Requires a Python 2.7 installation, not compatible with Python 3. Python 2.6 might work but is not supported.

Uses rpyc for the connection backend (automatically installed).

From the Windows command line (or the special Anaconda or Python command if pip is not in your path by default), run: (If you are using a virtual environment, remember to activate it first.)

pip install gh-python-remote --upgrade --process-dependency-links --no-binary=:all:
python -m ghpythonremote._configure_ironpython_installation

The first line installs gh-python-remote in the current Python interpreter. The second tries to find your Rhinoceros IronPython installation, and install gh-python-remote there. The extra options are necessary to be able to get the pre-release version of rpyc.

If you do not use the standard Rhinoceros IronPython installation (%APPDATA%\McNeel\Rhinoceros\5.0\Plug-ins\IronPython (814d908a-e25c-493d-97e9-ee3861957f49)\settings\lib), you can specify a target directory to use like so: python -m ghpythonremote._configure_ironpython_installation "location".

This will also install the gh-python-remote UserObject in Grasshopper.

All the examples files are copied in the %APPDATA%\Grasshopper\UserObjects\gh-python-remote\examples folder. You can also download them from the github repo.

1. Open the example file GH_python_remote.ghx in Python, or drop the gh-python-remote component on the canvas.

2. Use the location input to define the location of the Python interpreter you want to connect to. You can use the path to a folder containing python, the full path to a python executable, or conda:// followed by the name of an Anaconda virtual environment.

3. Use the modules input to define the modules you want to access in the GHPython component. Anything that can follow an import statement in the remote Python works. If you need to import a submodule inside a package (like import this.that), the parent package has to be imported first.

4. Change run to True to connect.

5. In the GHPython component, the imported modules will be available via the sticky dictionary. For example if you are trying to use Numpy:

   import scriptcontext
   np = scriptcontext.sticky['numpy']

6. Done!

Creating remote array-like objects from large local lists can be slow. For example, np.array(range(10000)) takes more than 10 seconds on most computers. To solve this, you need to send the list first to the remote Python interpreter, then create the array from this remote object:

import scriptcontext as sc
import rpyc
np = sc.sticky['numpy']
rpy = sc.sticky['rpy']

r_range = rpyc.utils.classic.deliver(rpy, range(10000))
np.array(r_range)

There is also an issue that Grasshopper does not recognize remote list objects as lists. They need to be recovered to the local interpreter first:

import scriptcontext as sc
import rpyc
from ghpythonlib.treehelpers import list_to_tree # Rhino 6 only!
np = sc.sticky['numpy']

a = np.arange(15).reshape((3,5))
a = rpyc.utils.classic.obtain(a.tolist())
a = list_to_tree(a, source=[0,0])

ghpythonlib.treehelpers is Rhino 6 only, see the treehelpers gist for an equivalent implementation:

def list_to_tree(input, none_and_holes=True, source=[0]):
    """Transforms nestings of lists or tuples to a Grasshopper DataTree"""
    from Grasshopper import DataTree as Tree
    from Grasshopper.Kernel.Data import GH_Path as Path
    from System import Array
    def proc(input,tree,track):
        path = Path(Array[int](track))
        if len(input) == 0 and none_and_holes: tree.EnsurePath(path); return
        for i,item in enumerate(input):
            if hasattr(item, '__iter__'): #if list or tuple
                track.append(i); proc(item,tree,track); track.pop()
            else:
                if none_and_holes: tree.Insert(item,path,i)
                elif item is not None: tree.Add(item,path)
    if input is not None: t=Tree[object]();proc(input,t,source[:]);return t

* marks an input that is only available by editing the gh-python-remote UserObject, or in GH_python_remote.ghx.

You can also use gh-python-remote to programmatically control a Rhinoceros instance, and connect to it via Python. Have a look at examples/python_to_GH.py for a full working example.

Licensed under the MIT license.