Computational Design

This repo contains all my submission for the "Computional Design I" module at the University of Kassel. The seminar was hosted by the chair of Prof. Philipp Eversmann, EDEK.

There were 11 assignments. Every exercise is in its own named folder. There were no strict boundaries for the assignments, only topics – you basically had to make your own tasks for the week.
Students were expected to submit a self-containing Grasshopper file, and a describing PDF. This repo also holds the code by itself as seperate .py files.

All exercises are made with Rhino 6 and should run without any issues.

On Python

My recommendation is to actually script in VSCode (or any other editor you like for that matter). Use the Python stubs provided by McNeel, plus these, to have access to autocomplete. Then use the read file + code input at the Python component in Grasshopper.

Since Rhino heavily works with .NET the suiting Python interpreter is IronPython. Remember, this means Python 2.7! So don't be tempted to use any of the cool new stuff. Grasshopper Python seems to do an implicit from __future__ import *. However I'm not sure to what extend. print() is accepted in any case.

Cheat sheet

Helpful, if you're a beginner.

Cheat sheet

• Python strings are immutable!
• ... but Lists, Points, Surfaces, Breps, Meshes… are all REFERENCE Type based:
  myCopyList = myList : is not copying the contents of one list into the other but actually linking both variables to the same data.
  Use myCopyList = copy.deepcopy(myList) instead.
• Class names should start with Uppercase
• print(sys.path) to see where Python looks for modules to import right now
• You can append to sys.path!
• To get the current file path where GH doc is stored: file_name = ghenv.Component.OnPingDocument().FilePath
• See the ghpy_helpers file to see how to flatten lists in lists.

Slicing

Slicing, the valuable technique of taking ranges out of lists.

• list[:-1] whole list but without last item
• list[1:] start at the second item
• list[1::2] start with second item, always skip one then

Managing different Python versions

Either

• via Miniconda
  • Installing a different python besides the existing one:
    conda create -n python36 python=3.6 anaconda
    python36 is hereby the environment name, python the package name (which could be any one, like numpy)
  • activate by typing activate envname
  • check by typing python --version
  • use conda env list for a list of all installed environments
• or pipenv

---

License

All code is released under the BOML.