I’m doing this with (at least) one hand tied behind my back: I don’t know Stella, beyond a cursory understanding of what some things do and how to run models. So, I’m reliant on someone else’s translation script to make sure we get all the parts in the right place. It’s like they have provided at least a blue print of the work.
The other half-tied hand is that I’m out of practice with Python, but in this debugging practice, it’s coming back to me. In general, I do know how to do some things, and I know in broad terms what is bad coding practice and what isn’t. I don’t want to bag on someone else’s work, because I have no right. They did something open source and put it out there on the internet, for me to then download and co-opt. Thank you to the original authors for the start; it’s a good one.
I mostly blame Stella: it’s closed source, and downright sloppy syntax. I really don’t get how the equation layer works; I assume it’s just “for show” and something the authors can look at to understand in some term what’s going on (but I have no idea). I just don’t understand how it can work if the syntax is seemingly so inconsistent from place to place in the script. It makes translating it a bear (but hey, Stella probably doesn’t want me to translate it).
There are a few things going on with this translation script that I’m trying to deal with, in order…
- lack of consideration for things brought up by the giant, as-complex-as-can-be model I was handed
- poor structure/programming paradigm/coding practice
In handling all this, here are some things I have changed, in case you download the script and want a heads up on the tweaks I’m making that I envision will comprise the next version of this script. I plan to send it back to the authors, and maybe host it on my own website.