Question

I have a directory structure for my project as

<some_path>/code/project/
                        /conf
                        /data
                        /src/
                            a/x.py
                            b/
                             c/y.py
                            ... 

• each of x.py, y.py can run in isolation, so has to append the project path in its own script as

sys.path.append(‘~/code/project’)

• This becomes unmanageable when your code needs to run on multiple machines on cluster and I end up changing the paths manually 🙁

• Its a bad design and I want to learn what is the best way to append the project

• what is the best way to append the project path in this case so that

  • I dont write sys.path.append in each of the .py file
  • I can define a master configuration that runs every time a .py file runs

Thank you

Answer 1

Add the path to your project to the PYTHONPATH environment variable.

See the documentation on the Module Search Path for how exactly this is used.