Installation with pip (recommended)

Server software requirements

To use MI Scripting Toolkit 4.2, your Granta MI server must be running 2023 R1 or later, which includes Service Layer Interface 22/04.

Client software requirements

MI Scripting Toolkit for pip is supported on the following platforms:

Platform	Python 3.10	Python 3.11	Python 3.12	Python 3.13
Windows 1	Certified	Certified	Certified	Certified
Linux	Ubuntu 22.04 2	Certified	Certified	Certified
Ubuntu 24.04 3	Supported	Supported	Certified	Supported
SUSE Linux Enterprise 15 SP4+ 4	Supported	Supported	Certified	Supported
RHEL 8.8+, 9.3+ 5	Supported	Supported	Certified	Supported
Rocky Linux 8.9+, 9.3+ 6	Supported	Supported	Certified	Supported

Test platform details:

• [1] Automated testing performed on Windows Server 2022. Manual testing performed on Windows Server 2019 and Windows 11.
• [2] Automated testing performed on Ubuntu 22.04 development release. Manual testing performed on Ubuntu 22.04.4.
• [3] Automated testing performed on Ubuntu 24.04 development release. Manual testing performed on Ubuntu 24.04.2.
• [4] Automated testing performed on SLES 15 SP4. Manual testing performed on SLES 15 SP5.
• [5] Automated testing performed on RHEL 8.8 and 9.3. Manual testing performed on RHEL 8.10.
• [6] Automated testing performed on Rocky Linux 8.9 and 9.3. Manual testing performed on Rocky Linux 9.4.

Linux requirements

The wheel provided for installation via pip on Linux is a ‘manylinux’ wheel and is compatible with any manylinux 2.28-compliant Linux platform. This typically means that the platform must be based on glibc>=2.28.

NOTE

The use of MI Scripting Toolkit on other Linux platforms that meet this threshold is not supported or tested, but it is anticipated that it will function as expected.

For Linux platforms that do not meet this requirement, it is recommended to either use a compatible docker container or to install via conda. Whilst the use of conda packages in other platforms is not tested, it is anticipated that they will function as expected.

MI Scripting Toolkit on Linux requires additional shared libraries beyond the glibc runtime. The required shared library names are listed in the left-hand column; these usually represent symbolic links to shared libraries. The full name of the shared library on our build environment is listed in the right-hand column.

Required shared library name	Full library path on the ManyLinux 2.28 build image
libc.so.6	/lib64/libc.so.6
libdl.so.2	/lib64/libdl.so.2
libgcc_s.so.1	/lib64/libgcc_s.so.1
libm.so.6	/lib64/libm.so.6
libpthread.so.0	/lib64/libpthread.so.0
librt.so.1	/lib64/librt.so.1
libstdc++.so.6	/lib64/libstdc++.so.6
libuuid.so.1	/lib64/libuuid.so.1
libz.so.1	/lib64/libz.so.1

Required third-party packages

MI Scripting Toolkit depends on the following third-party packages:

• filetype (>=1.2.0,2.0.0)
• packaging (>=24.0,25.0)
• python-dateutil (>=2.9.0,3.0.0)

NOTE

The list above include only the direct dependencies of the MI Scripting Toolkit package. The transitive dependencies (dependencies of dependencies) are not given here, since in general they depend on the specific version of the dependency that is selected, and as a result will change over time.

To determine the full requirements at a particular point in time, install the Scripting Toolkit into an empty virtual environment (see https://docs.python.org/3/library/venv.html for details), and then use pip list or pip freeze to display the installed packages.

These third-party packages are not provided with MI Scripting Toolkit, but must be installed before use. Follow the installation instructions in Installing and upgrading MI Scripting Toolkit to automatically download compatible versions of these packages from PyPI and install them alongside MI Scripting Toolkit. To install MI Scripting Toolkit without access to PyPI, see Offline installation.

Installation preparation

MI Scripting Toolkit is provided as a zip file, and is available from the Downloads page on the Ansys Customer Portal. All instructions on this page assume a Windows environment, and that the zip file has been downloaded and unzipped to the location C:\Users\user\scripting_toolkit. Replace this path as required for your environment.

To install MI Scripting Toolkit in a Linux environment, replace the example Windows path with the appropriate Linux path. For example, the path C:\Users\user\scripting_toolkit\pip in Windows would be equivalent to /home/user/scripting_toolkit/pip on Linux.

Installing and upgrading MI Scripting Toolkit

Install MI Scripting Toolkit in a virtual environment. If you are installing MI Scripting Toolkit in an activated virtual environment, enter the following:

$ pip install --upgrade granta_miscriptingtoolkit --find-links="C:\Users\user\scripting_toolkit\pip"

If you are installing MI Scripting Toolkit in a Python environment which is not available on the path, enter the following:

$ path/to/python -m pip install --upgrade granta_miscriptingtoolkit --find-links="C:\Users\user\scripting_toolkit\pip"

When installation is complete, the MI Scripting Toolkit files are installed into the Python site-packages location.

pip extras

MI Scripting Toolkit defines additional optional dependencies in the form of extras:

• [notebooks]: Required to execute the Example notebooks.
• [oidc]: Required to use OpenID Connect (OIDC) authentication.

To install an extra, include the extra name after the package name when installing MI Scripting Toolkit. For example, to install MI Scripting Toolkit with the additional notebook dependencies:

$ pip install --upgrade granta_miscriptingtoolkit[notebooks] --find-links="C:\Users\user\scripting_toolkit\pip"

Or to install both the notebook and OIDC dependencies:

$ pip install --upgrade granta_miscriptingtoolkit[notebooks,oidc] --find-links="C:\Users\user\scripting_toolkit\pip"

The [notebooks] extra includes the following additional dependencies:

• Linux and Windows platforms: : - jupyterlab (>=4.0.0,5.0.0)
  • matplotlib (>=3.6.0,4.0.0)
  • numpy (>=2.0.0,3.0.0)
  • pandas (>=2.1,3.0)
  • python-docx (>=1.1.0,2.0.0)
  • python-pptx (>=1.0.0,2.0.0)
  • scipy (>=1.11.0,2.0.0)
  • seaborn (>=0.13,0.14)
• Windows only: : - docx2pdf (>=0.1.8,0.2.0)
  • tqdm[notebook] (>=4.0,5.0)

The [oidc] extra includes the following additional dependencies:

• keyring (>23,26)

If you encounter dependency errors when installing the [notebooks] extra, you may need to first upgrade pip:

python -m pip install --upgrade pip

Upgrading from MI Scripting Toolkit v3.0 or earlier

WARNING

The MI Scripting Toolkit package name changed with v3.1 from miscriptingtoolkit to GRANTA_MIScriptingToolkit. If you are upgrading from MI Scripting Toolkit v3.0 or earlier, you must follow the instructions in this section or pip install may fail with dependency conflicts.

If you are using a virtual environment, you should delete your virtual environment and create a new one.

If you are not using a virtual environment, you should uninstall older version of MI Scripting Toolkit by entering the following into the command line:

$ pip uninstall miscriptingtoolkit

Or if you are using a Python environment which is not available on the path:

$ path/to/python -m pip uninstall miscriptingtoolkit

Offline installation

To install MI Scripting Toolkit without access to PyPI, first download the required third-party dependencies using a Python environment with PyPI access. Enter the following into the command line to download the wheels into the C:\Users\user\all_wheels directory:

$ pip download GRANTA_MIScriptingToolkit --find-links=pip --dest="C:\Users\user\all_wheels"

Copy the all_wheels directory to the machine without access to PyPI, and enter the following into the command line to install MI Scripting Toolkit using the packages in the all_wheels directory:

$ pip install GRANTA_MIScriptingToolkit --find-links="C:\Users\user\all_wheels"

These instructions describe how to download the required packages for the current platform only. To download packages for a different platform, for example to download Linux wheels in a Windows environment, see the pip download documentation.