Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Best way to install CCBlade when using RAFT as standalone #17

Open
gbarter opened this issue Sep 19, 2022 · 6 comments
Open

Best way to install CCBlade when using RAFT as standalone #17

gbarter opened this issue Sep 19, 2022 · 6 comments

Comments

@gbarter
Copy link
Member

gbarter commented Sep 19, 2022

Hi Matt, Stein, and the greater RAFT team,

What is your recommended way of installing CCBlade when using RAFT as a standalone? To use the standalone CCBlade (my hypothesis)? Install WISDEM from source? From conda install wisdem? Just use WEIS and get everything together?

This is essentially the question being asked in this WISDEM issue. Feel free to post a reply back there if that is easiest.

Cheers,
Garrett

@mattEhall
Copy link
Contributor

Hi Garrett,
Getting CCBlade by doing the full WISDEM install as described on the WISDEM repo is what I know works. The last time I thought about this was in the middle of the WEIS project, and at that time I think there were differences between the CCBlade in WISDEM and the standalone CCBlade that made the standalone CCBlade incompatible. But I don't remember exactly, and maybe it has changed since then. Certainly I think it would be ideal if RAFT could work with the standalone CCBlade as well.

Cheers,
Matt

@gbarter
Copy link
Member Author

gbarter commented Sep 21, 2022

Perhaps there was a point in the development cycle where the WISDEM & CCBlade repos were out of sync, but I just did a diff on the Fortran and Python code in both and they are equivalent.

@mattEhall
Copy link
Contributor

This is great news! I'll have to try it out with just CCBlade installed and see if there are any other hurdles. It will be great to have a minimal-dependencies option for people.

@mattEhall
Copy link
Contributor

mattEhall commented Jan 7, 2023

After a couple hurdles and delays, @shousner and I had success in running RAFT with standalone CCBlade. The only tricky bit was installation dependencies for CCBlade, which ended up being very easy (after I made some mistakes and overcomplicated everything). With the latest commit, RAFT will now look for standalone CCBlade before looking for WISDEM.CCBlade.

In case it's helpful to other novices like me, here's what worked for me for the standalone install:
conda create --name raft python=3.9
conda install numpy
conda install pytest
conda install scipy
conda install libpython
conda install -c conda-forge m2w64-toolchain

I then installed each of CCBlade, pyHAMS, MoorPy, and RAFT using
python setup.py develop

I'll wait for feedback on the above before adding this to the readme.

@gbarter
Copy link
Member Author

gbarter commented Jan 7, 2023

Hi Matt- Sorry if this took you a lot of time to iterate on as I could have saved you some pain. This recipe is pretty much the core set we use for WISDEM and WEIS.

My only suggestion is to use the conda-forge channel instead of msys2, as that is a bit better supported and self-consistent with the other packages.

Cheers,
Garrett

@mattEhall
Copy link
Contributor

Not at all - it was mostly just me being slow to remember the most basic things and then getting sidetracked by other things. I've edited the above to use conda-forge, hopefully correctly!
Cheers,
Matt

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants