,do
           if (atomtypes[1] != nelements && atomtypes[1] != jtype) {
             continue;
           }
-          double *y3 = &expr[jj][p+3];
+          double *y3 = expr[jj].data() + p + 3;
-          double *y4 = &expr[jj][p];
+          double *y4 = expr[jj].data() + p;
           ai = n;
           yprod = expr[jj][ai];
           for (a2=0;a2<a;a2++) {
@@ -572,16 +572,18 @@ void Fingerprint_bond::do3bodyfeatureset_doubleneighborloop(double * features,do
   i = ilist[ii];
   itype = pair->map[type[i]];
   int f = pair->net[itype].dimensions[0];
-  double expr[jnum][kmax];
+  std::vector<double> row(kmax, 0.0);
-  double y[jnum][3];
+  std::vector<std::vector<double>> expr(jnum, row);
-  double ri[jnum];
+  std::vector<double> yrow(3, 0.0);
-  double dfc[jnum];
+  std::vector<std::vector<double>> y(jnum, yrow);
+  std::vector<double> ri(jnum, 0.0);
+  std::vector<double> dfc(jnum, 0.0);
   int kb = kmax;
   int mb = mlength;
-  double c41[kmax];
+  std::vector<double> c41(kmax, 0.0);
-  double c51[kmax];
+  std::vector<double> c51(kmax, 0.0);
-  double c61[kmax];
+  std::vector<double> c61(kmax, 0.0);
-  double ct[kmax];
+  std::vector<double> ct(kmax, 0.0);
   for (jj = 0; jj < jnum; jj++) {
     jtype = tn[jj];
     if (jtypes != nelements && jtypes != jtype && ktypes != nelements && ktypes != jtype) {

src/ML-RANN/rann_fingerprint_bondscreened.cpp

@@ -36,6 +36,8 @@ DISTRIBUTION A. Approved for public release; distribution unlimited. OPSEC#4918
 
 using namespace LAMMPS_NS::RANN;
 
+static constexpr double SMALL = 1.0e-12;
+
 Fingerprint_bondscreened::Fingerprint_bondscreened(PairRANN *_pair) : Fingerprint(_pair)
 {
   n_body_type = 3;
@@ -319,7 +321,8 @@ void Fingerprint_bondscreened::do3bodyfeatureset_singleneighborloop(double * fea
   i = ilist[ii];
   itype = pair->map[type[i]];
   int f = pair->net[itype].dimensions[0];
-  double expr[jnum][kmax+12];
+  std::vector<double> row(kmax+12, 0.0);
+  std::vector<std::vector<double>> expr(jnum, row);
   int p = kmax;
   int countmb=((mlength)*(mlength+1))>>1;
   // calculate interpolation expr, rinvs and dfc, for each neighbor
@@ -358,13 +361,13 @@ void Fingerprint_bondscreened::do3bodyfeatureset_singleneighborloop(double * fea
     expr[jj][p+1]=dely*rinvs;
     expr[jj][p+2]=delz*rinvs;
     //Hack to avoid nan when x y or z component of radial vector is exactly 0. Shouldn't affect accuracy.
-    if (expr[jj][p]*expr[jj][p]<0.000000000001) {
+    if (expr[jj][p]*expr[jj][p] < SMALL) {
       expr[jj][p] = 0.000001;
     }
-    if (expr[jj][p+1]*expr[jj][p+1]<0.000000000001) {
+    if (expr[jj][p+1]*expr[jj][p+1] < SMALL) {
       expr[jj][p+1] = 0.000001;
     }
-    if (expr[jj][p+2]*expr[jj][p+2]<0.000000000001) {
+    if (expr[jj][p+2]*expr[jj][p+2] < SMALL) {
       expr[jj][p+2] = 0.000001;
     }
     expr[jj][p+3] = -dfc*expr[jj][p]-dSikx[jj];
@@ -381,20 +384,20 @@ void Fingerprint_bondscreened::do3bodyfeatureset_singleneighborloop(double * fea
   int kb = kmax;
   int mb = mlength;
   count = startingneuron;
-  double Bb[mb];
+  std::vector<double> Bb(mb, 0.0);
   double dBbx;
   double dBby;
   double dBbz;
-  double dBbx1[mb];
+  std::vector<double> dBbx1(mb, 0.0);
-  double dBby1[mb];
+  std::vector<double> dBby1(mb, 0.0);
-  double dBbz1[mb];
+  std::vector<double> dBbz1(mb, 0.0);
   double yprod;
   for (mcount=0;mcount<countmb;mcount++) {
     int *M = Mf[mcount];
     int *_coeffx = coeffx[mcount];
-    int *_coeffy =  coeffy[mcount];
+    int *_coeffy = coeffy[mcount];
-    int *_coeffz =  coeffz[mcount];
+    int *_coeffz = coeffz[mcount];
-    int *_coeff =  coeff[mcount];
+    int *_coeff = coeff[mcount];
     a = mb+1;
     for (a1=0;a1<mb;a1++) {
       if (Mf[mcount][a1+1]==0) {
@@ -420,7 +423,7 @@ void Fingerprint_bondscreened::do3bodyfeatureset_singleneighborloop(double * fea
           continue;
         }
         double yprod = expr[jj][ai];
-        double *y4 = &expr[jj][p];
+        double *y4 = expr[jj].data() + p;
         for (a2=0;a2<a;a2++) {
           yprod *= y4[M[a2+1]];
         }
@@ -430,10 +433,7 @@ void Fingerprint_bondscreened::do3bodyfeatureset_singleneighborloop(double * fea
         }
       }
       if (atomtypes[1]!=atomtypes[2]) {//Bb!=Bg
-        double Bg[mb];
+        std::vector<double> Bg(mb, 0.0);
-        for (a1=0;a1<mb;a1++) {
-          Bg[a1]=0;
-        }
         ai = n;
         double y1 = alpha_k[ai]/re;
         //loop over ktype to get Bg
@@ -445,7 +445,7 @@ void Fingerprint_bondscreened::do3bodyfeatureset_singleneighborloop(double * fea
             continue;
           }
           double yprod = expr[jj][ai];
-          double *y4 = &expr[jj][p];
+          double *y4 = expr[jj].data() + p;
           for (a2=0;a2<a;a2++) {
             yprod *= y4[M[a2+1]];
           }
@@ -463,8 +463,8 @@ void Fingerprint_bondscreened::do3bodyfeatureset_singleneighborloop(double * fea
           if (atomtypes[2] != nelements && atomtypes[2] != jtype) {
             continue;
           }
-          double *y3 = &expr[jj][p+3];
+          double *y3 = expr[jj].data() + p + 3;
-          double *y4 = &expr[jj][p];
+          double *y4 = expr[jj].data() + p;
           ai = n;
           yprod = expr[jj][ai];
           for (a2=0;a2<a;a2++) {
@@ -500,8 +500,8 @@ void Fingerprint_bondscreened::do3bodyfeatureset_singleneighborloop(double * fea
           if (atomtypes[1] != nelements && atomtypes[1] != jtype) {
             continue;
           }
-          double *y3 = &expr[jj][p+3];
+          double *y3 = expr[jj].data() + p + 3;
-          double *y4 = &expr[jj][p];
+          double *y4 = expr[jj].data() + p;
           ai = n;
           yprod = expr[jj][ai];
           for (a2=0;a2<a;a2++) {
@@ -545,8 +545,8 @@ void Fingerprint_bondscreened::do3bodyfeatureset_singleneighborloop(double * fea
           if (atomtypes[1] != nelements && atomtypes[1] != jtype) {
             continue;
           }
-          double *y3 = &expr[jj][p+3];
+          double *y3 = expr[jj].data() + p + 3;
-          double *y4 = &expr[jj][p];
+          double *y4 = expr[jj].data() + p;
           ai = n;
           yprod = expr[jj][ai];
           for (a2=0;a2<a;a2++) {
@@ -615,17 +615,20 @@ void Fingerprint_bondscreened::do3bodyfeatureset_doubleneighborloop(double * fea
   i = ilist[ii];
   itype = pair->map[type[i]];
   int f = pair->net[itype].dimensions[0];
-  double expr[jnum][kmax];
+  std::vector<double> row(kmax, 0.0);
-  double y[jnum][3];
+  std::vector<std::vector<double>> expr(jnum, row);
-  double ri[jnum];
+  std::vector<double> yrow(3, 0.0);
-  double dfc[jnum];
+  std::vector<std::vector<double>> y(jnum, yrow);
+  std::vector<double> ri(jnum, 0.0);
+  std::vector<double> dfc(jnum, 0.0);
   int kb = kmax;
   int mb = mlength;
-  double Bijk[kb][mb];
+  std::vector<double> brow(mb, 0.0);
-  double c41[kmax];
+  std::vector<std::vector<double>> Bijk(kb, brow);
-  double c51[kmax];
+  std::vector<double> c41(kmax, 0.0);
-  double c61[kmax];
+  std::vector<double> c51(kmax, 0.0);
-  double ct[kmax];
+  std::vector<double> c61(kmax, 0.0);
+  std::vector<double> ct(kmax, 0.0);
   for (jj = 0; jj < jnum; jj++) {
     if (Bij[jj]==false) {continue;}
     jtype = tn[jj];

src/ML-RANN/rann_fingerprint_bondscreenedspin.cpp

@@ -35,6 +35,8 @@ DISTRIBUTION A. Approved for public release; distribution unlimited. OPSEC#4918
 
 using namespace LAMMPS_NS::RANN;
 
+static constexpr double SMALL = 1.0e-12;
+
 Fingerprint_bondscreenedspin::Fingerprint_bondscreenedspin(PairRANN *_pair) : Fingerprint(_pair)
 {
   n_body_type = 3;
@@ -320,7 +322,8 @@ void Fingerprint_bondscreenedspin::do3bodyfeatureset_singleneighborloop(double *
   i = ilist[ii];
   itype = pair->map[type[i]];
   int f = pair->net[itype].dimensions[0];
-  double expr[jnum][kmax+12];
+  std::vector<double> row(kmax+12, 0.0);
+  std::vector<std::vector<double>> expr(jnum, row);
   int p = kmax;
   int countmb=((mlength)*(mlength+1))>>1;
   double *si = sim->s[i];
@@ -360,13 +363,13 @@ void Fingerprint_bondscreenedspin::do3bodyfeatureset_singleneighborloop(double *
     expr[jj][p+1]=dely*rinvs;
     expr[jj][p+2]=delz*rinvs;
     //Hack to avoid nan when x y or z component of radial vector is exactly 0. Shouldn't affect accuracy.
-    if (expr[jj][p]*expr[jj][p]<0.000000000001) {
+    if (expr[jj][p]*expr[jj][p] < SMALL) {
       expr[jj][p] = 0.000001;
     }
-    if (expr[jj][p+1]*expr[jj][p+1]<0.000000000001) {
+    if (expr[jj][p+1]*expr[jj][p+1] < SMALL) {
       expr[jj][p+1] = 0.000001;
     }
-    if (expr[jj][p+2]*expr[jj][p+2]<0.000000000001) {
+    if (expr[jj][p+2]*expr[jj][p+2] < SMALL) {
       expr[jj][p+2] = 0.000001;
     }
     expr[jj][p+3] = -dfc*expr[jj][p]-dSikx[jj];
@@ -383,8 +386,8 @@ void Fingerprint_bondscreenedspin::do3bodyfeatureset_singleneighborloop(double *
   int kb = kmax;
   int mb = mlength;
   count = startingneuron;
-  double Bb[mb];
+  std::vector<double> Bb(mb, 0.0);
-  double Bbs[mb];
+  std::vector<double> Bbs(mb, 0.0);
   double dBbx;
   double dBby;
   double dBbz;
@@ -421,7 +424,7 @@ void Fingerprint_bondscreenedspin::do3bodyfeatureset_singleneighborloop(double *
         double *sj = sim->s[j];
         double sp = si[0]*sj[0]+si[1]*sj[1]+si[2]*sj[2];
         double yprod = expr[jj][ai];
-        double *y4 = &expr[jj][p];
+        double *y4 = expr[jj].data() + p;
         for (a2=0;a2<a;a2++) {
           yprod *= y4[M[a2+1]];
         }
@@ -432,12 +435,8 @@ void Fingerprint_bondscreenedspin::do3bodyfeatureset_singleneighborloop(double *
         }
       }
       if (atomtypes[1]!=atomtypes[2]) {//Bb!=Bg
-        double Bg[mb];
+        std::vector<double> Bg(mb, 0.0);
-        double Bgs[mb];
+        std::vector<double> Bgs(mb, 0.0);
-        for (a1=0;a1<mb;a1++) {
-          Bg[a1]=0;
-          Bgs[a1]=0;
-        }
         ai = n;
         double y1 = alpha_k[ai]/re;
         //loop over ktype to get Bg
@@ -452,7 +451,7 @@ void Fingerprint_bondscreenedspin::do3bodyfeatureset_singleneighborloop(double *
           double *sj = sim->s[j];
           double sp = si[0]*sj[0]+si[1]*sj[1]+si[2]*sj[2];
           double yprod = expr[jj][ai];
-          double *y4 = &expr[jj][p];
+          double *y4 = expr[jj].data() + p;
           for (a2=0;a2<a;a2++) {
             yprod *= y4[M[a2+1]];
           }
@@ -474,8 +473,8 @@ void Fingerprint_bondscreenedspin::do3bodyfeatureset_singleneighborloop(double *
           j = jl[jj];
           double *sj = sim->s[j];
           double sp = si[0]*sj[0]+si[1]*sj[1]+si[2]*sj[2];
-          double *y3 = &expr[jj][p+3];
+          double *y3 = expr[jj].data() + p + 3;
-          double *y4 = &expr[jj][p];
+          double *y4 = expr[jj].data() + p;
           ai = n;
           yprod = expr[jj][ai];
           for (a2=0;a2<a;a2++) {
@@ -517,8 +516,8 @@ void Fingerprint_bondscreenedspin::do3bodyfeatureset_singleneighborloop(double *
           j = jl[jj];
           double *sj = sim->s[j];
           double sp = si[0]*sj[0]+si[1]*sj[1]+si[2]*sj[2];
-          double *y3 = &expr[jj][p+3];
+          double *y3 = expr[jj].data() + p + 3;
-          double *y4 = &expr[jj][p];
+          double *y4 = expr[jj].data() + p;
           ai = n;
           yprod = expr[jj][ai];
           for (a2=0;a2<a;a2++) {
@@ -568,8 +567,8 @@ void Fingerprint_bondscreenedspin::do3bodyfeatureset_singleneighborloop(double *
           j = jl[jj];
           double *sj = sim->s[j];
           double sp = si[0]*sj[0]+si[1]*sj[1]+si[2]*sj[2];
-          double *y3 = &expr[jj][p+3];
+          double *y3 = expr[jj].data() + p + 3;
-          double *y4 = &expr[jj][p];
+          double *y4 = expr[jj].data() + p;
           ai = n;
           yprod = expr[jj][ai];
           for (a2=0;a2<a;a2++) {
@@ -641,18 +640,21 @@ void Fingerprint_bondscreenedspin::do3bodyfeatureset_doubleneighborloop(double *
   i = ilist[ii];
   itype = pair->map[type[i]];
   int f = pair->net[itype].dimensions[0];
-  double expr[jnum][kmax];
+  std::vector<double> row(kmax, 0.0);
-  double y[jnum][3];
+  std::vector<std::vector<double>> expr(jnum, row);
-  double ri[jnum];
+  std::vector<double> yrow(3, 0.0);
-  double dfc[jnum];
+  std::vector<std::vector<double>> y(jnum, yrow);
+  std::vector<double> ri(jnum, 0.0);
+  std::vector<double> dfc(jnum, 0.0);
   int kb = kmax;
   int mb = mlength;
-  double Bijk[kb][mb];
-  double c41[kmax];
-  double c51[kmax];
-  double c61[kmax];
-  double ct[kmax];
   double *si = sim->s[i];
+  std::vector<double> brow(mb, 0.0);
+  std::vector<std::vector<double>> Bijk(kb, brow);
+  std::vector<double> c41(kmax, 0.0);
+  std::vector<double> c51(kmax, 0.0);
+  std::vector<double> c61(kmax, 0.0);
+  std::vector<double> ct(kmax, 0.0);
   for (jj = 0; jj < jnum; jj++) {
     if (Bij[jj]==false) {continue;}
       jtype = tn[jj];

src/ML-RANN/rann_fingerprint_bondspin.cpp

@@ -35,6 +35,8 @@ DISTRIBUTION A. Approved for public release; distribution unlimited. OPSEC#4918
 
 using namespace LAMMPS_NS::RANN;
 
+static constexpr double SMALL = 1.0e-12;
+
 Fingerprint_bondspin::Fingerprint_bondspin(PairRANN *_pair) : Fingerprint(_pair)
 {
   n_body_type = 3;
@@ -317,7 +319,8 @@ void Fingerprint_bondspin::do3bodyfeatureset_singleneighborloop(double * feature
   i = ilist[ii];
   itype = pair->map[type[i]];
   int f = pair->net[itype].dimensions[0];
-  double expr[jnum][kmax+12];
+  std::vector<double> row(kmax+12, 0.0);
+  std::vector<std::vector<double>> expr(jnum, row);
   int p = kmax;
   int countmb=((mlength)*(mlength+1))>>1;
   double *si = sim->s[i];
@@ -355,13 +358,13 @@ void Fingerprint_bondspin::do3bodyfeatureset_singleneighborloop(double * feature
     expr[jj][p+1]=dely*rinvs;
     expr[jj][p+2]=delz*rinvs;
     //Hack to avoid nan when x y or z component of radial vector is exactly 0. Shouldn't affect accuracy.
-    if (expr[jj][p]*expr[jj][p]<0.000000000001) {
+    if (expr[jj][p]*expr[jj][p] < SMALL) {
       expr[jj][p] = 0.000001;
     }
-    if (expr[jj][p+1]*expr[jj][p+1]<0.000000000001) {
+    if (expr[jj][p+1]*expr[jj][p+1] < SMALL) {
       expr[jj][p+1] = 0.000001;
     }
-    if (expr[jj][p+2]*expr[jj][p+2]<0.000000000001) {
+    if (expr[jj][p+2]*expr[jj][p+2] < SMALL) {
       expr[jj][p+2] = 0.000001;
     }
     expr[jj][p+3] = -dfc*expr[jj][p];
@@ -378,8 +381,8 @@ void Fingerprint_bondspin::do3bodyfeatureset_singleneighborloop(double * feature
   int kb = kmax;
   int mb = mlength;
   count = startingneuron;
-  double Bb[mb];
+  std::vector<double> Bb(mb, 0.0);
-  double Bbs[mb];
+  std::vector<double> Bbs(mb, 0.0);
   double dBbx;
   double dBby;
   double dBbz;
@@ -415,7 +418,7 @@ void Fingerprint_bondspin::do3bodyfeatureset_singleneighborloop(double * feature
         double *sj = sim->s[j];
         double sp = si[0]*sj[0]+si[1]*sj[1]+si[2]*sj[2];
         double yprod = expr[jj][ai];
-        double *y4 = &expr[jj][p];
+        double *y4 = expr[jj].data() + p;
         for (a2=0;a2<a;a2++) {
           yprod *= y4[M[a2+1]];
         }
@@ -426,12 +429,8 @@ void Fingerprint_bondspin::do3bodyfeatureset_singleneighborloop(double * feature
         }
       }
       if (atomtypes[1]!=atomtypes[2]) {//Bb!=Bg
-        double Bg[mb];
+        std::vector<double> Bg(mb, 0.0);
-        double Bgs[mb];
+        std::vector<double> Bgs(mb, 0.0);
-        for (a1=0;a1<mb;a1++) {
-          Bg[a1]=0;
-          Bgs[a1]=0;
-        }
         ai = n;
         double y1 = alpha_k[ai]/re;
         //loop over ktype to get Bg
@@ -445,7 +444,7 @@ void Fingerprint_bondspin::do3bodyfeatureset_singleneighborloop(double * feature
           double *sj = sim->s[j];
           double sp = si[0]*sj[0]+si[1]*sj[1]+si[2]*sj[2];
           double yprod = expr[jj][ai];
-          double *y4 = &expr[jj][p];
+          double *y4 = expr[jj].data() + p;
           for (a2=0;a2<a;a2++) {
             yprod *= y4[M[a2+1]];
           }
@@ -466,8 +465,8 @@ void Fingerprint_bondspin::do3bodyfeatureset_singleneighborloop(double * feature
           j = jl[jj];
           double *sj = sim->s[j];
           double sp = si[0]*sj[0]+si[1]*sj[1]+si[2]*sj[2];
-          double *y3 = &expr[jj][p+3];
+          double *y3 = expr[jj].data() + p + 3;
-          double *y4 = &expr[jj][p];
+          double *y4 = expr[jj].data() + p;
           ai = n;
           yprod = expr[jj][ai];
           for (a2=0;a2<a;a2++) {
@@ -502,8 +501,8 @@ void Fingerprint_bondspin::do3bodyfeatureset_singleneighborloop(double * feature
           j = jl[jj];
           double *sj = sim->s[j];
           double sp = si[0]*sj[0]+si[1]*sj[1]+si[2]*sj[2];
-          double *y3 = &expr[jj][p+3];
+          double *y3 = expr[jj].data() + p + 3;
-          double *y4 = &expr[jj][p];
+          double *y4 = expr[jj].data() + p;
           ai = n;
           yprod = expr[jj][ai];
           for (a2=0;a2<a;a2++) {
@@ -546,8 +545,8 @@ void Fingerprint_bondspin::do3bodyfeatureset_singleneighborloop(double * feature
           j = jl[jj];
           double *sj = sim->s[j];
           double sp = si[0]*sj[0]+si[1]*sj[1]+si[2]*sj[2];
-          double *y3 = &expr[jj][p+3];
+          double *y3 = expr[jj].data() + p + 3;
-          double *y4 = &expr[jj][p];
+          double *y4 = expr[jj].data() + p;
           ai = n;
           yprod = expr[jj][ai];
           for (a2=0;a2<a;a2++) {
@@ -612,16 +611,18 @@ void Fingerprint_bondspin::do3bodyfeatureset_doubleneighborloop(double * feature
   i = ilist[ii];
   itype = pair->map[type[i]];
   int f = pair->net[itype].dimensions[0];
-  double expr[jnum][kmax];
+  std::vector<double> row(kmax, 0.0);
-  double y[jnum][3];
+  std::vector<std::vector<double>> expr(jnum, row);
-  double ri[jnum];
+  std::vector<double> yrow(3, 0.0);
-  double dfc[jnum];
+  std::vector<std::vector<double>> y(jnum, yrow);
+  std::vector<double> ri(jnum, 0.0);
+  std::vector<double> dfc(jnum, 0.0);
   int kb = kmax;
   int mb = mlength;
-  double c41[kmax];
+  std::vector<double> c41(kmax, 0.0);
-  double c51[kmax];
+  std::vector<double> c51(kmax, 0.0);
-  double c61[kmax];
+  std::vector<double> c61(kmax, 0.0);
-  double ct[kmax];
+  std::vector<double> ct(kmax, 0.0);
   double *si = sim->s[i];
   for (jj = 0; jj < jnum; jj++) {
     jtype = tn[jj];

src/Makefile

@@ -2,6 +2,12 @@
 
 SHELL = /bin/bash
 PYTHON = python
+PYTHON_VERSION=$(word 2,$(shell $(PYTHON) -V 2>&1 | tr '.' ' '))
+ifeq ($(PYTHON_VERSION),2)
+PYTHON_CHECK=@echo LAMMPS requires Python 3. Try: 'make PYTHON=python3 ...' ; exit 1
+else
+PYTHON_CHECK=@echo
+endif
 DYN_LIB = -ldl
 
 #.IGNORE:
@@ -276,7 +282,7 @@ LIBDIR = $(@:lib-%=%)
 # List of all targets
 
 help:
-	@echo ''
+	$(PYTHON_CHECK)
 	@echo 'make clean-all           delete all object files'
 	@echo 'make clean-machine       delete object files for one machine'
 	@echo 'make mpi-stubs           build dummy MPI library in STUBS'
@@ -378,6 +384,7 @@ gitversion:
 # shared = shared compile in Obj_shared_machine
 
 .DEFAULT:
+	$(PYTHON_CHECK)
 	@if [ $@ = "serial" ]; \
 	  then cd STUBS; $(MAKE); cd ..; fi
 	@test -f MAKE/Makefile.$@ -o -f MAKE/OPTIONS/Makefile.$@ -o \
@@ -465,6 +472,7 @@ mpi-stubs:
 
 sinclude ../lib/python/Makefile.lammps
 install-python:
+	$(PYTHON_CHECK)
 	@rm -rf ../python/build
 	@$(PYTHON) ../python/install.py -p ../python/lammps -l ../src/liblammps.so -w $(PWD) -v $(PWD)/version.h
 
@@ -483,49 +491,59 @@ tar:
 check: check-whitespace check-permissions check-homepage check-errordocs check-fmtlib check-docs check-version
 
 check-whitespace:
+	$(PYTHON_CHECK)
 	$(PYTHON) ../tools/coding_standard/whitespace.py ..
 
 fix-whitespace:
+	$(PYTHON_CHECK)
 	$(PYTHON) ../tools/coding_standard/whitespace.py .. -f
 
 check-permissions:
+	$(PYTHON_CHECK)
 	$(PYTHON) ../tools/coding_standard/permissions.py ..
 
 fix-permissions:
+	$(PYTHON_CHECK)
 	$(PYTHON) ../tools/coding_standard/permissions.py .. -f
 
 check-homepage:
+	$(PYTHON_CHECK)
 	$(PYTHON) ../tools/coding_standard/homepage.py ..
 
 fix-homepage:
+	$(PYTHON_CHECK)
 	$(PYTHON) ../tools/coding_standard/homepage.py .. -f
 
 check-errordocs:
+	$(PYTHON_CHECK)
 	$(PYTHON) ../tools/coding_standard/errordocs.py ..
 
 fix-errordocs:
+	$(PYTHON_CHECK)
 	$(PYTHON) ../tools/coding_standard/errordocs.py .. -f
 
 check-fmtlib:
+	$(PYTHON_CHECK)
 	$(PYTHON) ../tools/coding_standard/fmtlib.py ..
 
 fix-fmtlib:
+	$(PYTHON_CHECK)
 	$(PYTHON) ../tools/coding_standard/fmtlib.py .. -f
 
 check-docs:
+	$(PYTHON_CHECK)
 	$(MAKE) $(MFLAGS) -C ../doc anchor_check style_check package_check role_check
 
 check-version:
+	$(PYTHON_CHECK)
 	$(PYTHON) ../tools/coding_standard/versiontags.py .. || echo
 
-
 format-src:
 	clang-format -i --verbose --style=file *.cpp *.h */*.cpp */*.h
 
 format-tests:
 	clang-format -i --verbose --style=file ../unittest/*/*.cpp ../unittest/*/*.h
 
-
 # Package management
 
 package:

src/PYTHON/fix_python_invoke.cpp

@@ -70,7 +70,7 @@ FixPythonInvoke::FixPythonInvoke(LAMMPS *lmp, int narg, char **arg) :
     error->all(FLERR,"Could not find Python function");
   }
 
-  lmpPtr = PY_VOID_POINTER(lmp);
+  lmpPtr = PyCapsule_New((void *)lmp, nullptr, nullptr);
 
   // nvalid = next step on which end_of_step or post_force does something
   // add nextvalid() to all computes that store invocation times

src/PYTHON/fix_python_move.cpp

@@ -44,7 +44,7 @@ FixPythonMove::FixPythonMove(LAMMPS *lmp, int narg, char **arg) :
 
   // add current directory to PYTHONPATH
   PyObject *py_path = PySys_GetObject((char *)"path");
-  PyList_Append(py_path, PY_STRING_FROM_STRING("."));
+  PyList_Append(py_path, PyUnicode_FromString("."));
 
 
   // create integrator instance
@@ -73,7 +73,7 @@ FixPythonMove::FixPythonMove(LAMMPS *lmp, int narg, char **arg) :
     error->all(FLERR,"Could not find integrator class {} in module {}", cls_name, module_name);
   }
 
-  PyObject *ptr = PY_VOID_POINTER(lmp);
+  PyObject *ptr = PyCapsule_New((void *)lmp, nullptr, nullptr);
   PyObject *py_move_obj = PyObject_CallFunction(py_move_type, (char *)"O", ptr);
   Py_CLEAR(ptr);
 

src/PYTHON/pair_python.cpp

@@ -54,14 +54,14 @@ PairPython::PairPython(LAMMPS *lmp) : Pair(lmp) {
 
   PyUtils::GIL lock;
   PyObject *py_path = PySys_GetObject((char *)"path");
-  PyList_Append(py_path, PY_STRING_FROM_STRING("."));
+  PyList_Append(py_path, PyUnicode_FromString("."));
 
   // if LAMMPS_POTENTIALS environment variable is set,
   // add it to PYTHONPATH as well
 
   const char *potentials_path = getenv("LAMMPS_POTENTIALS");
   if (potentials_path != nullptr) {
-    PyList_Append(py_path, PY_STRING_FROM_STRING(potentials_path));
+    PyList_Append(py_path, PyUnicode_FromString(potentials_path));
   }
 }
 

src/PYTHON/python_compat.h

@@ -16,22 +16,11 @@
 
 #include <Python.h>
 
-// Wrap API changes between Python 2 and 3 using macros
+#if PY_VERSION_HEX < 0x030600f0
-#if PY_MAJOR_VERSION == 2
+#error Python version 3.6 or later is required by LAMMPS
-#if defined(_MSC_VER) || defined(__MINGW32__)
-#define PY_INT_FROM_LONG(X) PyLong_FromLongLong(X)
-#define PY_INT_AS_LONG(X) PyLong_AsLongLong(X)
-#define PY_LONG_FROM_STRING(X) std::stoll(X)
-#else
-#define PY_INT_FROM_LONG(X) PyInt_FromLong(X)
-#define PY_INT_AS_LONG(X) PyInt_AsLong(X)
-#define PY_LONG_FROM_STRING(X) std::stol(X)
 #endif
-#define PY_STRING_FROM_STRING(X) PyString_FromString(X)
-#define PY_VOID_POINTER(X) PyCObject_FromVoidPtr((void *) X, nullptr)
-#define PY_STRING_AS_STRING(X) PyString_AsString(X)
 
-#elif PY_MAJOR_VERSION == 3
+// Wrap API differences between platforms using macros
 #if defined(_MSC_VER) || defined(__MINGW32__)
 #define PY_INT_FROM_LONG(X) PyLong_FromLongLong(X)
 #define PY_INT_AS_LONG(X) PyLong_AsLongLong(X)
@@ -41,9 +30,5 @@
 #define PY_INT_AS_LONG(X) PyLong_AsLong(X)
 #define PY_LONG_FROM_STRING(X) std::stol(X)
 #endif
-#define PY_STRING_FROM_STRING(X) PyUnicode_FromString(X)
-#define PY_VOID_POINTER(X) PyCapsule_New((void *) X, nullptr, nullptr)
-#define PY_STRING_AS_STRING(X) PyUnicode_AsUTF8(X)
-#endif
 
 #endif

src/PYTHON/python_impl.cpp

@@ -361,12 +361,12 @@ void PythonImpl::invoke_function(int ifunc, char *result)
         if (!str)
           error->all(FLERR, "Could not evaluate Python function {} input variable: {}",
                      pfuncs[ifunc].name, pfuncs[ifunc].svalue[i]);
-        pValue = PY_STRING_FROM_STRING(str);
+        pValue = PyUnicode_FromString(str);
       } else {
-        pValue = PY_STRING_FROM_STRING(pfuncs[ifunc].svalue[i]);
+        pValue = PyUnicode_FromString(pfuncs[ifunc].svalue[i]);
       }
     } else if (itype == PTR) {
-      pValue = PY_VOID_POINTER(lmp);
+      pValue = PyCapsule_New((void *)lmp, nullptr, nullptr);
     } else {
       error->all(FLERR, "Unsupported variable type: {}", itype);
     }
@@ -397,7 +397,7 @@ void PythonImpl::invoke_function(int ifunc, char *result)
       auto value = fmt::format("{:.15g}", PyFloat_AsDouble(pValue));
       strncpy(result, value.c_str(), Variable::VALUELENGTH - 1);
     } else if (otype == STRING) {
-      const char *pystr = PY_STRING_AS_STRING(pValue);
+      const char *pystr = PyUnicode_AsUTF8(pValue);
       if (pfuncs[ifunc].longstr)
         strncpy(pfuncs[ifunc].longstr, pystr, pfuncs[ifunc].length_longstr);
       else

src/REAXFF/fix_reaxff_species.cpp

@@ -965,6 +965,7 @@ void FixReaxFFSpecies::DeleteSpecies(int Nmole, int Nspec)
   int *mask = atom->mask;
   double *mass = atom->mass;
   double *rmass = atom->rmass;
+  int *type = atom->type;
   double localmass, totalmass;
   std::string species_str;
 
@@ -1013,12 +1014,12 @@ void FixReaxFFSpecies::DeleteSpecies(int Nmole, int Nspec)
       if (!(mask[i] & groupbit)) continue;
       cid = nint(clusterID[i]);
       if (cid == m) {
-        itype = ele2uele[atom->type[i] - 1];
+        itype = ele2uele[type[i] - 1];
         Name[itype]++;
         count++;
         marklist[nmarklist++] = i;
         if (rmass) localmass += rmass[i];
-        else localmass += atom->mass[atom->type[i]];
+        else localmass += mass[type[i]];
       }
     }
 

src/angle_write.cpp

@@ -26,6 +26,7 @@
 #include "force.h"
 #include "input.h"
 #include "math_const.h"
+#include "safe_pointers.h"
 #include "update.h"
 
 #include <cmath>
@@ -79,13 +80,12 @@ void AngleWrite::command(int narg, char **arg)
   // otherwise make certain that units are consistent
   // print header in format used by angle_style table
 
-  FILE *fp = nullptr;
+  SafeFilePtr fp;
   std::string coeffs_file = table_file + ".tmp.coeffs";
   if (comm->me == 0) {
 
-    fp = fopen(coeffs_file.c_str(), "w");
+    SafeFilePtr coeffs = fopen(coeffs_file.c_str(), "w");
-    force->angle->write_data(fp);
+    force->angle->write_data(coeffs);
-    fclose(fp);
 
     // units sanity check:
     // - if this is the first time we write to this potential file,
@@ -145,16 +145,14 @@ void AngleWrite::command(int narg, char **arg)
     writer->input->one("pair_coeff * *");
     writer->input->one("mass * 1.0");
     writer->input->one(fmt::format("angle_style {}", force->angle_style));
-    FILE *coeffs;
     char line[MAXLINE] = {'\0'};
-    coeffs = fopen(coeffs_file.c_str(), "r");
+    SafeFilePtr coeffs = fopen(coeffs_file.c_str(), "r");
     if (!coeffs)
       error->one(FLERR, "Unable to open temporary file {}: {}", coeffs_file, utils::getsyserror());
     for (int i = 0; i < atom->nangletypes; ++i) {
       utils::sfgets(FLERR, line, MAXLINE, coeffs, coeffs_file.c_str(), error);
       writer->input->one(fmt::format("angle_coeff {}", line));
     }
-    fclose(coeffs);
     platform::unlink(coeffs_file);
 
     // initialize system
@@ -221,7 +219,6 @@ void AngleWrite::command(int narg, char **arg)
 
     // clean up
     delete writer;
-    fclose(fp);
   }
   MPI_Comm_free(&singlecomm);
 }

src/bond.cpp

@@ -20,6 +20,7 @@
 #include "force.h"
 #include "memory.h"
 #include "neighbor.h"
+#include "safe_pointers.h"
 #include "suffix.h"
 #include "update.h"
 
@@ -365,7 +366,7 @@ void Bond::write_file(int narg, char **arg)
   // add line with DATE: and UNITS: tag when creating new file
   // print header in format used by bond_style table
 
-  FILE *fp = nullptr;
+  SafeFilePtr fp;
   if (comm->me == 0) {
     std::string table_file = arg[4];
 
@@ -419,7 +420,6 @@ void Bond::write_file(int narg, char **arg)
       e = single(btype, r * r, itype, jtype, f);
       fprintf(fp, "%8d %- 22.15g %- 22.15g %- 22.15g\n", i + 1, r, e, f * r);
     }
-    fclose(fp);
   }
 }
 

src/create_atoms.cpp

@@ -34,6 +34,7 @@
 #include "random_mars.h"
 #include "random_park.h"
 #include "region.h"
+#include "safe_pointers.h"
 #include "special.h"
 #include "text_file_reader.h"
 #include "variable.h"
@@ -1082,8 +1083,9 @@ void CreateAtoms::add_mesh(const char *filename)
     molid = maxmol + 1;
   }
 
-  FILE *fp = fopen(filename, "rb");
+  SafeFilePtr fp = fopen(filename, "rb");
-  if (fp == nullptr) error->one(FLERR, "Cannot open file {}: {}", filename, utils::getsyserror());
+  if (fp == nullptr)
+    error->one(FLERR, "Cannot open STL mesh file {}: {}", filename, utils::getsyserror());
 
   // first try reading the file in ASCII format
 
@@ -1184,7 +1186,7 @@ void CreateAtoms::add_mesh(const char *filename)
         }
       }
     } else {
-      error->all(FLERR, "Error reading triangles from file {}: {}", filename, e.what());
+      error->all(FLERR, "Error reading triangles from STL mesh file {}: {}", filename, e.what());
     }
   }
 
@@ -1195,7 +1197,6 @@ void CreateAtoms::add_mesh(const char *filename)
       utils::logmesg(lmp, "  read {} triangles with {:.2f} atoms per triangle added in {} mode\n",
                      ntriangle, ratio, mesh_name[mesh_style]);
   }
-  if (fp) fclose(fp);
 }
 
 /* ----------------------------------------------------------------------

src/fix_adapt.cpp

@@ -54,7 +54,7 @@ FixAdapt::FixAdapt(LAMMPS *lmp, int narg, char **arg) :
 {
   if (narg < 5) utils::missing_cmd_args(FLERR,"fix adapt", error);
   nevery = utils::inumeric(FLERR,arg[3],false,lmp);
-  if (nevery < 0) error->all(FLERR,"Illegal fix adapt every value {}", nevery);
+  if (nevery < 0) error->all(FLERR, 3, "Illegal fix adapt every value {}", nevery);
 
   dynamic_group_allow = 1;
   create_attribute = 1;
@@ -92,7 +92,7 @@ FixAdapt::FixAdapt(LAMMPS *lmp, int narg, char **arg) :
     } else break;
   }
 
-  if (nadapt == 0) error->all(FLERR,"Nothing to adapt in fix adapt command");
+  if (nadapt == 0) error->all(FLERR, 3, "Nothing to adapt in fix adapt command");
   adapt = new Adapt[nadapt];
 
   // parse keywords
@@ -125,7 +125,7 @@ FixAdapt::FixAdapt(LAMMPS *lmp, int narg, char **arg) :
 
       if (utils::strmatch(arg[iarg+5],"^v_")) {
         adapt[nadapt].var = utils::strdup(arg[iarg+5]+2);
-      } else error->all(FLERR,"Argument #{} must be variable not {}", iarg+6, arg[iarg+5]);
+      } else error->all(FLERR, iarg+5, "Argument must be variable not {}", arg[iarg+5]);
       nadapt++;
       iarg += 6;
 
@@ -138,7 +138,7 @@ FixAdapt::FixAdapt(LAMMPS *lmp, int narg, char **arg) :
                               adapt[nadapt].ilo, adapt[nadapt].ihi, lmp, Atom::BOND);
       if (utils::strmatch(arg[iarg+4],"^v_")) {
         adapt[nadapt].var = utils::strdup(arg[iarg+4]+2);
-      } else error->all(FLERR,"Argument #{} must be variable not {}", iarg+5, arg[iarg+4]);
+      } else error->all(FLERR, iarg+4, "Argument must be variable not {}", arg[iarg+4]);
       nadapt++;
       iarg += 5;
 
@@ -151,7 +151,7 @@ FixAdapt::FixAdapt(LAMMPS *lmp, int narg, char **arg) :
                               adapt[nadapt].ilo, adapt[nadapt].ihi, lmp, Atom::ANGLE);
       if (utils::strmatch(arg[iarg+4],"^v_")) {
         adapt[nadapt].var = utils::strdup(arg[iarg+4]+2);
-      } else error->all(FLERR,"Argument #{} must be variable not {}", iarg+5, arg[iarg+4]);
+      } else error->all(FLERR,iarg+4, "Argument must be variable not {}", arg[iarg+4]);
       nadapt++;
       iarg += 5;
 
@@ -172,7 +172,7 @@ FixAdapt::FixAdapt(LAMMPS *lmp, int narg, char **arg) :
       adapt[nadapt].which = KSPACE;
       if (utils::strmatch(arg[iarg+1],"^v_")) {
         adapt[nadapt].var = utils::strdup(arg[iarg+1]+2);
-      } else error->all(FLERR,"Argument #{} must be variable not {}", iarg+2, arg[iarg+1]);
+      } else error->all(FLERR, iarg+1, "Argument must be variable not {}", arg[iarg+1]);
       nadapt++;
       iarg += 2;
 
@@ -187,10 +187,11 @@ FixAdapt::FixAdapt(LAMMPS *lmp, int narg, char **arg) :
       } else if (strcmp(arg[iarg+1],"charge") == 0) {
         adapt[nadapt].atomparam = CHARGE;
         chgflag = 1;
-      } else error->all(FLERR,"Unsupported per-atom property {} for fix adapt", arg[iarg+1]);
+      } else error->all(FLERR, iarg+1, "Unsupported per-atom property {} for fix adapt",
+                        arg[iarg+1]);
       if (utils::strmatch(arg[iarg+2],"^v_")) {
         adapt[nadapt].var = utils::strdup(arg[iarg+2]+2);
-      } else error->all(FLERR,"Argument #{} must be variable not {}", iarg+3, arg[iarg+2]);
+      } else error->all(FLERR, iarg+2, "Argument must be variable not {}", arg[iarg+2]);
       nadapt++;
       iarg += 3;
     } else break;
@@ -359,7 +360,8 @@ void FixAdapt::init()
   if (group->dynamic[igroup])
     for (i = 0; i < nadapt; i++)
       if (adapt[i].which == ATOM)
-        error->all(FLERR,"Cannot use dynamic group {} with fix adapt atom", group->names[igroup]);
+        error->all(FLERR, Error::NOLASTLINE,
+                   "Cannot use dynamic group {} with fix adapt atom", group->names[igroup]);
 
   // setup and error checks
 
@@ -373,9 +375,11 @@ void FixAdapt::init()
 
     ad->ivar = input->variable->find(ad->var);
     if (ad->ivar < 0)
-      error->all(FLERR,"Variable name {} for fix adapt does not exist", ad->var);
+      error->all(FLERR, Error::NOLASTLINE,
+                 "Variable name {} for fix adapt does not exist", ad->var);
     if (!input->variable->equalstyle(ad->ivar))
-      error->all(FLERR,"Variable {} for fix adapt is invalid style", ad->var);
+      error->all(FLERR, Error::NOLASTLINE,
+                 "Variable {} for fix adapt is invalid style", ad->var);
 
     if (ad->which == PAIR) {
       anypair = 1;
@@ -400,17 +404,20 @@ void FixAdapt::init()
           ad->pair = force->pair_match(fmt::format("{}/{}",pstyle,lmp->suffix2),1,nsub);
       }
       if (ad->pair == nullptr) ad->pair = force->pair_match(pstyle,1,nsub);
-      if (ad->pair == nullptr) error->all(FLERR,"Fix adapt pair style {} not found", pstyle);
+      if (ad->pair == nullptr)
+        error->all(FLERR, Error::NOLASTLINE, "Fix adapt pair style {} not found", pstyle);
 
       void *ptr = ad->pair->extract(ad->pparam,ad->pdim);
       if (ptr == nullptr)
-        error->all(FLERR,"Fix adapt pair style {} param {} not supported", ad->pstyle, ad->pparam);
+        error->all(FLERR, Error::NOLASTLINE,
+                   "Fix adapt pair style {} param {} not supported", ad->pstyle, ad->pparam);
 
       // for pair styles only parameters that are 2-d arrays in atom types or
       // scalars are supported
 
       if (ad->pdim != 2 && ad->pdim != 0)
-        error->all(FLERR,"Pair style parameter {} is not compatible with fix adapt", ad->pparam);
+        error->all(FLERR, Error::NOLASTLINE,
+                   "Pair style parameter {} is not compatible with fix adapt", ad->pparam);
 
       if (ad->pdim == 2) ad->array = (double **) ptr;
       if (ad->pdim == 0) ad->scalar = (double *) ptr;
@@ -423,7 +430,7 @@ void FixAdapt::init()
           for (i = ad->ilo; i <= ad->ihi; i++) {
             for (j = MAX(ad->jlo,i); j <= ad->jhi; j++) {
               if (!pair->check_ijtype(i,j,pstyle))
-                error->all(FLERR,"Fix adapt type pair range is not valid "
+                error->all(FLERR, Error::NOLASTLINE, "Fix adapt type pair range is not valid "
                            "for pair hybrid sub-style {}", pstyle);
             }
           }
@@ -442,12 +449,13 @@ void FixAdapt::init()
 
       if (ad->bond == nullptr) ad->bond = force->bond_match(bstyle);
       if (ad->bond == nullptr )
-        error->all(FLERR,"Fix adapt bond style {} does not exist", bstyle);
+        error->all(FLERR, Error::NOLASTLINE,"Fix adapt bond style {} does not exist", bstyle);
 
       void *ptr = ad->bond->extract(ad->bparam,ad->bdim);
 
       if (ptr == nullptr)
-        error->all(FLERR,"Fix adapt bond style parameter {} not supported", ad->bparam);
+        error->all(FLERR, Error::NOLASTLINE,
+                   "Fix adapt bond style parameter {} not supported", ad->bparam);
 
       // for bond styles, use a vector
 
@@ -458,8 +466,8 @@ void FixAdapt::init()
         if (bond) {
           for (i = ad->ilo; i <= ad->ihi; i++) {
             if (!bond->check_itype(i,bstyle))
-              error->all(FLERR,"Fix adapt type bond range is not valid "
+              error->all(FLERR, Error::NOLASTLINE, "Fix adapt type bond range is not valid "
-                         "for pair hybrid sub-style {}", bstyle);
+                         "for bond hybrid sub-style {}", bstyle);
           }
         }
       }
@@ -476,12 +484,13 @@ void FixAdapt::init()
 
       if (ad->angle == nullptr) ad->angle = force->angle_match(astyle);
       if (ad->angle == nullptr )
-        error->all(FLERR,"Fix adapt angle style {} does not exist", astyle);
+        error->all(FLERR, Error::NOLASTLINE, "Fix adapt angle style {} does not exist", astyle);
 
       void *ptr = ad->angle->extract(ad->aparam,ad->adim);
 
       if (ptr == nullptr)
-        error->all(FLERR,"Fix adapt angle style parameter {} not supported", ad->aparam);
+        error->all(FLERR, Error::NOLASTLINE,
+                   "Fix adapt angle style parameter {} not supported", ad->aparam);
 
       // for angle styles, use a vector
 
@@ -492,8 +501,8 @@ void FixAdapt::init()
         if (angle) {
           for (i = ad->ilo; i <= ad->ihi; i++) {
             if (!angle->check_itype(i,astyle))
-              error->all(FLERR,"Fix adapt type angle range is not valid "
+              error->all(FLERR, Error::NOLASTLINE, "Fix adapt type angle range is not valid "
-                         "for pair hybrid sub-style {}", astyle);
+                         "for angle hybrid sub-style {}", astyle);
           }
         }
       }
@@ -536,22 +545,23 @@ void FixAdapt::init()
 
     } else if (ad->which == KSPACE) {
       if (force->kspace == nullptr)
-        error->all(FLERR,"Fix adapt expected a kspace style but none was defined");
+        error->all(FLERR, Error::NOLASTLINE,
+                   "Fix adapt expected a kspace style but none was defined");
       kspace_scale = (double *) force->kspace->extract("scale");
 
     } else if (ad->which == ATOM) {
       if (ad->atomparam == DIAMETER) {
         if (!atom->radius_flag)
-          error->all(FLERR,"Fix adapt requires atom attribute diameter");
+          error->all(FLERR, Error::NOLASTLINE, "Fix adapt requires atom attribute diameter");
         if (!atom->rmass_flag)
-          error->all(FLERR,"Fix adapt requires atom attribute mass");
+          error->all(FLERR, Error::NOLASTLINE, "Fix adapt requires atom attribute mass");
         if (discflag && domain->dimension != 2)
-          error->all(FLERR,"Fix adapt requires 2d simulation");
+          error->all(FLERR, Error::NOLASTLINE, "Fix adapt requires 2d simulation");
         if (!restart_reset) previous_diam_scale = 1.0;
       }
       if (ad->atomparam == CHARGE) {
         if (!atom->q_flag)
-          error->all(FLERR,"Fix adapt requires atom attribute charge");
+          error->all(FLERR, Error::NOLASTLINE, "Fix adapt requires atom attribute charge");
         if (!restart_reset) previous_chg_scale = 1.0;
       }
     }
@@ -589,11 +599,15 @@ void FixAdapt::init()
 
   if (id_fix_diam) {
     fix_diam = dynamic_cast<FixStoreAtom *>(modify->get_fix_by_id(id_fix_diam));
-    if (!fix_diam) error->all(FLERR,"Could not find fix adapt storage fix ID {}", id_fix_diam);
+    if (!fix_diam)
+      error->all(FLERR, Error::NOLASTLINE,
+                 "Could not find fix adapt storage fix ID {}", id_fix_diam);
   }
   if (id_fix_chg) {
     fix_chg = dynamic_cast<FixStoreAtom *>(modify->get_fix_by_id(id_fix_chg));
-    if (!fix_chg) error->all(FLERR,"Could not find fix adapt storage fix ID {}", id_fix_chg);
+    if (!fix_chg)
+      error->all(FLERR, Error::NOLASTLINE,
+                 "Could not find fix adapt storage fix ID {}", id_fix_chg);
   }
 
   if (utils::strmatch(update->integrate_style,"^respa"))

src/region.cpp

@@ -65,27 +65,39 @@ void Region::init()
 {
   if (xstr) {
     xvar = input->variable->find(xstr);
-    if (xvar < 0) error->all(FLERR, "Variable {} for region does not exist", xstr);
+    if (xvar < 0)
+      error->all(FLERR, Error::NOLASTLINE, "Variable {} for region {} {} does not exist", xstr,
+                 style, id);
     if (!input->variable->equalstyle(xvar))
-      error->all(FLERR, "Variable {} for region is invalid style", xstr);
+      error->all(FLERR, Error::NOLASTLINE, "Variable {} for region {} {} is invalid style", xstr,
+                 style, id);
   }
   if (ystr) {
     yvar = input->variable->find(ystr);
-    if (yvar < 0) error->all(FLERR, "Variable {} for region does not exist", ystr);
+    if (yvar < 0)
+      error->all(FLERR, Error::NOLASTLINE, "Variable {} for region {} {} does not exist", ystr,
+                 style, id);
     if (!input->variable->equalstyle(yvar))
-      error->all(FLERR, "Variable {} for region is not equal style", ystr);
+      error->all(FLERR, Error::NOLASTLINE, "Variable {} for region {} {} is not equal style", ystr,
+                 style, id);
   }
   if (zstr) {
     zvar = input->variable->find(zstr);
-    if (zvar < 0) error->all(FLERR, "Variable {} for region does not exist", zstr);
+    if (zvar < 0)
+      error->all(FLERR, Error::NOLASTLINE, "Variable {} for region {} {} does not exist", zstr,
+                 style, id);
     if (!input->variable->equalstyle(zvar))
-      error->all(FLERR, "Variable {} for region is not equal style", zstr);
+      error->all(FLERR, Error::NOLASTLINE, "Variable {} for region {} {} is not equal style", zstr,
+                 style, id);
   }
   if (tstr) {
     tvar = input->variable->find(tstr);
-    if (tvar < 0) error->all(FLERR, "Variable {} for region does not exist", tstr);
+    if (tvar < 0)
+      error->all(FLERR, Error::NOLASTLINE, "Variable {} for region {} {} does not exist", tstr,
+                 style, id);
     if (!input->variable->equalstyle(tvar))
-      error->all(FLERR, "Variable {} for region is not equal style", tstr);
+      error->all(FLERR, Error::NOLASTLINE, "Variable {} for region {} {} is not equal style", tstr,
+                 style, id);
   }
   vel_timestep = -1;
 }
@@ -303,6 +315,16 @@ void Region::options(int narg, char **arg)
 {
   if (narg < 0) utils::missing_cmd_args(FLERR, "region", error);
 
+  int offset = -20;
+  if (input && input->arg && arg) {
+    for (int i = 0; i < input->narg; ++i) {
+      if (arg[0] == input->arg[i]) {
+        offset = i;
+        break;
+      }
+    }
+  }
+
   // option defaults
 
   interior = 1;
@@ -321,7 +343,7 @@ void Region::options(int narg, char **arg)
       else if (strcmp(arg[iarg + 1], "lattice") == 0)
         scaleflag = 1;
       else
-        error->all(FLERR, "Illegal region units: {}", arg[iarg + 1]);
+        error->all(FLERR, iarg + 1 + offset, "Unknown region units: {}", arg[iarg + 1]);
       iarg += 2;
     } else if (strcmp(arg[iarg], "side") == 0) {
       if (iarg + 2 > narg) utils::missing_cmd_args(FLERR, "region side", error);
@@ -330,24 +352,27 @@ void Region::options(int narg, char **arg)
       else if (strcmp(arg[iarg + 1], "out") == 0)
         interior = 0;
       else
-        error->all(FLERR, "Illegal region side: {}", arg[iarg + 1]);
+        error->all(FLERR, iarg + 1 + offset, "Unknown region side setting: {}", arg[iarg + 1]);
       iarg += 2;
 
     } else if (strcmp(arg[iarg], "move") == 0) {
       if (iarg + 4 > narg) utils::missing_cmd_args(FLERR, "region move", error);
       if (strcmp(arg[iarg + 1], "NULL") != 0) {
         if (strstr(arg[iarg + 1], "v_") != arg[iarg + 1])
-          error->all(FLERR, "Illegal region move x displacement variable: {}", arg[iarg + 1]);
+          error->all(FLERR, iarg + 1 + offset, "Illegal region move x displacement variable: {}",
+                     arg[iarg + 1]);
         xstr = utils::strdup(&arg[iarg + 1][2]);
       }
       if (strcmp(arg[iarg + 2], "NULL") != 0) {
         if (strstr(arg[iarg + 2], "v_") != arg[iarg + 2])
-          error->all(FLERR, "Illegal region move y displacement variable: {}", arg[iarg + 2]);
+          error->all(FLERR, iarg + 2 + offset, "Illegal region move y displacement variable: {}",
+                     arg[iarg + 2]);
         ystr = utils::strdup(&arg[iarg + 2][2]);
       }
       if (strcmp(arg[iarg + 3], "NULL") != 0) {
         if (strstr(arg[iarg + 3], "v_") != arg[iarg + 3])
-          error->all(FLERR, "Illegal region move z displacement variable: {}", arg[iarg + 3]);
+          error->all(FLERR, iarg + 3 + offset, "Illegal region move z displacement variable: {}",
+                     arg[iarg + 3]);
         zstr = utils::strdup(&arg[iarg + 3][2]);
       }
       moveflag = 1;
@@ -369,19 +394,20 @@ void Region::options(int narg, char **arg)
     } else if (strcmp(arg[iarg], "open") == 0) {
       if (iarg + 2 > narg) utils::missing_cmd_args(FLERR, "region open", error);
       int iface = utils::inumeric(FLERR, arg[iarg + 1], false, lmp);
-      if (iface < 1 || iface > 6) error->all(FLERR, "Illegal region open face index: {}", iface);
+      if (iface < 1 || iface > 6)
+        error->all(FLERR, iarg + 1 + offset, "Illegal region open face index: {}", iface);
       // additional checks on valid face index are done by region classes
       open_faces[iface - 1] = 1;
       openflag = 1;
       iarg += 2;
     } else
-      error->all(FLERR, "Illegal region command argument: {}", arg[iarg]);
+      error->all(FLERR, iarg + offset, "Unknown region command argument: {}", arg[iarg]);
   }
 
   // error check
 
   if ((moveflag || rotateflag) && (strcmp(style, "union") == 0 || strcmp(style, "intersect") == 0))
-    error->all(FLERR, "Region union or intersect cannot be dynamic");
+    error->all(FLERR, 1, "Region union or intersect cannot be dynamic");
 
   // setup scaling
 
@@ -402,7 +428,7 @@ void Region::options(int narg, char **arg)
 
   if (rotateflag) {
     double len = sqrt(axis[0] * axis[0] + axis[1] * axis[1] + axis[2] * axis[2]);
-    if (len == 0.0) error->all(FLERR, "Region cannot have 0 length rotation vector");
+    if (len == 0.0) error->all(FLERR, Error::NOPOINTER, "Region cannot have 0 length rotation vector");
     runit[0] = axis[0] / len;
     runit[1] = axis[1] / len;
     runit[2] = axis[2] / len;

src/region_block.h

@@ -26,6 +26,7 @@ namespace LAMMPS_NS {
 
 class RegBlock : public Region {
   friend class FixPour;
+  friend class Region2VMD;
 
  public:
   RegBlock(class LAMMPS *, int, char **);

src/region_cone.cpp

@@ -41,10 +41,11 @@ RegCone::RegCone(LAMMPS *lmp, int narg, char **arg) :
 
   if (openflag)
     for (int i = 3; i < 6; i++)
-      if (open_faces[i]) error->all(FLERR, "Illegal region cone open face: {}", i + 1);
+      if (open_faces[i])
+        error->all(FLERR, Error::NOPOINTER, "Illegal region cone open face: {}", i + 1);
 
   if (strcmp(arg[2], "x") != 0 && strcmp(arg[2], "y") != 0 && strcmp(arg[2], "z") != 0)
-    error->all(FLERR, "Illegal region cone axis: {}", arg[2]);
+    error->all(FLERR, 2, "Illegal region cone axis: {}", arg[2]);
   axis = arg[2][0];
 
   if (axis == 'x') {
@@ -256,8 +257,8 @@ RegCone::RegCone(LAMMPS *lmp, int narg, char **arg) :
 
   // error check
 
-  if (radiuslo < 0.0) error->all(FLERR, "Illegal radius in region cone command");
+  if (radiuslo < 0.0) error->all(FLERR, 5, "Illegal lower radius in region cone command");
-  if (radiushi < 0.0) error->all(FLERR, "Illegal radius in region cone command");
+  if (radiushi < 0.0) error->all(FLERR, 6, "Illegal upper radius in region cone command");
   if (radiuslo == 0.0 && radiushi == 0.0)
     error->all(FLERR, "Illegal radius in region cone command");
   if (hi <= lo) error->all(FLERR, "Illegal cone length in region cone command");

src/region_cone.h

@@ -25,6 +25,8 @@ RegionStyle(cone,RegCone);
 namespace LAMMPS_NS {
 
 class RegCone : public Region {
+  friend class Region2VMD;
+
  public:
   RegCone(class LAMMPS *, int, char **);
   ~RegCone() override;

src/region_cylinder.h

@@ -26,6 +26,7 @@ namespace LAMMPS_NS {
 
 class RegCylinder : public Region {
   friend class FixPour;
+  friend class Region2VMD;
 
  public:
   RegCylinder(class LAMMPS *, int, char **);

src/region_ellipsoid.h

@@ -25,6 +25,8 @@ RegionStyle(ellipsoid,RegEllipsoid);
 namespace LAMMPS_NS {
 
 class RegEllipsoid : public Region {
+  friend class Region2VMD;
+
  public:
   RegEllipsoid(class LAMMPS *, int, char **);
   ~RegEllipsoid() override;

src/region_plane.h

@@ -25,6 +25,8 @@ RegionStyle(plane,RegPlane);
 namespace LAMMPS_NS {
 
 class RegPlane : public Region {
+  friend class Region2VMD;
+
  public:
   RegPlane(class LAMMPS *, int, char **);
   ~RegPlane() override;

src/region_prism.h

@@ -26,6 +26,7 @@ namespace LAMMPS_NS {
 
 class RegPrism : public Region {
   friend class CreateBox;
+  friend class Region2VMD;
 
  public:
   RegPrism(class LAMMPS *, int, char **);

src/region_sphere.h

@@ -23,8 +23,9 @@ RegionStyle(sphere,RegSphere);
 #include "region.h"
 
 namespace LAMMPS_NS {
-
 class RegSphere : public Region {
+  friend class Region2VMD;
+
  public:
   RegSphere(class LAMMPS *, int, char **);
   ~RegSphere() override;

src/safe_pointers.h

@@ -0,0 +1,61 @@
+/* -*- c++ -*- ----------------------------------------------------------
+   LAMMPS - Large-scale Atomic/Molecular Massively Parallel Simulator
+   https://www.lammps.org/, Sandia National Laboratories
+   LAMMPS development team: developers@lammps.org
+
+   Copyright (2003) Sandia Corporation.  Under the terms of Contract
+   DE-AC04-94AL85000 with Sandia Corporation, the U.S. Government retains
+   certain rights in this software.  This software is distributed under
+   the GNU General Public License.
+
+   See the README file in the top-level LAMMPS directory.
+------------------------------------------------------------------------- */
+
+#ifndef LMP_SAFE_POINTERS_H
+#define LMP_SAFE_POINTERS_H
+
+// collection of smart pointers for specific purposes
+
+#include <cstdio>
+#include <utility>
+
+namespace LAMMPS_NS {
+
+/** Class to automatically close a FILE pointer when a class instance goes out of scope.
+
+\verbatim embed:rst
+
+Drop in replacement for ``FILE *``. Use as ``SafeFilePtr fp;`` instead of
+``FILE *fp = nullptr;`` and there is no more need to explicitly call
+``fclose(fp)``.
+
+\endverbatim
+*/
+class SafeFilePtr {
+ public:
+  SafeFilePtr() : fp(nullptr) {};
+  SafeFilePtr(FILE *_fp) : fp(_fp) {};
+
+  SafeFilePtr(const SafeFilePtr &) = delete;
+  SafeFilePtr(SafeFilePtr &&o) noexcept : fp(std::exchange(o.fp, nullptr)) {}
+  SafeFilePtr &operator=(const SafeFilePtr &) = delete;
+
+  ~SafeFilePtr()
+  {
+    if (fp) fclose(fp);
+  }
+
+  SafeFilePtr &operator=(FILE *_fp)
+  {
+    if (fp && (fp != _fp)) fclose(fp);
+    fp = _fp;
+    return *this;
+  }
+  operator FILE *() const { return fp; }
+
+ private:
+  FILE *fp;
+};
+}    // namespace LAMMPS_NS
+
+#endif

src/text_file_reader.h

@@ -26,6 +26,7 @@
 
 namespace LAMMPS_NS {
 class TextFileReader {
+ private:
   std::string filetype;
   bool closefp;
   int bufsize;

src/version.h

@@ -1 +1,2 @@
 #define LAMMPS_VERSION "4 Feb 2025"
+#define LAMMPS_UPDATE "Development"

tools/lammps-gui/CMakeLists.txt

@@ -279,7 +279,7 @@ if(APPLE)
     MACOSX_BUNDLE_BUNDLE_VERSION ${PROJECT_VERSION}
     MACOSX_BUNDLE_SHORT_VERSION_STRING ${PROJECT_VERSION_MAJOR}.${PROJECT_VERSION_MINOR}
     MACOSX_BUNDLE_ICON_FILE lammps.icns
-    MACOSX_BUNDLE_COPYRIGHT "(c) 2003 - 2024, The LAMMPS Developers"
+    MACOSX_BUNDLE_COPYRIGHT "(c) 2003 - 2025, The LAMMPS Developers"
     MACOSX_BUNDLE TRUE
   )
   # additional targets to populate the bundle tree and create the .dmg image file

tools/singularity/README.md

@@ -1,6 +1,6 @@
 # Apptainer (aka Singularity) container definitions for compiling/testing LAMMPS
 
-The *.def files in this folder can be used to build container images
+The \*.def files in this folder can be used to build container images
 for [Apptainer](https://apptainer.org) (previously called
 [Singularity](https://sylabs.io)), suitable for compiling and testing
 LAMMPS on a variety of OS variants with support for most standard

tools/singularity/fedora41_musl_mingw.def

@@ -0,0 +1,109 @@
+BootStrap: docker
+From: fedora:41
+
+%setup
+   curl -L -o musl-gcc-f37.tar.gz https://download.lammps.org/static/musl-gcc-f37.tar.gz
+   cp musl-gcc-f37.tar.gz ${APPTAINER_ROOTFS}
+
+%post
+        dnf -y update
+        dnf -y install vim-enhanced git file make cmake patch which file \
+               ninja-build libomp-devel diffutils dos2unix findutils rsync \
+               ccache gcc-c++ gcc-gfortran gdb valgrind python3-pyyaml \
+               enchant enchant2 python3-enchant python3-virtualenv doxygen latexmk \
+               texlive-latex-fonts texlive-pslatex texlive-collection-latexrecommended \
+               texlive-latex texlive-latexconfig doxygen-latex texlive-collection-latex \
+               texlive-latex-bin texlive-lualatex-math texlive-fncychap texlive-tabulary \
+               texlive-framed texlive-wrapfig texlive-upquote texlive-capt-of texlive-pict2e \
+               texlive-needspace texlive-titlesec texlive-anysize texlive-dvipng texlive-xindy \
+               texlive-fontawesome texlive-ellipse texlive-tex-gyre \
+               mingw-filesystem-base mingw32-nsis mingw-binutils-generic \
+               mingw64-filesystem mingw64-pkg-config \
+               mingw64-crt mingw64-headers mingw64-binutils \
+               mingw64-cpp mingw64-gcc mingw64-gcc-gfortran mingw64-gcc-c++ \
+               mingw64-curl \
+               mingw64-libgomp \
+               mingw64-winpthreads \
+               mingw64-eigen3 \
+               mingw64-fftw \
+               mingw64-libjpeg-turbo \
+               mingw64-libpng \
+               mingw64-python3 \
+               mingw64-python3-numpy \
+               mingw64-python3-pyyaml \
+               mingw64-python3-setuptools \
+               mingw64-readline \
+               mingw64-termcap \
+               mingw64-tcl \
+               mingw64-tk \
+               mingw64-zlib \
+               mingw64-zstd \
+               mingw64-qt5-qtdeclarative \
+               mingw64-qt5-qmldevtools \
+               mingw64-qt5-qmldevtools-devel \
+               mingw64-qt5-qttools-tools \
+               mingw64-qt5-qtcharts \
+               mingw64-qt5-qttools \
+               mingw64-qt5-qmake \
+               mingw64-qt5-qtbase \
+               mingw64-qt5-qtbase-devel \
+               mingw64-qt5-qtbase-static \
+
+        dnf clean all
+
+        # install musl-libc Linux-2-Linux cross-compiler
+        tar -C /usr/ -xvf /musl-gcc-f37.tar.gz
+        rm -f /musl-gcc-f37.tar.gz
+
+        # install NSIS EnVar plugin
+        curl -L -o EnVar_plugin.zip https://nsis.sourceforge.io/mediawiki/images/7/7f/EnVar_plugin.zip
+        unzip -d /usr/share/nsis EnVar_plugin.zip
+        rm EnVar_plugin.zip
+
+        # create missing termcap pkgconfig file
+        cat > /usr/x86_64-w64-mingw32/sys-root/mingw/lib/pkgconfig/termcap.pc <<EOF
+prefix=/usr/x86_64-w64-mingw32/sys-root/mingw
+exec_prefix=/usr/x86_64-w64-mingw32/sys-root/mingw
+libdir=/usr/x86_64-w64-mingw32/sys-root/mingw/lib
+includedir=/usr/x86_64-w64-mingw32/sys-root/mingw/include
+
+Name: Termcap
+Description: GNU/MinGW terminal feature database
+URL: ftp://ftp.gnu.org/gnu/termcap/
+Version: 1.3
+Libs: -L\${libdir} -ltermcap
+Cflags:
+EOF
+        # set custom prompt indicating the container name
+        CUSTOM_PROMPT_ENV=/.singularity.d/env/99-zz_custom_prompt.sh
+        cat >$CUSTOM_PROMPT_ENV <<EOF
+#!/bin/bash
+PS1="[fedora41/musl-mingw:\u@\h] \W> "
+EOF
+        chmod 755 $CUSTOM_PROMPT_ENV
+
+%environment
+        # we need to reset any module variables
+        # inherited from the host.
+        unset __LMOD_REF_COUNT__LMFILES_
+        unset __LMOD_REF_COUNT_PATH
+        unset __LMOD_REF_COUNT_LD_LIBRARY_PATH
+        unset __LMOD_REF_COUNT_MANPATH
+        unset __LMOD_REF_COUNT_MODULEPATH
+        unset __LMOD_REF_COUNT_LOADEDMODULES
+        unset _LMFILES_
+        unset MODULEPATH
+        unset MODULESHOME
+        unset MODULEPATH_ROOT
+        unset LOADEDMODULES
+        unset LMOD_SYSTEM_DEFAULT_MODULES
+
+        . /etc/profile
+        # tell OpenMPI to not try using Infiniband
+        OMPI_MCA_btl="^openib"
+        # do not warn about unused components as this messes up testing
+        OMPI_MCA_btl_base_warn_component_unused="0"
+        export OMPI_MCA_btl OMPI_MCA_btl_base_warn_component_unused
+
+%labels
+        Author akohlmey