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

Creation of User Defined Types With Arrays #65

Open
rustam-azimov opened this issue Aug 3, 2020 · 5 comments
Open

Creation of User Defined Types With Arrays #65

rustam-azimov opened this issue Aug 3, 2020 · 5 comments

Comments

@rustam-azimov
Copy link

Is it possible to create user defined types for matrix elements that contain int array, for example?
If so, what types should be entered instead of double and uint64_t in the following line?

members = ['double w', 'uint64_t h', 'uint64_t pi']

@michelp
Copy link
Collaborator

michelp commented Aug 4, 2020

Hi @rustam-azimov,

Unfortunately at the moment matrix elements can only be scalar elements of a fixed size. This is a limitation of the underlying GraphBLAS spec. There has been talk of having variable length elements, or event elements that are sub-matrices themselves but that's still in the discussion phase.

@szarnyasg
Copy link
Contributor

I think this will be possible with the C++ API but (I think) providing a solution for this is not even on the roadmap for the C API. So we might close this issue as a wontfix for the time being.

@rustam-azimov
Copy link
Author

Yes, thank you.

@rustam-azimov
Copy link
Author

I would like to share that we were able to implement our algorithm for CFPQ that can restore all founded paths. We use C realization of GraphBLAS and C++ structure with int* array and int size members for matrix elements. It will be great if in the future it will be possible to use int* in matrix element types in pygraphblas.

@rustam-azimov rustam-azimov reopened this Feb 10, 2021
@rustam-azimov
Copy link
Author

rustam-azimov commented Feb 10, 2021

How you think, is it possible or there are some limitations from cffi or Python itself?

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

3 participants