Using snappy with python3.7 or higher

Hi,

Is someone working in a conda env with python > 3.6. I am trying to set snappy to work with at least python 3.7. From the documentation I see:

Standard Python (CPython) Approach

With this approach you can use a standard Python (CPython) interpreter installed on your computer (SNAP does not include a CPython interpreter.) The supported versions are Python 2.7, 3.3 to 3.6 64-bit (Linux + Darwin) and both 32-bit and 64-bit (Windows) as well as Anaconda distributions.

https://senbox.atlassian.net/wiki/spaces/SNAP/pages/19300362/How+to+use+the+SNAP+API+from+Python

You should mention your platform. If you are comfortable with development tools, you may be able to build jpy for newer python versions. When snappyutil.py fails to find a suitable jpy binary wheel it prints instructions for building a binary wheel with your choice of python., but you may need to deal with changes made to more recent python versions and others won’t be able to help with any issues you encounter as they won’t have the same version of python. Users who lack development experience often struggle to understand build failures.

Snappy is widely used with python-3.6, so is relatively well proven and limitations are understood.

Python 3.6 is readily available for all the platforms that run SNAP natively. With Anaconda you can install a python-3.6 environment for use with snappy. If you are hoping to embed snappy in a large python 3.n (n>6) system, you might want to review About java-python integration (Eclipse.org). If you are planning to use a particular python library you can search this forum for posts that mention the library and snappy together.

@gnwiii thanks for the details. Facing the jpy issue in 3.9 indeed and having some issues to make it work. No problem to stick to 3.6 in a conda env although I am having some issues to make rasterio and geopandas work there, but I guess those are not SNAP related obviously… Btw, working on a Windows 10 machine

Any known plans to extend snappy to more recent python versions. This may be a potential issue with time.

M