Installing mdb

Windows Support

Note

mdb does not run on Windows. This is due to a core dependency pexpect which does not currently support windows for the pexpect.spawn and pexpect.run methods (see here for more information).

Standard Installation (UNIX)

The easiest way to install mdb is to clone the repo and install it into a virtual environment of your choice. For the following example, I will use Python’s built-in venv virtual environment module.

First we will create and activate a virtual environment for mdb.

$ python3 -m venv .mdb
$ source .mdb/bin/activate

You then have two options for install. Either from GitHub source (recommended for developers) or using PyPI (recommended for developers).

pip install

To install mdb with the optional package termgraph use the following command (Note: the name mdb is already in use on PyPI, so I have decided to call the repo mdb-debugger)

$ pip install mdb-debugger[termgraph]

GitHub install

Then we can download and install the mdb code. You can download/clone the mdb code anywhere but in this example I will choose the home directory.

$ cd ~
$ git clone https://github.com/TomMelt/mdb.git
$ cd mdb
$ pip install .[termgraph]

That’s it! We now have mdb installed and ready to go. You can test the install was successfully by running mdb --help. You should see the following output. Specifying [termgraph] is optional but it will give pretty plots straight to the terminal.

$ mdb --help
Usage: mdb [OPTIONS] COMMAND [ARGS]...

Options:
  --help  Show this message and exit.

  Commands:
    attach
    launch

For Developers

If you plan to carry out development it will be handy to install some additional dependencies. This can be done in the following way.

$ pip install -e .[docs,develop,termgraph]

The -e (or --editable) flag installs the project in editable mode meaning that your changes will be reflected when you save the source files and restart mdb. docs installs packages required to build the documentation and develop installs packages required for verifying the quality of the code. Please also familiarize yourself with the CONTRIBUTING.md guide, for more details on how to best contribute to mdb.

The additional develop dependencies are;

black is used for formatting the source code.
flake8 is used to check the style and quality of some python code.
mypy is a static type checker and is used to help ensure that variables and functions are being used correctly.
pytest for creating and running unit and integration tests
pytest-cov for producing coverage reports.